Upload folder using huggingface_hub
Browse files- checkpoint-430/README.md +202 -0
- checkpoint-430/adapter_config.json +37 -0
- checkpoint-430/adapter_model.safetensors +3 -0
- checkpoint-430/optimizer.pt +3 -0
- checkpoint-430/rng_state.pth +3 -0
- checkpoint-430/scheduler.pt +3 -0
- checkpoint-430/trainer_state.json +3043 -0
- checkpoint-430/training_args.bin +3 -0
checkpoint-430/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: rinna/gemma-2-baku-2b
|
| 3 |
+
library_name: peft
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.13.2
|
checkpoint-430/adapter_config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "rinna/gemma-2-baku-2b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 32,
|
| 14 |
+
"lora_dropout": 0,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": [
|
| 18 |
+
"lm_head",
|
| 19 |
+
"embed_tokens"
|
| 20 |
+
],
|
| 21 |
+
"peft_type": "LORA",
|
| 22 |
+
"r": 128,
|
| 23 |
+
"rank_pattern": {},
|
| 24 |
+
"revision": null,
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"k_proj",
|
| 27 |
+
"q_proj",
|
| 28 |
+
"down_proj",
|
| 29 |
+
"up_proj",
|
| 30 |
+
"gate_proj",
|
| 31 |
+
"o_proj",
|
| 32 |
+
"v_proj"
|
| 33 |
+
],
|
| 34 |
+
"task_type": "CAUSAL_LM",
|
| 35 |
+
"use_dora": false,
|
| 36 |
+
"use_rslora": true
|
| 37 |
+
}
|
checkpoint-430/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1646f26d3155f7251807d70b65966852e4e3d6b3185c5ce64035547f0ee178ae
|
| 3 |
+
size 7742666784
|
checkpoint-430/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9df37195d52f070947cd1aa4fa2560e07036aa7470def99a00cb76b9842cbaa
|
| 3 |
+
size 2734115724
|
checkpoint-430/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a70ab7d1c2c41e85a659e7c584c1c5343dc1f14d8e6a7f4dd6bbd2caf8058e08
|
| 3 |
+
size 14244
|
checkpoint-430/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de87a72fabe441e7f456adcc1b58a281744229a9396e890b6ae8b225e8692f12
|
| 3 |
+
size 1064
|
checkpoint-430/trainer_state.json
ADDED
|
@@ -0,0 +1,3043 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.99953509995351,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 430,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0023245002324500234,
|
| 13 |
+
"grad_norm": 24.9628849029541,
|
| 14 |
+
"learning_rate": 7.222222222222221e-06,
|
| 15 |
+
"loss": 2.9398,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004649000464900047,
|
| 20 |
+
"grad_norm": 27.249839782714844,
|
| 21 |
+
"learning_rate": 1.4444444444444442e-05,
|
| 22 |
+
"loss": 3.0436,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.00697350069735007,
|
| 27 |
+
"grad_norm": 29.58257293701172,
|
| 28 |
+
"learning_rate": 2.1666666666666664e-05,
|
| 29 |
+
"loss": 2.5493,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009298000929800094,
|
| 34 |
+
"grad_norm": 21.416799545288086,
|
| 35 |
+
"learning_rate": 2.8888888888888885e-05,
|
| 36 |
+
"loss": 1.5315,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.011622501162250116,
|
| 41 |
+
"grad_norm": 6.383184909820557,
|
| 42 |
+
"learning_rate": 3.611111111111111e-05,
|
| 43 |
+
"loss": 1.2681,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01394700139470014,
|
| 48 |
+
"grad_norm": 3.6736369132995605,
|
| 49 |
+
"learning_rate": 4.333333333333333e-05,
|
| 50 |
+
"loss": 1.1031,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.016271501627150162,
|
| 55 |
+
"grad_norm": 2.394563913345337,
|
| 56 |
+
"learning_rate": 5.055555555555555e-05,
|
| 57 |
+
"loss": 1.1564,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.018596001859600187,
|
| 62 |
+
"grad_norm": 2.2405858039855957,
|
| 63 |
+
"learning_rate": 5.777777777777777e-05,
|
| 64 |
+
"loss": 1.1077,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.02092050209205021,
|
| 69 |
+
"grad_norm": 2.2083897590637207,
|
| 70 |
+
"learning_rate": 6.5e-05,
|
| 71 |
+
"loss": 1.0652,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.023245002324500233,
|
| 76 |
+
"grad_norm": 2.0182244777679443,
|
| 77 |
+
"learning_rate": 6.499909512851264e-05,
|
| 78 |
+
"loss": 1.0865,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.025569502556950254,
|
| 83 |
+
"grad_norm": 1.935754418373108,
|
| 84 |
+
"learning_rate": 6.499638056443783e-05,
|
| 85 |
+
"loss": 1.1143,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.02789400278940028,
|
| 90 |
+
"grad_norm": 1.7560354471206665,
|
| 91 |
+
"learning_rate": 6.499185645893443e-05,
|
| 92 |
+
"loss": 1.0293,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.030218503021850304,
|
| 97 |
+
"grad_norm": 1.882704496383667,
|
| 98 |
+
"learning_rate": 6.498552306392452e-05,
|
| 99 |
+
"loss": 1.1189,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.032543003254300325,
|
| 104 |
+
"grad_norm": 1.6570043563842773,
|
| 105 |
+
"learning_rate": 6.49773807320794e-05,
|
| 106 |
+
"loss": 1.024,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03486750348675035,
|
| 111 |
+
"grad_norm": 1.6351194381713867,
|
| 112 |
+
"learning_rate": 6.496742991679994e-05,
|
| 113 |
+
"loss": 1.1053,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.037192003719200374,
|
| 118 |
+
"grad_norm": 1.693750023841858,
|
| 119 |
+
"learning_rate": 6.495567117219131e-05,
|
| 120 |
+
"loss": 1.0443,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.03951650395165039,
|
| 125 |
+
"grad_norm": 1.6009650230407715,
|
| 126 |
+
"learning_rate": 6.494210515303213e-05,
|
| 127 |
+
"loss": 0.9832,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.04184100418410042,
|
| 132 |
+
"grad_norm": 1.6692452430725098,
|
| 133 |
+
"learning_rate": 6.492673261473802e-05,
|
| 134 |
+
"loss": 0.9498,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.04416550441655044,
|
| 139 |
+
"grad_norm": 1.6801658868789673,
|
| 140 |
+
"learning_rate": 6.490955441331956e-05,
|
| 141 |
+
"loss": 0.9983,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.046490004649000466,
|
| 146 |
+
"grad_norm": 1.5920137166976929,
|
| 147 |
+
"learning_rate": 6.489057150533456e-05,
|
| 148 |
+
"loss": 1.0731,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.04881450488145049,
|
| 153 |
+
"grad_norm": 1.5493357181549072,
|
| 154 |
+
"learning_rate": 6.486978494783486e-05,
|
| 155 |
+
"loss": 0.9685,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05113900511390051,
|
| 160 |
+
"grad_norm": 1.4238895177841187,
|
| 161 |
+
"learning_rate": 6.484719589830741e-05,
|
| 162 |
+
"loss": 0.9721,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05346350534635053,
|
| 167 |
+
"grad_norm": 1.572223424911499,
|
| 168 |
+
"learning_rate": 6.482280561460986e-05,
|
| 169 |
+
"loss": 1.0263,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.05578800557880056,
|
| 174 |
+
"grad_norm": 1.4538493156433105,
|
| 175 |
+
"learning_rate": 6.479661545490052e-05,
|
| 176 |
+
"loss": 1.0008,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.05811250581125058,
|
| 181 |
+
"grad_norm": 1.4649128913879395,
|
| 182 |
+
"learning_rate": 6.47686268775627e-05,
|
| 183 |
+
"loss": 1.0391,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06043700604370061,
|
| 188 |
+
"grad_norm": 1.4427037239074707,
|
| 189 |
+
"learning_rate": 6.473884144112351e-05,
|
| 190 |
+
"loss": 0.9842,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06276150627615062,
|
| 195 |
+
"grad_norm": 1.5365402698516846,
|
| 196 |
+
"learning_rate": 6.470726080416707e-05,
|
| 197 |
+
"loss": 1.027,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.06508600650860065,
|
| 202 |
+
"grad_norm": 1.58499014377594,
|
| 203 |
+
"learning_rate": 6.467388672524219e-05,
|
| 204 |
+
"loss": 1.0134,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.06741050674105067,
|
| 209 |
+
"grad_norm": 1.4400190114974976,
|
| 210 |
+
"learning_rate": 6.463872106276442e-05,
|
| 211 |
+
"loss": 0.9216,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.0697350069735007,
|
| 216 |
+
"grad_norm": 1.5131999254226685,
|
| 217 |
+
"learning_rate": 6.46017657749125e-05,
|
| 218 |
+
"loss": 0.9949,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07205950720595072,
|
| 223 |
+
"grad_norm": 1.4283956289291382,
|
| 224 |
+
"learning_rate": 6.456302291951948e-05,
|
| 225 |
+
"loss": 0.9467,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.07438400743840075,
|
| 230 |
+
"grad_norm": 1.48639714717865,
|
| 231 |
+
"learning_rate": 6.452249465395796e-05,
|
| 232 |
+
"loss": 1.007,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.07670850767085077,
|
| 237 |
+
"grad_norm": 1.3868178129196167,
|
| 238 |
+
"learning_rate": 6.448018323502007e-05,
|
| 239 |
+
"loss": 0.9728,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.07903300790330078,
|
| 244 |
+
"grad_norm": 1.442160725593567,
|
| 245 |
+
"learning_rate": 6.443609101879175e-05,
|
| 246 |
+
"loss": 0.9678,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.08135750813575081,
|
| 251 |
+
"grad_norm": 1.334540843963623,
|
| 252 |
+
"learning_rate": 6.439022046052158e-05,
|
| 253 |
+
"loss": 0.9515,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.08368200836820083,
|
| 258 |
+
"grad_norm": 1.3423985242843628,
|
| 259 |
+
"learning_rate": 6.434257411448404e-05,
|
| 260 |
+
"loss": 0.9431,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.08600650860065086,
|
| 265 |
+
"grad_norm": 1.3876471519470215,
|
| 266 |
+
"learning_rate": 6.429315463383725e-05,
|
| 267 |
+
"loss": 0.9427,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.08833100883310088,
|
| 272 |
+
"grad_norm": 1.3787826299667358,
|
| 273 |
+
"learning_rate": 6.424196477047534e-05,
|
| 274 |
+
"loss": 0.956,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.09065550906555091,
|
| 279 |
+
"grad_norm": 1.393442153930664,
|
| 280 |
+
"learning_rate": 6.418900737487508e-05,
|
| 281 |
+
"loss": 0.9742,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.09298000929800093,
|
| 286 |
+
"grad_norm": 1.3465863466262817,
|
| 287 |
+
"learning_rate": 6.413428539593724e-05,
|
| 288 |
+
"loss": 0.9512,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.09530450953045096,
|
| 293 |
+
"grad_norm": 1.3853354454040527,
|
| 294 |
+
"learning_rate": 6.407780188082231e-05,
|
| 295 |
+
"loss": 0.9802,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.09762900976290098,
|
| 300 |
+
"grad_norm": 1.506191372871399,
|
| 301 |
+
"learning_rate": 6.401955997478094e-05,
|
| 302 |
+
"loss": 0.9839,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.099953509995351,
|
| 307 |
+
"grad_norm": 1.346642255783081,
|
| 308 |
+
"learning_rate": 6.395956292097865e-05,
|
| 309 |
+
"loss": 0.9612,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.10227801022780102,
|
| 314 |
+
"grad_norm": 1.4252711534500122,
|
| 315 |
+
"learning_rate": 6.389781406031533e-05,
|
| 316 |
+
"loss": 0.966,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.10460251046025104,
|
| 321 |
+
"grad_norm": 1.3196333646774292,
|
| 322 |
+
"learning_rate": 6.383431683123921e-05,
|
| 323 |
+
"loss": 0.9783,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.10692701069270107,
|
| 328 |
+
"grad_norm": 1.334774136543274,
|
| 329 |
+
"learning_rate": 6.376907476955534e-05,
|
| 330 |
+
"loss": 0.9661,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.10925151092515109,
|
| 335 |
+
"grad_norm": 1.3319814205169678,
|
| 336 |
+
"learning_rate": 6.370209150822871e-05,
|
| 337 |
+
"loss": 0.9326,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.11157601115760112,
|
| 342 |
+
"grad_norm": 1.3688098192214966,
|
| 343 |
+
"learning_rate": 6.3633370777182e-05,
|
| 344 |
+
"loss": 0.9141,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.11390051139005114,
|
| 349 |
+
"grad_norm": 1.375450849533081,
|
| 350 |
+
"learning_rate": 6.356291640308783e-05,
|
| 351 |
+
"loss": 0.937,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.11622501162250116,
|
| 356 |
+
"grad_norm": 1.4250009059906006,
|
| 357 |
+
"learning_rate": 6.349073230915569e-05,
|
| 358 |
+
"loss": 0.9604,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.11854951185495119,
|
| 363 |
+
"grad_norm": 1.3212535381317139,
|
| 364 |
+
"learning_rate": 6.34168225149135e-05,
|
| 365 |
+
"loss": 0.9432,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.12087401208740121,
|
| 370 |
+
"grad_norm": 1.3916277885437012,
|
| 371 |
+
"learning_rate": 6.334119113598371e-05,
|
| 372 |
+
"loss": 1.0155,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.12319851231985124,
|
| 377 |
+
"grad_norm": 1.278640627861023,
|
| 378 |
+
"learning_rate": 6.326384238385426e-05,
|
| 379 |
+
"loss": 0.9462,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.12552301255230125,
|
| 384 |
+
"grad_norm": 1.3200207948684692,
|
| 385 |
+
"learning_rate": 6.318478056564394e-05,
|
| 386 |
+
"loss": 0.9506,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.1278475127847513,
|
| 391 |
+
"grad_norm": 1.2866770029067993,
|
| 392 |
+
"learning_rate": 6.310401008386256e-05,
|
| 393 |
+
"loss": 0.9299,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.1301720130172013,
|
| 398 |
+
"grad_norm": 1.3080073595046997,
|
| 399 |
+
"learning_rate": 6.302153543616591e-05,
|
| 400 |
+
"loss": 0.9416,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.13249651324965134,
|
| 405 |
+
"grad_norm": 1.306003451347351,
|
| 406 |
+
"learning_rate": 6.293736121510519e-05,
|
| 407 |
+
"loss": 0.96,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.13482101348210135,
|
| 412 |
+
"grad_norm": 1.2852219343185425,
|
| 413 |
+
"learning_rate": 6.285149210787132e-05,
|
| 414 |
+
"loss": 0.9077,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.13714551371455136,
|
| 419 |
+
"grad_norm": 1.312796711921692,
|
| 420 |
+
"learning_rate": 6.276393289603395e-05,
|
| 421 |
+
"loss": 0.96,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.1394700139470014,
|
| 426 |
+
"grad_norm": 1.3223918676376343,
|
| 427 |
+
"learning_rate": 6.267468845527521e-05,
|
| 428 |
+
"loss": 0.9349,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.1417945141794514,
|
| 433 |
+
"grad_norm": 1.2744101285934448,
|
| 434 |
+
"learning_rate": 6.258376375511813e-05,
|
| 435 |
+
"loss": 0.9634,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.14411901441190145,
|
| 440 |
+
"grad_norm": 1.2165145874023438,
|
| 441 |
+
"learning_rate": 6.249116385865003e-05,
|
| 442 |
+
"loss": 0.883,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.14644351464435146,
|
| 447 |
+
"grad_norm": 1.3080806732177734,
|
| 448 |
+
"learning_rate": 6.239689392224053e-05,
|
| 449 |
+
"loss": 0.9183,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.1487680148768015,
|
| 454 |
+
"grad_norm": 1.3560563325881958,
|
| 455 |
+
"learning_rate": 6.230095919525438e-05,
|
| 456 |
+
"loss": 0.8918,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.1510925151092515,
|
| 461 |
+
"grad_norm": 1.3362810611724854,
|
| 462 |
+
"learning_rate": 6.220336501975922e-05,
|
| 463 |
+
"loss": 0.9218,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.15341701534170155,
|
| 468 |
+
"grad_norm": 1.4007967710494995,
|
| 469 |
+
"learning_rate": 6.21041168302281e-05,
|
| 470 |
+
"loss": 0.9743,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.15574151557415156,
|
| 475 |
+
"grad_norm": 1.2750542163848877,
|
| 476 |
+
"learning_rate": 6.200322015323678e-05,
|
| 477 |
+
"loss": 0.9101,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.15806601580660157,
|
| 482 |
+
"grad_norm": 1.3171645402908325,
|
| 483 |
+
"learning_rate": 6.190068060715615e-05,
|
| 484 |
+
"loss": 0.9494,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.1603905160390516,
|
| 489 |
+
"grad_norm": 1.2462354898452759,
|
| 490 |
+
"learning_rate": 6.179650390183923e-05,
|
| 491 |
+
"loss": 0.9074,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.16271501627150162,
|
| 496 |
+
"grad_norm": 1.2816126346588135,
|
| 497 |
+
"learning_rate": 6.169069583830324e-05,
|
| 498 |
+
"loss": 0.9672,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.16503951650395166,
|
| 503 |
+
"grad_norm": 1.360295057296753,
|
| 504 |
+
"learning_rate": 6.158326230840664e-05,
|
| 505 |
+
"loss": 0.9216,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.16736401673640167,
|
| 510 |
+
"grad_norm": 1.454693078994751,
|
| 511 |
+
"learning_rate": 6.1474209294521e-05,
|
| 512 |
+
"loss": 0.9648,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.1696885169688517,
|
| 517 |
+
"grad_norm": 1.2523744106292725,
|
| 518 |
+
"learning_rate": 6.136354286919789e-05,
|
| 519 |
+
"loss": 0.9098,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.17201301720130172,
|
| 524 |
+
"grad_norm": 1.265788197517395,
|
| 525 |
+
"learning_rate": 6.125126919483069e-05,
|
| 526 |
+
"loss": 0.8955,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.17433751743375175,
|
| 531 |
+
"grad_norm": 1.2782336473464966,
|
| 532 |
+
"learning_rate": 6.113739452331156e-05,
|
| 533 |
+
"loss": 0.9257,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.17666201766620176,
|
| 538 |
+
"grad_norm": 1.3308346271514893,
|
| 539 |
+
"learning_rate": 6.102192519568312e-05,
|
| 540 |
+
"loss": 0.9581,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.1789865178986518,
|
| 545 |
+
"grad_norm": 1.3193368911743164,
|
| 546 |
+
"learning_rate": 6.0904867641785524e-05,
|
| 547 |
+
"loss": 0.9376,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.18131101813110181,
|
| 552 |
+
"grad_norm": 1.2379558086395264,
|
| 553 |
+
"learning_rate": 6.078622837989834e-05,
|
| 554 |
+
"loss": 0.8897,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.18363551836355183,
|
| 559 |
+
"grad_norm": 1.314134120941162,
|
| 560 |
+
"learning_rate": 6.0666014016377575e-05,
|
| 561 |
+
"loss": 0.9065,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.18596001859600186,
|
| 566 |
+
"grad_norm": 1.3453565835952759,
|
| 567 |
+
"learning_rate": 6.0544231245287847e-05,
|
| 568 |
+
"loss": 0.9992,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.18828451882845187,
|
| 573 |
+
"grad_norm": 1.2375950813293457,
|
| 574 |
+
"learning_rate": 6.042088684802962e-05,
|
| 575 |
+
"loss": 0.8979,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.1906090190609019,
|
| 580 |
+
"grad_norm": 1.2984182834625244,
|
| 581 |
+
"learning_rate": 6.029598769296152e-05,
|
| 582 |
+
"loss": 0.9209,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.19293351929335192,
|
| 587 |
+
"grad_norm": 1.264028787612915,
|
| 588 |
+
"learning_rate": 6.0169540735017986e-05,
|
| 589 |
+
"loss": 0.9202,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.19525801952580196,
|
| 594 |
+
"grad_norm": 1.2835426330566406,
|
| 595 |
+
"learning_rate": 6.004155301532189e-05,
|
| 596 |
+
"loss": 0.9103,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.19758251975825197,
|
| 601 |
+
"grad_norm": 1.3315404653549194,
|
| 602 |
+
"learning_rate": 5.991203166079251e-05,
|
| 603 |
+
"loss": 0.9246,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.199907019990702,
|
| 608 |
+
"grad_norm": 1.2901430130004883,
|
| 609 |
+
"learning_rate": 5.978098388374867e-05,
|
| 610 |
+
"loss": 0.9187,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.20223152022315202,
|
| 615 |
+
"grad_norm": 1.2905139923095703,
|
| 616 |
+
"learning_rate": 5.964841698150709e-05,
|
| 617 |
+
"loss": 0.8755,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.20455602045560203,
|
| 622 |
+
"grad_norm": 1.284635305404663,
|
| 623 |
+
"learning_rate": 5.95143383359761e-05,
|
| 624 |
+
"loss": 0.9376,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.20688052068805207,
|
| 629 |
+
"grad_norm": 1.274927020072937,
|
| 630 |
+
"learning_rate": 5.93787554132445e-05,
|
| 631 |
+
"loss": 0.8953,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.20920502092050208,
|
| 636 |
+
"grad_norm": 1.275505542755127,
|
| 637 |
+
"learning_rate": 5.92416757631659e-05,
|
| 638 |
+
"loss": 0.9458,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.21152952115295212,
|
| 643 |
+
"grad_norm": 1.249010443687439,
|
| 644 |
+
"learning_rate": 5.910310701893825e-05,
|
| 645 |
+
"loss": 0.9195,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.21385402138540213,
|
| 650 |
+
"grad_norm": 1.2102458477020264,
|
| 651 |
+
"learning_rate": 5.896305689667885e-05,
|
| 652 |
+
"loss": 0.9237,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.21617852161785217,
|
| 657 |
+
"grad_norm": 1.2785658836364746,
|
| 658 |
+
"learning_rate": 5.882153319499459e-05,
|
| 659 |
+
"loss": 0.9474,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.21850302185030218,
|
| 664 |
+
"grad_norm": 1.2559230327606201,
|
| 665 |
+
"learning_rate": 5.867854379454777e-05,
|
| 666 |
+
"loss": 0.8634,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.22082752208275222,
|
| 671 |
+
"grad_norm": 1.1916835308074951,
|
| 672 |
+
"learning_rate": 5.8534096657617286e-05,
|
| 673 |
+
"loss": 0.8826,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.22315202231520223,
|
| 678 |
+
"grad_norm": 1.2259260416030884,
|
| 679 |
+
"learning_rate": 5.838819982765514e-05,
|
| 680 |
+
"loss": 0.8864,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.22547652254765224,
|
| 685 |
+
"grad_norm": 1.2614275217056274,
|
| 686 |
+
"learning_rate": 5.824086142883867e-05,
|
| 687 |
+
"loss": 0.8649,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.22780102278010228,
|
| 692 |
+
"grad_norm": 1.3275364637374878,
|
| 693 |
+
"learning_rate": 5.8092089665618096e-05,
|
| 694 |
+
"loss": 0.9218,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2301255230125523,
|
| 699 |
+
"grad_norm": 1.2847840785980225,
|
| 700 |
+
"learning_rate": 5.7941892822259686e-05,
|
| 701 |
+
"loss": 0.8976,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.23245002324500233,
|
| 706 |
+
"grad_norm": 1.2817707061767578,
|
| 707 |
+
"learning_rate": 5.779027926238441e-05,
|
| 708 |
+
"loss": 0.9278,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.23477452347745234,
|
| 713 |
+
"grad_norm": 1.2560011148452759,
|
| 714 |
+
"learning_rate": 5.7637257428502266e-05,
|
| 715 |
+
"loss": 0.9113,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.23709902370990238,
|
| 720 |
+
"grad_norm": 1.222548484802246,
|
| 721 |
+
"learning_rate": 5.748283584154215e-05,
|
| 722 |
+
"loss": 0.8772,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.2394235239423524,
|
| 727 |
+
"grad_norm": 1.2913312911987305,
|
| 728 |
+
"learning_rate": 5.7327023100377346e-05,
|
| 729 |
+
"loss": 0.9342,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.24174802417480243,
|
| 734 |
+
"grad_norm": 1.1847671270370483,
|
| 735 |
+
"learning_rate": 5.716982788134676e-05,
|
| 736 |
+
"loss": 0.8713,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.24407252440725244,
|
| 741 |
+
"grad_norm": 1.2654389142990112,
|
| 742 |
+
"learning_rate": 5.70112589377717e-05,
|
| 743 |
+
"loss": 0.8818,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.24639702463970248,
|
| 748 |
+
"grad_norm": 1.2268662452697754,
|
| 749 |
+
"learning_rate": 5.6851325099468526e-05,
|
| 750 |
+
"loss": 0.8754,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2487215248721525,
|
| 755 |
+
"grad_norm": 1.30552339553833,
|
| 756 |
+
"learning_rate": 5.669003527225696e-05,
|
| 757 |
+
"loss": 0.946,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.2510460251046025,
|
| 762 |
+
"grad_norm": 1.239740252494812,
|
| 763 |
+
"learning_rate": 5.65273984374641e-05,
|
| 764 |
+
"loss": 0.9061,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.2533705253370525,
|
| 769 |
+
"grad_norm": 1.2037532329559326,
|
| 770 |
+
"learning_rate": 5.6363423651424416e-05,
|
| 771 |
+
"loss": 0.9221,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2556950255695026,
|
| 776 |
+
"grad_norm": 1.3354305028915405,
|
| 777 |
+
"learning_rate": 5.6198120044975337e-05,
|
| 778 |
+
"loss": 0.9359,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.2580195258019526,
|
| 783 |
+
"grad_norm": 1.3739912509918213,
|
| 784 |
+
"learning_rate": 5.603149682294886e-05,
|
| 785 |
+
"loss": 0.9305,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.2603440260344026,
|
| 790 |
+
"grad_norm": 1.2662415504455566,
|
| 791 |
+
"learning_rate": 5.586356326365904e-05,
|
| 792 |
+
"loss": 0.8996,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.2626685262668526,
|
| 797 |
+
"grad_norm": 1.3930978775024414,
|
| 798 |
+
"learning_rate": 5.5694328718385214e-05,
|
| 799 |
+
"loss": 0.9477,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.2649930264993027,
|
| 804 |
+
"grad_norm": 1.276788592338562,
|
| 805 |
+
"learning_rate": 5.552380261085135e-05,
|
| 806 |
+
"loss": 0.886,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.2673175267317527,
|
| 811 |
+
"grad_norm": 1.2634483575820923,
|
| 812 |
+
"learning_rate": 5.5351994436701326e-05,
|
| 813 |
+
"loss": 0.9207,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.2696420269642027,
|
| 818 |
+
"grad_norm": 1.3280141353607178,
|
| 819 |
+
"learning_rate": 5.5178913762970074e-05,
|
| 820 |
+
"loss": 0.916,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.2719665271966527,
|
| 825 |
+
"grad_norm": 1.2456930875778198,
|
| 826 |
+
"learning_rate": 5.500457022755095e-05,
|
| 827 |
+
"loss": 0.8524,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.2742910274291027,
|
| 832 |
+
"grad_norm": 1.2636594772338867,
|
| 833 |
+
"learning_rate": 5.4828973538658974e-05,
|
| 834 |
+
"loss": 0.9525,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.2766155276615528,
|
| 839 |
+
"grad_norm": 1.223232626914978,
|
| 840 |
+
"learning_rate": 5.465213347429026e-05,
|
| 841 |
+
"loss": 0.8556,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.2789400278940028,
|
| 846 |
+
"grad_norm": 1.414523720741272,
|
| 847 |
+
"learning_rate": 5.4474059881677566e-05,
|
| 848 |
+
"loss": 0.9719,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.2812645281264528,
|
| 853 |
+
"grad_norm": 1.2748632431030273,
|
| 854 |
+
"learning_rate": 5.42947626767419e-05,
|
| 855 |
+
"loss": 0.9076,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.2835890283589028,
|
| 860 |
+
"grad_norm": 1.1875295639038086,
|
| 861 |
+
"learning_rate": 5.411425184354042e-05,
|
| 862 |
+
"loss": 0.8475,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.2859135285913529,
|
| 867 |
+
"grad_norm": 1.168205976486206,
|
| 868 |
+
"learning_rate": 5.393253743371041e-05,
|
| 869 |
+
"loss": 0.8827,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.2882380288238029,
|
| 874 |
+
"grad_norm": 1.2521716356277466,
|
| 875 |
+
"learning_rate": 5.3749629565909614e-05,
|
| 876 |
+
"loss": 0.9238,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.2905625290562529,
|
| 881 |
+
"grad_norm": 1.3135292530059814,
|
| 882 |
+
"learning_rate": 5.356553842525278e-05,
|
| 883 |
+
"loss": 0.9562,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.2928870292887029,
|
| 888 |
+
"grad_norm": 1.1815370321273804,
|
| 889 |
+
"learning_rate": 5.3380274262744447e-05,
|
| 890 |
+
"loss": 0.8571,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.2952115295211529,
|
| 895 |
+
"grad_norm": 1.2776923179626465,
|
| 896 |
+
"learning_rate": 5.319384739470821e-05,
|
| 897 |
+
"loss": 0.9234,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.297536029753603,
|
| 902 |
+
"grad_norm": 1.3336985111236572,
|
| 903 |
+
"learning_rate": 5.300626820221224e-05,
|
| 904 |
+
"loss": 0.9257,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.299860529986053,
|
| 909 |
+
"grad_norm": 1.2534116506576538,
|
| 910 |
+
"learning_rate": 5.2817547130491154e-05,
|
| 911 |
+
"loss": 0.8918,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.302185030218503,
|
| 916 |
+
"grad_norm": 1.2334403991699219,
|
| 917 |
+
"learning_rate": 5.262769468836446e-05,
|
| 918 |
+
"loss": 0.8407,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.304509530450953,
|
| 923 |
+
"grad_norm": 1.2081224918365479,
|
| 924 |
+
"learning_rate": 5.2436721447651325e-05,
|
| 925 |
+
"loss": 0.8976,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3068340306834031,
|
| 930 |
+
"grad_norm": 1.3717191219329834,
|
| 931 |
+
"learning_rate": 5.224463804258194e-05,
|
| 932 |
+
"loss": 0.9156,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3091585309158531,
|
| 937 |
+
"grad_norm": 1.2610055208206177,
|
| 938 |
+
"learning_rate": 5.205145516920531e-05,
|
| 939 |
+
"loss": 0.925,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.3114830311483031,
|
| 944 |
+
"grad_norm": 1.2727302312850952,
|
| 945 |
+
"learning_rate": 5.185718358479369e-05,
|
| 946 |
+
"loss": 0.9217,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3138075313807531,
|
| 951 |
+
"grad_norm": 1.2891085147857666,
|
| 952 |
+
"learning_rate": 5.166183410724353e-05,
|
| 953 |
+
"loss": 0.9583,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.31613203161320313,
|
| 958 |
+
"grad_norm": 1.1891586780548096,
|
| 959 |
+
"learning_rate": 5.146541761447313e-05,
|
| 960 |
+
"loss": 0.8329,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.3184565318456532,
|
| 965 |
+
"grad_norm": 1.203373670578003,
|
| 966 |
+
"learning_rate": 5.1267945043816886e-05,
|
| 967 |
+
"loss": 0.8716,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.3207810320781032,
|
| 972 |
+
"grad_norm": 1.2259798049926758,
|
| 973 |
+
"learning_rate": 5.106942739141625e-05,
|
| 974 |
+
"loss": 0.9246,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.3231055323105532,
|
| 979 |
+
"grad_norm": 1.237679123878479,
|
| 980 |
+
"learning_rate": 5.086987571160746e-05,
|
| 981 |
+
"loss": 0.8769,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.32543003254300323,
|
| 986 |
+
"grad_norm": 1.2703795433044434,
|
| 987 |
+
"learning_rate": 5.0669301116305886e-05,
|
| 988 |
+
"loss": 0.8838,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3277545327754533,
|
| 993 |
+
"grad_norm": 1.2701797485351562,
|
| 994 |
+
"learning_rate": 5.0467714774387386e-05,
|
| 995 |
+
"loss": 0.8438,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.3300790330079033,
|
| 1000 |
+
"grad_norm": 1.2709927558898926,
|
| 1001 |
+
"learning_rate": 5.0265127911066295e-05,
|
| 1002 |
+
"loss": 0.9252,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3324035332403533,
|
| 1007 |
+
"grad_norm": 1.291172981262207,
|
| 1008 |
+
"learning_rate": 5.0061551807270376e-05,
|
| 1009 |
+
"loss": 0.93,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.33472803347280333,
|
| 1014 |
+
"grad_norm": 1.2257442474365234,
|
| 1015 |
+
"learning_rate": 4.9856997799012677e-05,
|
| 1016 |
+
"loss": 0.8957,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.33705253370525334,
|
| 1021 |
+
"grad_norm": 1.2257241010665894,
|
| 1022 |
+
"learning_rate": 4.965147727676027e-05,
|
| 1023 |
+
"loss": 0.9213,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3393770339377034,
|
| 1028 |
+
"grad_norm": 1.16828191280365,
|
| 1029 |
+
"learning_rate": 4.944500168479995e-05,
|
| 1030 |
+
"loss": 0.8625,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.3417015341701534,
|
| 1035 |
+
"grad_norm": 1.2331054210662842,
|
| 1036 |
+
"learning_rate": 4.9237582520600986e-05,
|
| 1037 |
+
"loss": 0.9318,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.34402603440260343,
|
| 1042 |
+
"grad_norm": 1.2196905612945557,
|
| 1043 |
+
"learning_rate": 4.902923133417496e-05,
|
| 1044 |
+
"loss": 0.8709,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.34635053463505344,
|
| 1049 |
+
"grad_norm": 1.2139414548873901,
|
| 1050 |
+
"learning_rate": 4.8819959727432496e-05,
|
| 1051 |
+
"loss": 0.9081,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.3486750348675035,
|
| 1056 |
+
"grad_norm": 1.204951524734497,
|
| 1057 |
+
"learning_rate": 4.86097793535373e-05,
|
| 1058 |
+
"loss": 0.8331,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.3509995350999535,
|
| 1063 |
+
"grad_norm": 1.1808617115020752,
|
| 1064 |
+
"learning_rate": 4.839870191625722e-05,
|
| 1065 |
+
"loss": 0.865,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.35332403533240353,
|
| 1070 |
+
"grad_norm": 1.1742596626281738,
|
| 1071 |
+
"learning_rate": 4.818673916931252e-05,
|
| 1072 |
+
"loss": 0.8266,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.35564853556485354,
|
| 1077 |
+
"grad_norm": 1.2377365827560425,
|
| 1078 |
+
"learning_rate": 4.797390291572145e-05,
|
| 1079 |
+
"loss": 0.841,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.3579730357973036,
|
| 1084 |
+
"grad_norm": 1.2187622785568237,
|
| 1085 |
+
"learning_rate": 4.7760205007142904e-05,
|
| 1086 |
+
"loss": 0.875,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.3602975360297536,
|
| 1091 |
+
"grad_norm": 1.182664394378662,
|
| 1092 |
+
"learning_rate": 4.754565734321654e-05,
|
| 1093 |
+
"loss": 0.8742,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.36262203626220363,
|
| 1098 |
+
"grad_norm": 1.2196147441864014,
|
| 1099 |
+
"learning_rate": 4.733027187090013e-05,
|
| 1100 |
+
"loss": 0.8849,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.36494653649465364,
|
| 1105 |
+
"grad_norm": 1.2542427778244019,
|
| 1106 |
+
"learning_rate": 4.711406058380429e-05,
|
| 1107 |
+
"loss": 0.847,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.36727103672710365,
|
| 1112 |
+
"grad_norm": 1.2659164667129517,
|
| 1113 |
+
"learning_rate": 4.689703552152467e-05,
|
| 1114 |
+
"loss": 0.8557,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.3695955369595537,
|
| 1119 |
+
"grad_norm": 1.1970021724700928,
|
| 1120 |
+
"learning_rate": 4.667920876897147e-05,
|
| 1121 |
+
"loss": 0.8474,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.3719200371920037,
|
| 1126 |
+
"grad_norm": 1.2822171449661255,
|
| 1127 |
+
"learning_rate": 4.646059245569654e-05,
|
| 1128 |
+
"loss": 0.9273,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.37424453742445374,
|
| 1133 |
+
"grad_norm": 1.233355164527893,
|
| 1134 |
+
"learning_rate": 4.6241198755217925e-05,
|
| 1135 |
+
"loss": 0.8528,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.37656903765690375,
|
| 1140 |
+
"grad_norm": 1.2058496475219727,
|
| 1141 |
+
"learning_rate": 4.602103988434206e-05,
|
| 1142 |
+
"loss": 0.86,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.3788935378893538,
|
| 1147 |
+
"grad_norm": 1.3541826009750366,
|
| 1148 |
+
"learning_rate": 4.5800128102483376e-05,
|
| 1149 |
+
"loss": 0.9412,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.3812180381218038,
|
| 1154 |
+
"grad_norm": 1.1209250688552856,
|
| 1155 |
+
"learning_rate": 4.5578475710981745e-05,
|
| 1156 |
+
"loss": 0.8332,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.38354253835425384,
|
| 1161 |
+
"grad_norm": 1.2058111429214478,
|
| 1162 |
+
"learning_rate": 4.535609505241742e-05,
|
| 1163 |
+
"loss": 0.857,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.38586703858670385,
|
| 1168 |
+
"grad_norm": 1.2212828397750854,
|
| 1169 |
+
"learning_rate": 4.513299850992376e-05,
|
| 1170 |
+
"loss": 0.8301,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.38819153881915386,
|
| 1175 |
+
"grad_norm": 1.2507742643356323,
|
| 1176 |
+
"learning_rate": 4.4909198506497725e-05,
|
| 1177 |
+
"loss": 0.88,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.3905160390516039,
|
| 1182 |
+
"grad_norm": 1.1851963996887207,
|
| 1183 |
+
"learning_rate": 4.4684707504308036e-05,
|
| 1184 |
+
"loss": 0.8504,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.39284053928405394,
|
| 1189 |
+
"grad_norm": 1.1980507373809814,
|
| 1190 |
+
"learning_rate": 4.4459538004001276e-05,
|
| 1191 |
+
"loss": 0.8413,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.39516503951650395,
|
| 1196 |
+
"grad_norm": 1.210808515548706,
|
| 1197 |
+
"learning_rate": 4.42337025440058e-05,
|
| 1198 |
+
"loss": 0.8669,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.39748953974895396,
|
| 1203 |
+
"grad_norm": 1.1731034517288208,
|
| 1204 |
+
"learning_rate": 4.400721369983352e-05,
|
| 1205 |
+
"loss": 0.9147,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.399814039981404,
|
| 1210 |
+
"grad_norm": 1.2603881359100342,
|
| 1211 |
+
"learning_rate": 4.3780084083379654e-05,
|
| 1212 |
+
"loss": 0.8829,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.40213854021385403,
|
| 1217 |
+
"grad_norm": 1.2504175901412964,
|
| 1218 |
+
"learning_rate": 4.3552326342220444e-05,
|
| 1219 |
+
"loss": 0.8924,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.40446304044630405,
|
| 1224 |
+
"grad_norm": 1.1380208730697632,
|
| 1225 |
+
"learning_rate": 4.332395315890887e-05,
|
| 1226 |
+
"loss": 0.7826,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.40678754067875406,
|
| 1231 |
+
"grad_norm": 1.2040411233901978,
|
| 1232 |
+
"learning_rate": 4.309497725026844e-05,
|
| 1233 |
+
"loss": 0.8476,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.40911204091120407,
|
| 1238 |
+
"grad_norm": 1.2187401056289673,
|
| 1239 |
+
"learning_rate": 4.2865411366685064e-05,
|
| 1240 |
+
"loss": 0.8605,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.41143654114365413,
|
| 1245 |
+
"grad_norm": 1.2218343019485474,
|
| 1246 |
+
"learning_rate": 4.2635268291397054e-05,
|
| 1247 |
+
"loss": 0.8408,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.41376104137610414,
|
| 1252 |
+
"grad_norm": 1.3197338581085205,
|
| 1253 |
+
"learning_rate": 4.240456083978329e-05,
|
| 1254 |
+
"loss": 0.8658,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.41608554160855415,
|
| 1259 |
+
"grad_norm": 1.3240693807601929,
|
| 1260 |
+
"learning_rate": 4.2173301858649605e-05,
|
| 1261 |
+
"loss": 0.9434,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.41841004184100417,
|
| 1266 |
+
"grad_norm": 1.2702267169952393,
|
| 1267 |
+
"learning_rate": 4.1941504225513445e-05,
|
| 1268 |
+
"loss": 0.8643,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.42073454207345423,
|
| 1273 |
+
"grad_norm": 1.2468559741973877,
|
| 1274 |
+
"learning_rate": 4.1709180847886745e-05,
|
| 1275 |
+
"loss": 0.8797,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.42305904230590424,
|
| 1280 |
+
"grad_norm": 1.2787050008773804,
|
| 1281 |
+
"learning_rate": 4.147634466255721e-05,
|
| 1282 |
+
"loss": 0.8794,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.42538354253835425,
|
| 1287 |
+
"grad_norm": 1.222352147102356,
|
| 1288 |
+
"learning_rate": 4.124300863486793e-05,
|
| 1289 |
+
"loss": 0.8459,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.42770804277080426,
|
| 1294 |
+
"grad_norm": 1.203860878944397,
|
| 1295 |
+
"learning_rate": 4.100918575799543e-05,
|
| 1296 |
+
"loss": 0.9002,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.4300325430032543,
|
| 1301 |
+
"grad_norm": 1.1671602725982666,
|
| 1302 |
+
"learning_rate": 4.077488905222613e-05,
|
| 1303 |
+
"loss": 0.8686,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.43235704323570434,
|
| 1308 |
+
"grad_norm": 1.172710657119751,
|
| 1309 |
+
"learning_rate": 4.0540131564231346e-05,
|
| 1310 |
+
"loss": 0.8665,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.43468154346815435,
|
| 1315 |
+
"grad_norm": 1.196799635887146,
|
| 1316 |
+
"learning_rate": 4.0304926366340746e-05,
|
| 1317 |
+
"loss": 0.874,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.43700604370060436,
|
| 1322 |
+
"grad_norm": 1.3162767887115479,
|
| 1323 |
+
"learning_rate": 4.006928655581446e-05,
|
| 1324 |
+
"loss": 0.9128,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.4393305439330544,
|
| 1329 |
+
"grad_norm": 1.1845192909240723,
|
| 1330 |
+
"learning_rate": 3.983322525411379e-05,
|
| 1331 |
+
"loss": 0.8354,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.44165504416550444,
|
| 1336 |
+
"grad_norm": 1.2140437364578247,
|
| 1337 |
+
"learning_rate": 3.959675560617049e-05,
|
| 1338 |
+
"loss": 0.8716,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.44397954439795445,
|
| 1343 |
+
"grad_norm": 1.1984777450561523,
|
| 1344 |
+
"learning_rate": 3.9359890779654846e-05,
|
| 1345 |
+
"loss": 0.8063,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.44630404463040446,
|
| 1350 |
+
"grad_norm": 1.1856093406677246,
|
| 1351 |
+
"learning_rate": 3.91226439642424e-05,
|
| 1352 |
+
"loss": 0.7776,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.4486285448628545,
|
| 1357 |
+
"grad_norm": 1.2061065435409546,
|
| 1358 |
+
"learning_rate": 3.8885028370879544e-05,
|
| 1359 |
+
"loss": 0.8475,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.4509530450953045,
|
| 1364 |
+
"grad_norm": 1.2898370027542114,
|
| 1365 |
+
"learning_rate": 3.864705723104786e-05,
|
| 1366 |
+
"loss": 0.8788,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.45327754532775455,
|
| 1371 |
+
"grad_norm": 1.2373768091201782,
|
| 1372 |
+
"learning_rate": 3.840874379602728e-05,
|
| 1373 |
+
"loss": 0.8424,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.45560204556020456,
|
| 1378 |
+
"grad_norm": 1.2365310192108154,
|
| 1379 |
+
"learning_rate": 3.817010133615827e-05,
|
| 1380 |
+
"loss": 0.8429,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.45792654579265457,
|
| 1385 |
+
"grad_norm": 1.2247048616409302,
|
| 1386 |
+
"learning_rate": 3.793114314010282e-05,
|
| 1387 |
+
"loss": 0.8806,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.4602510460251046,
|
| 1392 |
+
"grad_norm": 1.2044016122817993,
|
| 1393 |
+
"learning_rate": 3.769188251410454e-05,
|
| 1394 |
+
"loss": 0.8633,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.46257554625755465,
|
| 1399 |
+
"grad_norm": 1.1678348779678345,
|
| 1400 |
+
"learning_rate": 3.7452332781247633e-05,
|
| 1401 |
+
"loss": 0.8325,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.46490004649000466,
|
| 1406 |
+
"grad_norm": 1.2637572288513184,
|
| 1407 |
+
"learning_rate": 3.7212507280715056e-05,
|
| 1408 |
+
"loss": 0.8471,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.46722454672245467,
|
| 1413 |
+
"grad_norm": 1.1990036964416504,
|
| 1414 |
+
"learning_rate": 3.6972419367045726e-05,
|
| 1415 |
+
"loss": 0.8821,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.4695490469549047,
|
| 1420 |
+
"grad_norm": 1.1562861204147339,
|
| 1421 |
+
"learning_rate": 3.673208240939088e-05,
|
| 1422 |
+
"loss": 0.8486,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.4718735471873547,
|
| 1427 |
+
"grad_norm": 1.2566421031951904,
|
| 1428 |
+
"learning_rate": 3.6491509790769616e-05,
|
| 1429 |
+
"loss": 0.9117,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.47419804741980476,
|
| 1434 |
+
"grad_norm": 1.2128387689590454,
|
| 1435 |
+
"learning_rate": 3.625071490732366e-05,
|
| 1436 |
+
"loss": 0.8905,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.47652254765225477,
|
| 1441 |
+
"grad_norm": 1.2243783473968506,
|
| 1442 |
+
"learning_rate": 3.6009711167571434e-05,
|
| 1443 |
+
"loss": 0.8362,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.4788470478847048,
|
| 1448 |
+
"grad_norm": 1.2012783288955688,
|
| 1449 |
+
"learning_rate": 3.57685119916614e-05,
|
| 1450 |
+
"loss": 0.841,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.4811715481171548,
|
| 1455 |
+
"grad_norm": 1.1792044639587402,
|
| 1456 |
+
"learning_rate": 3.552713081062478e-05,
|
| 1457 |
+
"loss": 0.8673,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.48349604834960486,
|
| 1462 |
+
"grad_norm": 1.2149136066436768,
|
| 1463 |
+
"learning_rate": 3.528558106562759e-05,
|
| 1464 |
+
"loss": 0.8776,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.48582054858205487,
|
| 1469 |
+
"grad_norm": 1.2262108325958252,
|
| 1470 |
+
"learning_rate": 3.50438762072223e-05,
|
| 1471 |
+
"loss": 0.8598,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.4881450488145049,
|
| 1476 |
+
"grad_norm": 1.1815216541290283,
|
| 1477 |
+
"learning_rate": 3.4802029694598705e-05,
|
| 1478 |
+
"loss": 0.7776,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.4904695490469549,
|
| 1483 |
+
"grad_norm": 1.1577028036117554,
|
| 1484 |
+
"learning_rate": 3.456005499483456e-05,
|
| 1485 |
+
"loss": 0.8072,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.49279404927940496,
|
| 1490 |
+
"grad_norm": 1.1453114748001099,
|
| 1491 |
+
"learning_rate": 3.4317965582145684e-05,
|
| 1492 |
+
"loss": 0.8518,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.49511854951185497,
|
| 1497 |
+
"grad_norm": 1.2426217794418335,
|
| 1498 |
+
"learning_rate": 3.4075774937135554e-05,
|
| 1499 |
+
"loss": 0.88,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.497443049744305,
|
| 1504 |
+
"grad_norm": 1.2424201965332031,
|
| 1505 |
+
"learning_rate": 3.3833496546044746e-05,
|
| 1506 |
+
"loss": 0.8669,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.499767549976755,
|
| 1511 |
+
"grad_norm": 1.2596735954284668,
|
| 1512 |
+
"learning_rate": 3.3591143899999905e-05,
|
| 1513 |
+
"loss": 0.8546,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.502092050209205,
|
| 1518 |
+
"grad_norm": 1.2579164505004883,
|
| 1519 |
+
"learning_rate": 3.334873049426253e-05,
|
| 1520 |
+
"loss": 0.8954,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.504416550441655,
|
| 1525 |
+
"grad_norm": 1.1875728368759155,
|
| 1526 |
+
"learning_rate": 3.3106269827477484e-05,
|
| 1527 |
+
"loss": 0.8319,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.506741050674105,
|
| 1532 |
+
"grad_norm": 1.241857647895813,
|
| 1533 |
+
"learning_rate": 3.286377540092133e-05,
|
| 1534 |
+
"loss": 0.9209,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5090655509065551,
|
| 1539 |
+
"grad_norm": 1.1753859519958496,
|
| 1540 |
+
"learning_rate": 3.2621260717750515e-05,
|
| 1541 |
+
"loss": 0.8092,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5113900511390052,
|
| 1546 |
+
"grad_norm": 1.1673084497451782,
|
| 1547 |
+
"learning_rate": 3.237873928224948e-05,
|
| 1548 |
+
"loss": 0.831,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5137145513714552,
|
| 1553 |
+
"grad_norm": 1.2066866159439087,
|
| 1554 |
+
"learning_rate": 3.213622459907867e-05,
|
| 1555 |
+
"loss": 0.8413,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5160390516039052,
|
| 1560 |
+
"grad_norm": 1.1958924531936646,
|
| 1561 |
+
"learning_rate": 3.189373017252252e-05,
|
| 1562 |
+
"loss": 0.8648,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5183635518363552,
|
| 1567 |
+
"grad_norm": 1.1267814636230469,
|
| 1568 |
+
"learning_rate": 3.165126950573747e-05,
|
| 1569 |
+
"loss": 0.8551,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5206880520688052,
|
| 1574 |
+
"grad_norm": 1.1723332405090332,
|
| 1575 |
+
"learning_rate": 3.140885610000009e-05,
|
| 1576 |
+
"loss": 0.8732,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5230125523012552,
|
| 1581 |
+
"grad_norm": 1.2160817384719849,
|
| 1582 |
+
"learning_rate": 3.116650345395524e-05,
|
| 1583 |
+
"loss": 0.874,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5253370525337052,
|
| 1588 |
+
"grad_norm": 1.1341142654418945,
|
| 1589 |
+
"learning_rate": 3.0924225062864434e-05,
|
| 1590 |
+
"loss": 0.8044,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.5276615527661552,
|
| 1595 |
+
"grad_norm": 1.1896854639053345,
|
| 1596 |
+
"learning_rate": 3.0682034417854324e-05,
|
| 1597 |
+
"loss": 0.8364,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.5299860529986054,
|
| 1602 |
+
"grad_norm": 1.229346513748169,
|
| 1603 |
+
"learning_rate": 3.0439945005165433e-05,
|
| 1604 |
+
"loss": 0.8821,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.5323105532310554,
|
| 1609 |
+
"grad_norm": 1.1989872455596924,
|
| 1610 |
+
"learning_rate": 3.0197970305401306e-05,
|
| 1611 |
+
"loss": 0.8566,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.5346350534635054,
|
| 1616 |
+
"grad_norm": 1.248273491859436,
|
| 1617 |
+
"learning_rate": 2.9956123792777708e-05,
|
| 1618 |
+
"loss": 0.7907,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.5369595536959554,
|
| 1623 |
+
"grad_norm": 1.235098958015442,
|
| 1624 |
+
"learning_rate": 2.9714418934372402e-05,
|
| 1625 |
+
"loss": 0.8838,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.5392840539284054,
|
| 1630 |
+
"grad_norm": 1.2413426637649536,
|
| 1631 |
+
"learning_rate": 2.9472869189375224e-05,
|
| 1632 |
+
"loss": 0.8614,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.5416085541608554,
|
| 1637 |
+
"grad_norm": 1.2023251056671143,
|
| 1638 |
+
"learning_rate": 2.9231488008338592e-05,
|
| 1639 |
+
"loss": 0.8328,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.5439330543933054,
|
| 1644 |
+
"grad_norm": 1.2137963771820068,
|
| 1645 |
+
"learning_rate": 2.899028883242856e-05,
|
| 1646 |
+
"loss": 0.8174,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.5462575546257554,
|
| 1651 |
+
"grad_norm": 1.1812171936035156,
|
| 1652 |
+
"learning_rate": 2.8749285092676343e-05,
|
| 1653 |
+
"loss": 0.8636,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.5485820548582054,
|
| 1658 |
+
"grad_norm": 1.2329221963882446,
|
| 1659 |
+
"learning_rate": 2.8508490209230388e-05,
|
| 1660 |
+
"loss": 0.8419,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.5509065550906556,
|
| 1665 |
+
"grad_norm": 1.176352620124817,
|
| 1666 |
+
"learning_rate": 2.8267917590609115e-05,
|
| 1667 |
+
"loss": 0.8174,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.5532310553231056,
|
| 1672 |
+
"grad_norm": 1.2109792232513428,
|
| 1673 |
+
"learning_rate": 2.802758063295427e-05,
|
| 1674 |
+
"loss": 0.8718,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.5555555555555556,
|
| 1679 |
+
"grad_norm": 1.155416488647461,
|
| 1680 |
+
"learning_rate": 2.7787492719284935e-05,
|
| 1681 |
+
"loss": 0.8365,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.5578800557880056,
|
| 1686 |
+
"grad_norm": 1.2347787618637085,
|
| 1687 |
+
"learning_rate": 2.7547667218752358e-05,
|
| 1688 |
+
"loss": 0.9275,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.5602045560204556,
|
| 1693 |
+
"grad_norm": 1.2063627243041992,
|
| 1694 |
+
"learning_rate": 2.7308117485895463e-05,
|
| 1695 |
+
"loss": 0.8702,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.5625290562529056,
|
| 1700 |
+
"grad_norm": 1.1815749406814575,
|
| 1701 |
+
"learning_rate": 2.7068856859897173e-05,
|
| 1702 |
+
"loss": 0.8222,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.5648535564853556,
|
| 1707 |
+
"grad_norm": 1.1446418762207031,
|
| 1708 |
+
"learning_rate": 2.6829898663841733e-05,
|
| 1709 |
+
"loss": 0.7562,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.5671780567178056,
|
| 1714 |
+
"grad_norm": 1.117566466331482,
|
| 1715 |
+
"learning_rate": 2.659125620397272e-05,
|
| 1716 |
+
"loss": 0.7828,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.5695025569502556,
|
| 1721 |
+
"grad_norm": 1.1796785593032837,
|
| 1722 |
+
"learning_rate": 2.6352942768952144e-05,
|
| 1723 |
+
"loss": 0.8555,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.5718270571827058,
|
| 1728 |
+
"grad_norm": 1.1962978839874268,
|
| 1729 |
+
"learning_rate": 2.6114971629120454e-05,
|
| 1730 |
+
"loss": 0.8359,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.5741515574151558,
|
| 1735 |
+
"grad_norm": 1.2345088720321655,
|
| 1736 |
+
"learning_rate": 2.5877356035757598e-05,
|
| 1737 |
+
"loss": 0.857,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.5764760576476058,
|
| 1742 |
+
"grad_norm": 1.1942895650863647,
|
| 1743 |
+
"learning_rate": 2.564010922034516e-05,
|
| 1744 |
+
"loss": 0.8233,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.5788005578800558,
|
| 1749 |
+
"grad_norm": 1.2165658473968506,
|
| 1750 |
+
"learning_rate": 2.5403244393829505e-05,
|
| 1751 |
+
"loss": 0.8141,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.5811250581125058,
|
| 1756 |
+
"grad_norm": 1.2151166200637817,
|
| 1757 |
+
"learning_rate": 2.5166774745886202e-05,
|
| 1758 |
+
"loss": 0.8538,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.5834495583449558,
|
| 1763 |
+
"grad_norm": 1.203872799873352,
|
| 1764 |
+
"learning_rate": 2.493071344418553e-05,
|
| 1765 |
+
"loss": 0.8177,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.5857740585774058,
|
| 1770 |
+
"grad_norm": 1.2117934226989746,
|
| 1771 |
+
"learning_rate": 2.4695073633659245e-05,
|
| 1772 |
+
"loss": 0.8277,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.5880985588098558,
|
| 1777 |
+
"grad_norm": 1.1932367086410522,
|
| 1778 |
+
"learning_rate": 2.4459868435768644e-05,
|
| 1779 |
+
"loss": 0.8342,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.5904230590423059,
|
| 1784 |
+
"grad_norm": 1.1477673053741455,
|
| 1785 |
+
"learning_rate": 2.4225110947773856e-05,
|
| 1786 |
+
"loss": 0.751,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.592747559274756,
|
| 1791 |
+
"grad_norm": 1.243074655532837,
|
| 1792 |
+
"learning_rate": 2.3990814242004573e-05,
|
| 1793 |
+
"loss": 0.8784,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.595072059507206,
|
| 1798 |
+
"grad_norm": 1.1835278272628784,
|
| 1799 |
+
"learning_rate": 2.375699136513207e-05,
|
| 1800 |
+
"loss": 0.8597,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.597396559739656,
|
| 1805 |
+
"grad_norm": 1.2624067068099976,
|
| 1806 |
+
"learning_rate": 2.3523655337442798e-05,
|
| 1807 |
+
"loss": 0.8519,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.599721059972106,
|
| 1812 |
+
"grad_norm": 1.1431169509887695,
|
| 1813 |
+
"learning_rate": 2.3290819152113256e-05,
|
| 1814 |
+
"loss": 0.8444,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.602045560204556,
|
| 1819 |
+
"grad_norm": 1.1751455068588257,
|
| 1820 |
+
"learning_rate": 2.3058495774486553e-05,
|
| 1821 |
+
"loss": 0.8608,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.604370060437006,
|
| 1826 |
+
"grad_norm": 1.1753759384155273,
|
| 1827 |
+
"learning_rate": 2.2826698141350392e-05,
|
| 1828 |
+
"loss": 0.8567,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.606694560669456,
|
| 1833 |
+
"grad_norm": 1.22267746925354,
|
| 1834 |
+
"learning_rate": 2.2595439160216713e-05,
|
| 1835 |
+
"loss": 0.8707,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.609019060901906,
|
| 1840 |
+
"grad_norm": 1.1453746557235718,
|
| 1841 |
+
"learning_rate": 2.2364731708602947e-05,
|
| 1842 |
+
"loss": 0.831,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6113435611343561,
|
| 1847 |
+
"grad_norm": 1.1886532306671143,
|
| 1848 |
+
"learning_rate": 2.2134588633314937e-05,
|
| 1849 |
+
"loss": 0.763,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6136680613668062,
|
| 1854 |
+
"grad_norm": 1.3295722007751465,
|
| 1855 |
+
"learning_rate": 2.1905022749731563e-05,
|
| 1856 |
+
"loss": 0.8818,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6159925615992562,
|
| 1861 |
+
"grad_norm": 1.124739408493042,
|
| 1862 |
+
"learning_rate": 2.1676046841091133e-05,
|
| 1863 |
+
"loss": 0.7534,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6183170618317062,
|
| 1868 |
+
"grad_norm": 1.2315418720245361,
|
| 1869 |
+
"learning_rate": 2.144767365777955e-05,
|
| 1870 |
+
"loss": 0.8528,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.6206415620641562,
|
| 1875 |
+
"grad_norm": 1.248775601387024,
|
| 1876 |
+
"learning_rate": 2.121991591662034e-05,
|
| 1877 |
+
"loss": 0.8536,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.6229660622966062,
|
| 1882 |
+
"grad_norm": 1.2271955013275146,
|
| 1883 |
+
"learning_rate": 2.0992786300166472e-05,
|
| 1884 |
+
"loss": 0.8731,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.6252905625290562,
|
| 1889 |
+
"grad_norm": 1.1660230159759521,
|
| 1890 |
+
"learning_rate": 2.0766297455994204e-05,
|
| 1891 |
+
"loss": 0.8301,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.6276150627615062,
|
| 1896 |
+
"grad_norm": 1.2100571393966675,
|
| 1897 |
+
"learning_rate": 2.054046199599873e-05,
|
| 1898 |
+
"loss": 0.867,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.6299395629939563,
|
| 1903 |
+
"grad_norm": 1.1913954019546509,
|
| 1904 |
+
"learning_rate": 2.031529249569197e-05,
|
| 1905 |
+
"loss": 0.8322,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.6322640632264063,
|
| 1910 |
+
"grad_norm": 1.1850714683532715,
|
| 1911 |
+
"learning_rate": 2.0090801493502276e-05,
|
| 1912 |
+
"loss": 0.8616,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.6345885634588564,
|
| 1917 |
+
"grad_norm": 1.152343511581421,
|
| 1918 |
+
"learning_rate": 1.986700149007624e-05,
|
| 1919 |
+
"loss": 0.8155,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.6369130636913064,
|
| 1924 |
+
"grad_norm": 1.195383906364441,
|
| 1925 |
+
"learning_rate": 1.964390494758258e-05,
|
| 1926 |
+
"loss": 0.8021,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.6392375639237564,
|
| 1931 |
+
"grad_norm": 1.4834905862808228,
|
| 1932 |
+
"learning_rate": 1.9421524289018253e-05,
|
| 1933 |
+
"loss": 0.8191,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.6415620641562064,
|
| 1938 |
+
"grad_norm": 1.250483751296997,
|
| 1939 |
+
"learning_rate": 1.919987189751662e-05,
|
| 1940 |
+
"loss": 0.8497,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.6438865643886564,
|
| 1945 |
+
"grad_norm": 1.1749329566955566,
|
| 1946 |
+
"learning_rate": 1.8978960115657944e-05,
|
| 1947 |
+
"loss": 0.837,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.6462110646211064,
|
| 1952 |
+
"grad_norm": 1.1642099618911743,
|
| 1953 |
+
"learning_rate": 1.875880124478207e-05,
|
| 1954 |
+
"loss": 0.8454,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.6485355648535565,
|
| 1959 |
+
"grad_norm": 1.2152022123336792,
|
| 1960 |
+
"learning_rate": 1.853940754430346e-05,
|
| 1961 |
+
"loss": 0.8492,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.6508600650860065,
|
| 1966 |
+
"grad_norm": 1.1679767370224,
|
| 1967 |
+
"learning_rate": 1.8320791231028526e-05,
|
| 1968 |
+
"loss": 0.8454,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.6531845653184565,
|
| 1973 |
+
"grad_norm": 1.2204087972640991,
|
| 1974 |
+
"learning_rate": 1.8102964478475316e-05,
|
| 1975 |
+
"loss": 0.9385,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.6555090655509066,
|
| 1980 |
+
"grad_norm": 1.1549757719039917,
|
| 1981 |
+
"learning_rate": 1.7885939416195687e-05,
|
| 1982 |
+
"loss": 0.7963,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.6578335657833566,
|
| 1987 |
+
"grad_norm": 1.1434992551803589,
|
| 1988 |
+
"learning_rate": 1.766972812909988e-05,
|
| 1989 |
+
"loss": 0.7884,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.6601580660158066,
|
| 1994 |
+
"grad_norm": 1.151331901550293,
|
| 1995 |
+
"learning_rate": 1.7454342656783464e-05,
|
| 1996 |
+
"loss": 0.767,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.6624825662482566,
|
| 2001 |
+
"grad_norm": 1.1396957635879517,
|
| 2002 |
+
"learning_rate": 1.7239794992857097e-05,
|
| 2003 |
+
"loss": 0.8191,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.6648070664807066,
|
| 2008 |
+
"grad_norm": 1.1968632936477661,
|
| 2009 |
+
"learning_rate": 1.7026097084278556e-05,
|
| 2010 |
+
"loss": 0.8241,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.6671315667131567,
|
| 2015 |
+
"grad_norm": 1.1648362874984741,
|
| 2016 |
+
"learning_rate": 1.6813260830687472e-05,
|
| 2017 |
+
"loss": 0.7965,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.6694560669456067,
|
| 2022 |
+
"grad_norm": 1.226285457611084,
|
| 2023 |
+
"learning_rate": 1.660129808374279e-05,
|
| 2024 |
+
"loss": 0.9029,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.6717805671780567,
|
| 2029 |
+
"grad_norm": 1.178755760192871,
|
| 2030 |
+
"learning_rate": 1.6390220646462693e-05,
|
| 2031 |
+
"loss": 0.7986,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.6741050674105067,
|
| 2036 |
+
"grad_norm": 1.1674259901046753,
|
| 2037 |
+
"learning_rate": 1.618004027256749e-05,
|
| 2038 |
+
"loss": 0.7532,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.6764295676429568,
|
| 2043 |
+
"grad_norm": 1.196860671043396,
|
| 2044 |
+
"learning_rate": 1.5970768665825036e-05,
|
| 2045 |
+
"loss": 0.8043,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.6787540678754068,
|
| 2050 |
+
"grad_norm": 1.2226951122283936,
|
| 2051 |
+
"learning_rate": 1.5762417479399005e-05,
|
| 2052 |
+
"loss": 0.847,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.6810785681078568,
|
| 2057 |
+
"grad_norm": 1.1900945901870728,
|
| 2058 |
+
"learning_rate": 1.555499831520006e-05,
|
| 2059 |
+
"loss": 0.847,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.6834030683403068,
|
| 2064 |
+
"grad_norm": 1.176953911781311,
|
| 2065 |
+
"learning_rate": 1.5348522723239726e-05,
|
| 2066 |
+
"loss": 0.8145,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.6857275685727569,
|
| 2071 |
+
"grad_norm": 1.1760255098342896,
|
| 2072 |
+
"learning_rate": 1.5143002200987304e-05,
|
| 2073 |
+
"loss": 0.792,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.6880520688052069,
|
| 2078 |
+
"grad_norm": 1.1875512599945068,
|
| 2079 |
+
"learning_rate": 1.4938448192729617e-05,
|
| 2080 |
+
"loss": 0.8698,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.6903765690376569,
|
| 2085 |
+
"grad_norm": 1.142947793006897,
|
| 2086 |
+
"learning_rate": 1.4734872088933713e-05,
|
| 2087 |
+
"loss": 0.8222,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.6927010692701069,
|
| 2092 |
+
"grad_norm": 1.2026643753051758,
|
| 2093 |
+
"learning_rate": 1.4532285225612613e-05,
|
| 2094 |
+
"loss": 0.8494,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.695025569502557,
|
| 2099 |
+
"grad_norm": 1.1509863138198853,
|
| 2100 |
+
"learning_rate": 1.4330698883694115e-05,
|
| 2101 |
+
"loss": 0.8091,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.697350069735007,
|
| 2106 |
+
"grad_norm": 1.237417459487915,
|
| 2107 |
+
"learning_rate": 1.413012428839254e-05,
|
| 2108 |
+
"loss": 0.8465,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.699674569967457,
|
| 2113 |
+
"grad_norm": 1.1145235300064087,
|
| 2114 |
+
"learning_rate": 1.3930572608583745e-05,
|
| 2115 |
+
"loss": 0.7913,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.701999070199907,
|
| 2120 |
+
"grad_norm": 1.1738826036453247,
|
| 2121 |
+
"learning_rate": 1.3732054956183111e-05,
|
| 2122 |
+
"loss": 0.8164,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.704323570432357,
|
| 2127 |
+
"grad_norm": 1.1653589010238647,
|
| 2128 |
+
"learning_rate": 1.3534582385526876e-05,
|
| 2129 |
+
"loss": 0.7946,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.7066480706648071,
|
| 2134 |
+
"grad_norm": 1.1790781021118164,
|
| 2135 |
+
"learning_rate": 1.3338165892756468e-05,
|
| 2136 |
+
"loss": 0.8287,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.7089725708972571,
|
| 2141 |
+
"grad_norm": 1.1557509899139404,
|
| 2142 |
+
"learning_rate": 1.3142816415206304e-05,
|
| 2143 |
+
"loss": 0.7584,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.7112970711297071,
|
| 2148 |
+
"grad_norm": 1.2092468738555908,
|
| 2149 |
+
"learning_rate": 1.2948544830794682e-05,
|
| 2150 |
+
"loss": 0.8641,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7136215713621571,
|
| 2155 |
+
"grad_norm": 1.1953829526901245,
|
| 2156 |
+
"learning_rate": 1.275536195741805e-05,
|
| 2157 |
+
"loss": 0.8669,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.7159460715946072,
|
| 2162 |
+
"grad_norm": 1.1542140245437622,
|
| 2163 |
+
"learning_rate": 1.2563278552348676e-05,
|
| 2164 |
+
"loss": 0.8841,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.7182705718270572,
|
| 2169 |
+
"grad_norm": 1.2032291889190674,
|
| 2170 |
+
"learning_rate": 1.2372305311635541e-05,
|
| 2171 |
+
"loss": 0.8485,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.7205950720595072,
|
| 2176 |
+
"grad_norm": 1.1283597946166992,
|
| 2177 |
+
"learning_rate": 1.2182452869508834e-05,
|
| 2178 |
+
"loss": 0.7937,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.7229195722919572,
|
| 2183 |
+
"grad_norm": 1.133653163909912,
|
| 2184 |
+
"learning_rate": 1.1993731797787766e-05,
|
| 2185 |
+
"loss": 0.81,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.7252440725244073,
|
| 2190 |
+
"grad_norm": 1.1899718046188354,
|
| 2191 |
+
"learning_rate": 1.1806152605291783e-05,
|
| 2192 |
+
"loss": 0.8785,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.7275685727568573,
|
| 2197 |
+
"grad_norm": 1.182756781578064,
|
| 2198 |
+
"learning_rate": 1.1619725737255553e-05,
|
| 2199 |
+
"loss": 0.8157,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.7298930729893073,
|
| 2204 |
+
"grad_norm": 1.1462957859039307,
|
| 2205 |
+
"learning_rate": 1.143446157474723e-05,
|
| 2206 |
+
"loss": 0.8032,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.7322175732217573,
|
| 2211 |
+
"grad_norm": 1.1228857040405273,
|
| 2212 |
+
"learning_rate": 1.1250370434090382e-05,
|
| 2213 |
+
"loss": 0.7521,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.7345420734542073,
|
| 2218 |
+
"grad_norm": 1.1679176092147827,
|
| 2219 |
+
"learning_rate": 1.1067462566289592e-05,
|
| 2220 |
+
"loss": 0.8729,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.7368665736866574,
|
| 2225 |
+
"grad_norm": 1.1772230863571167,
|
| 2226 |
+
"learning_rate": 1.0885748156459578e-05,
|
| 2227 |
+
"loss": 0.8305,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.7391910739191074,
|
| 2232 |
+
"grad_norm": 1.1800023317337036,
|
| 2233 |
+
"learning_rate": 1.0705237323258088e-05,
|
| 2234 |
+
"loss": 0.8285,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.7415155741515574,
|
| 2239 |
+
"grad_norm": 1.1623144149780273,
|
| 2240 |
+
"learning_rate": 1.0525940118322434e-05,
|
| 2241 |
+
"loss": 0.8343,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.7438400743840075,
|
| 2246 |
+
"grad_norm": 1.2074713706970215,
|
| 2247 |
+
"learning_rate": 1.0347866525709737e-05,
|
| 2248 |
+
"loss": 0.8442,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.7461645746164575,
|
| 2253 |
+
"grad_norm": 1.1325371265411377,
|
| 2254 |
+
"learning_rate": 1.017102646134103e-05,
|
| 2255 |
+
"loss": 0.8072,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.7484890748489075,
|
| 2260 |
+
"grad_norm": 1.176443099975586,
|
| 2261 |
+
"learning_rate": 9.995429772449044e-06,
|
| 2262 |
+
"loss": 0.8585,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.7508135750813575,
|
| 2267 |
+
"grad_norm": 1.159180760383606,
|
| 2268 |
+
"learning_rate": 9.82108623702991e-06,
|
| 2269 |
+
"loss": 0.8226,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.7531380753138075,
|
| 2274 |
+
"grad_norm": 1.180493950843811,
|
| 2275 |
+
"learning_rate": 9.648005563298672e-06,
|
| 2276 |
+
"loss": 0.8364,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.7554625755462575,
|
| 2281 |
+
"grad_norm": 1.2017948627471924,
|
| 2282 |
+
"learning_rate": 9.476197389148645e-06,
|
| 2283 |
+
"loss": 0.8531,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.7577870757787076,
|
| 2288 |
+
"grad_norm": 1.156783103942871,
|
| 2289 |
+
"learning_rate": 9.305671281614786e-06,
|
| 2290 |
+
"loss": 0.8105,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.7601115760111576,
|
| 2295 |
+
"grad_norm": 1.145729660987854,
|
| 2296 |
+
"learning_rate": 9.136436736340962e-06,
|
| 2297 |
+
"loss": 0.8248,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.7624360762436077,
|
| 2302 |
+
"grad_norm": 1.2108309268951416,
|
| 2303 |
+
"learning_rate": 8.968503177051128e-06,
|
| 2304 |
+
"loss": 0.8042,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.7647605764760577,
|
| 2309 |
+
"grad_norm": 1.193647861480713,
|
| 2310 |
+
"learning_rate": 8.801879955024671e-06,
|
| 2311 |
+
"loss": 0.8598,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.7670850767085077,
|
| 2316 |
+
"grad_norm": 1.2257269620895386,
|
| 2317 |
+
"learning_rate": 8.636576348575587e-06,
|
| 2318 |
+
"loss": 0.8747,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.7694095769409577,
|
| 2323 |
+
"grad_norm": 1.2016220092773438,
|
| 2324 |
+
"learning_rate": 8.472601562535888e-06,
|
| 2325 |
+
"loss": 0.799,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.7717340771734077,
|
| 2330 |
+
"grad_norm": 1.243636131286621,
|
| 2331 |
+
"learning_rate": 8.309964727743042e-06,
|
| 2332 |
+
"loss": 0.8876,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.7740585774058577,
|
| 2337 |
+
"grad_norm": 1.2658741474151611,
|
| 2338 |
+
"learning_rate": 8.148674900531466e-06,
|
| 2339 |
+
"loss": 0.8943,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.7763830776383077,
|
| 2344 |
+
"grad_norm": 1.1937084197998047,
|
| 2345 |
+
"learning_rate": 7.988741062228303e-06,
|
| 2346 |
+
"loss": 0.8072,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.7787075778707578,
|
| 2351 |
+
"grad_norm": 1.188068151473999,
|
| 2352 |
+
"learning_rate": 7.830172118653236e-06,
|
| 2353 |
+
"loss": 0.8405,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.7810320781032078,
|
| 2358 |
+
"grad_norm": 1.1817395687103271,
|
| 2359 |
+
"learning_rate": 7.672976899622636e-06,
|
| 2360 |
+
"loss": 0.8474,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.7833565783356579,
|
| 2365 |
+
"grad_norm": 1.2230587005615234,
|
| 2366 |
+
"learning_rate": 7.517164158457843e-06,
|
| 2367 |
+
"loss": 0.8168,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.7856810785681079,
|
| 2372 |
+
"grad_norm": 1.1503418684005737,
|
| 2373 |
+
"learning_rate": 7.362742571497723e-06,
|
| 2374 |
+
"loss": 0.7769,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.7880055788005579,
|
| 2379 |
+
"grad_norm": 1.1554263830184937,
|
| 2380 |
+
"learning_rate": 7.209720737615588e-06,
|
| 2381 |
+
"loss": 0.7941,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.7903300790330079,
|
| 2386 |
+
"grad_norm": 1.167829155921936,
|
| 2387 |
+
"learning_rate": 7.058107177740316e-06,
|
| 2388 |
+
"loss": 0.861,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.7926545792654579,
|
| 2393 |
+
"grad_norm": 1.1501438617706299,
|
| 2394 |
+
"learning_rate": 6.907910334381895e-06,
|
| 2395 |
+
"loss": 0.7728,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.7949790794979079,
|
| 2400 |
+
"grad_norm": 1.2136327028274536,
|
| 2401 |
+
"learning_rate": 6.759138571161328e-06,
|
| 2402 |
+
"loss": 0.8745,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.7973035797303579,
|
| 2407 |
+
"grad_norm": 1.15041184425354,
|
| 2408 |
+
"learning_rate": 6.611800172344858e-06,
|
| 2409 |
+
"loss": 0.8245,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.799628079962808,
|
| 2414 |
+
"grad_norm": 1.2242484092712402,
|
| 2415 |
+
"learning_rate": 6.4659033423827105e-06,
|
| 2416 |
+
"loss": 0.8703,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.8019525801952581,
|
| 2421 |
+
"grad_norm": 1.2068572044372559,
|
| 2422 |
+
"learning_rate": 6.321456205452219e-06,
|
| 2423 |
+
"loss": 0.8453,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.8042770804277081,
|
| 2428 |
+
"grad_norm": 1.138300895690918,
|
| 2429 |
+
"learning_rate": 6.17846680500541e-06,
|
| 2430 |
+
"loss": 0.8205,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.8066015806601581,
|
| 2435 |
+
"grad_norm": 1.2071059942245483,
|
| 2436 |
+
"learning_rate": 6.0369431033211525e-06,
|
| 2437 |
+
"loss": 0.873,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.8089260808926081,
|
| 2442 |
+
"grad_norm": 1.2351478338241577,
|
| 2443 |
+
"learning_rate": 5.896892981061738e-06,
|
| 2444 |
+
"loss": 0.8453,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.8112505811250581,
|
| 2449 |
+
"grad_norm": 1.1644549369812012,
|
| 2450 |
+
"learning_rate": 5.758324236834091e-06,
|
| 2451 |
+
"loss": 0.8112,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.8135750813575081,
|
| 2456 |
+
"grad_norm": 1.2177077531814575,
|
| 2457 |
+
"learning_rate": 5.621244586755496e-06,
|
| 2458 |
+
"loss": 0.8216,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.8158995815899581,
|
| 2463 |
+
"grad_norm": 1.202233076095581,
|
| 2464 |
+
"learning_rate": 5.485661664023893e-06,
|
| 2465 |
+
"loss": 0.8821,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.8182240818224081,
|
| 2470 |
+
"grad_norm": 1.173756718635559,
|
| 2471 |
+
"learning_rate": 5.351583018492902e-06,
|
| 2472 |
+
"loss": 0.8337,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.8205485820548583,
|
| 2477 |
+
"grad_norm": 1.1963239908218384,
|
| 2478 |
+
"learning_rate": 5.219016116251332e-06,
|
| 2479 |
+
"loss": 0.8751,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.8228730822873083,
|
| 2484 |
+
"grad_norm": 1.1608357429504395,
|
| 2485 |
+
"learning_rate": 5.087968339207486e-06,
|
| 2486 |
+
"loss": 0.8938,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.8251975825197583,
|
| 2491 |
+
"grad_norm": 1.1277998685836792,
|
| 2492 |
+
"learning_rate": 4.958446984678114e-06,
|
| 2493 |
+
"loss": 0.7774,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.8275220827522083,
|
| 2498 |
+
"grad_norm": 1.2397881746292114,
|
| 2499 |
+
"learning_rate": 4.830459264982011e-06,
|
| 2500 |
+
"loss": 0.8269,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.8298465829846583,
|
| 2505 |
+
"grad_norm": 1.2230621576309204,
|
| 2506 |
+
"learning_rate": 4.704012307038478e-06,
|
| 2507 |
+
"loss": 0.8511,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.8321710832171083,
|
| 2512 |
+
"grad_norm": 1.1905261278152466,
|
| 2513 |
+
"learning_rate": 4.579113151970381e-06,
|
| 2514 |
+
"loss": 0.8424,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.8344955834495583,
|
| 2519 |
+
"grad_norm": 1.2389678955078125,
|
| 2520 |
+
"learning_rate": 4.455768754712141e-06,
|
| 2521 |
+
"loss": 0.8978,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.8368200836820083,
|
| 2526 |
+
"grad_norm": 1.2028136253356934,
|
| 2527 |
+
"learning_rate": 4.3339859836224235e-06,
|
| 2528 |
+
"loss": 0.8501,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.8391445839144583,
|
| 2533 |
+
"grad_norm": 1.1460210084915161,
|
| 2534 |
+
"learning_rate": 4.213771620101661e-06,
|
| 2535 |
+
"loss": 0.7621,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.8414690841469085,
|
| 2540 |
+
"grad_norm": 1.1692181825637817,
|
| 2541 |
+
"learning_rate": 4.095132358214476e-06,
|
| 2542 |
+
"loss": 0.8351,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.8437935843793585,
|
| 2547 |
+
"grad_norm": 1.1694318056106567,
|
| 2548 |
+
"learning_rate": 3.97807480431688e-06,
|
| 2549 |
+
"loss": 0.7781,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.8461180846118085,
|
| 2554 |
+
"grad_norm": 1.1537806987762451,
|
| 2555 |
+
"learning_rate": 3.862605476688435e-06,
|
| 2556 |
+
"loss": 0.7383,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.8484425848442585,
|
| 2561 |
+
"grad_norm": 1.1844308376312256,
|
| 2562 |
+
"learning_rate": 3.748730805169292e-06,
|
| 2563 |
+
"loss": 0.8434,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.8507670850767085,
|
| 2568 |
+
"grad_norm": 1.2213958501815796,
|
| 2569 |
+
"learning_rate": 3.6364571308021075e-06,
|
| 2570 |
+
"loss": 0.9058,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.8530915853091585,
|
| 2575 |
+
"grad_norm": 1.2063052654266357,
|
| 2576 |
+
"learning_rate": 3.5257907054789965e-06,
|
| 2577 |
+
"loss": 0.864,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.8554160855416085,
|
| 2582 |
+
"grad_norm": 1.2098087072372437,
|
| 2583 |
+
"learning_rate": 3.416737691593364e-06,
|
| 2584 |
+
"loss": 0.8397,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.8577405857740585,
|
| 2589 |
+
"grad_norm": 1.197662591934204,
|
| 2590 |
+
"learning_rate": 3.309304161696762e-06,
|
| 2591 |
+
"loss": 0.8456,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.8600650860065086,
|
| 2596 |
+
"grad_norm": 1.167975902557373,
|
| 2597 |
+
"learning_rate": 3.203496098160773e-06,
|
| 2598 |
+
"loss": 0.8325,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.8623895862389587,
|
| 2603 |
+
"grad_norm": 1.1775121688842773,
|
| 2604 |
+
"learning_rate": 3.0993193928438364e-06,
|
| 2605 |
+
"loss": 0.8519,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.8647140864714087,
|
| 2610 |
+
"grad_norm": 1.1968019008636475,
|
| 2611 |
+
"learning_rate": 2.996779846763207e-06,
|
| 2612 |
+
"loss": 0.8299,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.8670385867038587,
|
| 2617 |
+
"grad_norm": 1.1625730991363525,
|
| 2618 |
+
"learning_rate": 2.8958831697719116e-06,
|
| 2619 |
+
"loss": 0.7785,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.8693630869363087,
|
| 2624 |
+
"grad_norm": 1.2199307680130005,
|
| 2625 |
+
"learning_rate": 2.796634980240773e-06,
|
| 2626 |
+
"loss": 0.8107,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.8716875871687587,
|
| 2631 |
+
"grad_norm": 1.2188974618911743,
|
| 2632 |
+
"learning_rate": 2.6990408047456208e-06,
|
| 2633 |
+
"loss": 0.8457,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.8740120874012087,
|
| 2638 |
+
"grad_norm": 1.194596767425537,
|
| 2639 |
+
"learning_rate": 2.603106077759467e-06,
|
| 2640 |
+
"loss": 0.8295,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.8763365876336587,
|
| 2645 |
+
"grad_norm": 1.0834101438522339,
|
| 2646 |
+
"learning_rate": 2.508836141349955e-06,
|
| 2647 |
+
"loss": 0.7247,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.8786610878661087,
|
| 2652 |
+
"grad_norm": 1.1972309350967407,
|
| 2653 |
+
"learning_rate": 2.4162362448818634e-06,
|
| 2654 |
+
"loss": 0.9107,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.8809855880985588,
|
| 2659 |
+
"grad_norm": 1.1724853515625,
|
| 2660 |
+
"learning_rate": 2.3253115447247866e-06,
|
| 2661 |
+
"loss": 0.8098,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.8833100883310089,
|
| 2666 |
+
"grad_norm": 1.1860358715057373,
|
| 2667 |
+
"learning_rate": 2.2360671039660346e-06,
|
| 2668 |
+
"loss": 0.8301,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.8856345885634589,
|
| 2673 |
+
"grad_norm": 1.2200249433517456,
|
| 2674 |
+
"learning_rate": 2.148507892128674e-06,
|
| 2675 |
+
"loss": 0.8012,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.8879590887959089,
|
| 2680 |
+
"grad_norm": 1.118643045425415,
|
| 2681 |
+
"learning_rate": 2.062638784894803e-06,
|
| 2682 |
+
"loss": 0.7925,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.8902835890283589,
|
| 2687 |
+
"grad_norm": 1.1284154653549194,
|
| 2688 |
+
"learning_rate": 1.9784645638340827e-06,
|
| 2689 |
+
"loss": 0.825,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.8926080892608089,
|
| 2694 |
+
"grad_norm": 1.2449028491973877,
|
| 2695 |
+
"learning_rate": 1.8959899161374317e-06,
|
| 2696 |
+
"loss": 0.8187,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.8949325894932589,
|
| 2701 |
+
"grad_norm": 1.1662657260894775,
|
| 2702 |
+
"learning_rate": 1.8152194343560626e-06,
|
| 2703 |
+
"loss": 0.8131,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.897257089725709,
|
| 2708 |
+
"grad_norm": 1.2282719612121582,
|
| 2709 |
+
"learning_rate": 1.736157616145729e-06,
|
| 2710 |
+
"loss": 0.8428,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.899581589958159,
|
| 2715 |
+
"grad_norm": 1.112457275390625,
|
| 2716 |
+
"learning_rate": 1.6588088640162775e-06,
|
| 2717 |
+
"loss": 0.767,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.901906090190609,
|
| 2722 |
+
"grad_norm": 1.1861826181411743,
|
| 2723 |
+
"learning_rate": 1.5831774850865059e-06,
|
| 2724 |
+
"loss": 0.8122,
|
| 2725 |
+
"step": 388
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.9042305904230591,
|
| 2729 |
+
"grad_norm": 1.1628504991531372,
|
| 2730 |
+
"learning_rate": 1.5092676908443047e-06,
|
| 2731 |
+
"loss": 0.7754,
|
| 2732 |
+
"step": 389
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.9065550906555091,
|
| 2736 |
+
"grad_norm": 1.1673952341079712,
|
| 2737 |
+
"learning_rate": 1.4370835969121649e-06,
|
| 2738 |
+
"loss": 0.8309,
|
| 2739 |
+
"step": 390
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.9088795908879591,
|
| 2743 |
+
"grad_norm": 1.203006386756897,
|
| 2744 |
+
"learning_rate": 1.366629222817995e-06,
|
| 2745 |
+
"loss": 0.7955,
|
| 2746 |
+
"step": 391
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.9112040911204091,
|
| 2750 |
+
"grad_norm": 1.1598377227783203,
|
| 2751 |
+
"learning_rate": 1.2979084917712842e-06,
|
| 2752 |
+
"loss": 0.8337,
|
| 2753 |
+
"step": 392
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.9135285913528591,
|
| 2757 |
+
"grad_norm": 1.1832410097122192,
|
| 2758 |
+
"learning_rate": 1.2309252304446607e-06,
|
| 2759 |
+
"loss": 0.8024,
|
| 2760 |
+
"step": 393
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.9158530915853091,
|
| 2764 |
+
"grad_norm": 1.1913425922393799,
|
| 2765 |
+
"learning_rate": 1.1656831687607826e-06,
|
| 2766 |
+
"loss": 0.8127,
|
| 2767 |
+
"step": 394
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.9181775918177592,
|
| 2771 |
+
"grad_norm": 1.1691555976867676,
|
| 2772 |
+
"learning_rate": 1.1021859396846632e-06,
|
| 2773 |
+
"loss": 0.7681,
|
| 2774 |
+
"step": 395
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.9205020920502092,
|
| 2778 |
+
"grad_norm": 1.2271596193313599,
|
| 2779 |
+
"learning_rate": 1.0404370790213507e-06,
|
| 2780 |
+
"loss": 0.8653,
|
| 2781 |
+
"step": 396
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.9228265922826592,
|
| 2785 |
+
"grad_norm": 1.1940650939941406,
|
| 2786 |
+
"learning_rate": 9.804400252190588e-07,
|
| 2787 |
+
"loss": 0.8275,
|
| 2788 |
+
"step": 397
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.9251510925151093,
|
| 2792 |
+
"grad_norm": 1.322935938835144,
|
| 2793 |
+
"learning_rate": 9.221981191776801e-07,
|
| 2794 |
+
"loss": 0.9482,
|
| 2795 |
+
"step": 398
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.9274755927475593,
|
| 2799 |
+
"grad_norm": 1.226189374923706,
|
| 2800 |
+
"learning_rate": 8.657146040627655e-07,
|
| 2801 |
+
"loss": 0.8803,
|
| 2802 |
+
"step": 399
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.9298000929800093,
|
| 2806 |
+
"grad_norm": 1.2069227695465088,
|
| 2807 |
+
"learning_rate": 8.109926251249163e-07,
|
| 2808 |
+
"loss": 0.8547,
|
| 2809 |
+
"step": 400
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.9321245932124593,
|
| 2813 |
+
"grad_norm": 1.2206929922103882,
|
| 2814 |
+
"learning_rate": 7.58035229524658e-07,
|
| 2815 |
+
"loss": 0.8214,
|
| 2816 |
+
"step": 401
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.9344490934449093,
|
| 2820 |
+
"grad_norm": 1.2493535280227661,
|
| 2821 |
+
"learning_rate": 7.068453661627424e-07,
|
| 2822 |
+
"loss": 0.8802,
|
| 2823 |
+
"step": 402
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.9367735936773594,
|
| 2827 |
+
"grad_norm": 1.1659430265426636,
|
| 2828 |
+
"learning_rate": 6.574258855159656e-07,
|
| 2829 |
+
"loss": 0.803,
|
| 2830 |
+
"step": 403
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.9390980939098094,
|
| 2834 |
+
"grad_norm": 1.2101560831069946,
|
| 2835 |
+
"learning_rate": 6.097795394784138e-07,
|
| 2836 |
+
"loss": 0.8255,
|
| 2837 |
+
"step": 404
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.9414225941422594,
|
| 2841 |
+
"grad_norm": 1.1926182508468628,
|
| 2842 |
+
"learning_rate": 5.639089812082365e-07,
|
| 2843 |
+
"loss": 0.7878,
|
| 2844 |
+
"step": 405
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.9437470943747094,
|
| 2848 |
+
"grad_norm": 1.1893055438995361,
|
| 2849 |
+
"learning_rate": 5.198167649799214e-07,
|
| 2850 |
+
"loss": 0.8213,
|
| 2851 |
+
"step": 406
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.9460715946071595,
|
| 2855 |
+
"grad_norm": 1.2222219705581665,
|
| 2856 |
+
"learning_rate": 4.775053460420305e-07,
|
| 2857 |
+
"loss": 0.8765,
|
| 2858 |
+
"step": 407
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.9483960948396095,
|
| 2862 |
+
"grad_norm": 1.1839882135391235,
|
| 2863 |
+
"learning_rate": 4.369770804805123e-07,
|
| 2864 |
+
"loss": 0.7754,
|
| 2865 |
+
"step": 408
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.9507205950720595,
|
| 2869 |
+
"grad_norm": 1.1785272359848022,
|
| 2870 |
+
"learning_rate": 3.98234225087489e-07,
|
| 2871 |
+
"loss": 0.8291,
|
| 2872 |
+
"step": 409
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.9530450953045095,
|
| 2876 |
+
"grad_norm": 1.1746994256973267,
|
| 2877 |
+
"learning_rate": 3.6127893723558233e-07,
|
| 2878 |
+
"loss": 0.8198,
|
| 2879 |
+
"step": 410
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.9553695955369595,
|
| 2883 |
+
"grad_norm": 1.218136191368103,
|
| 2884 |
+
"learning_rate": 3.2611327475779545e-07,
|
| 2885 |
+
"loss": 0.8316,
|
| 2886 |
+
"step": 411
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.9576940957694096,
|
| 2890 |
+
"grad_norm": 1.2541922330856323,
|
| 2891 |
+
"learning_rate": 2.927391958329195e-07,
|
| 2892 |
+
"loss": 0.8747,
|
| 2893 |
+
"step": 412
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.9600185960018596,
|
| 2897 |
+
"grad_norm": 1.1385949850082397,
|
| 2898 |
+
"learning_rate": 2.6115855887648195e-07,
|
| 2899 |
+
"loss": 0.7893,
|
| 2900 |
+
"step": 413
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.9623430962343096,
|
| 2904 |
+
"grad_norm": 1.142892837524414,
|
| 2905 |
+
"learning_rate": 2.3137312243728834e-07,
|
| 2906 |
+
"loss": 0.7838,
|
| 2907 |
+
"step": 414
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.9646675964667596,
|
| 2911 |
+
"grad_norm": 1.2635773420333862,
|
| 2912 |
+
"learning_rate": 2.033845450994659e-07,
|
| 2913 |
+
"loss": 0.8358,
|
| 2914 |
+
"step": 415
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.9669920966992097,
|
| 2918 |
+
"grad_norm": 1.141421914100647,
|
| 2919 |
+
"learning_rate": 1.771943853901292e-07,
|
| 2920 |
+
"loss": 0.8355,
|
| 2921 |
+
"step": 416
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.9693165969316597,
|
| 2925 |
+
"grad_norm": 1.1934841871261597,
|
| 2926 |
+
"learning_rate": 1.5280410169258794e-07,
|
| 2927 |
+
"loss": 0.8171,
|
| 2928 |
+
"step": 417
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.9716410971641097,
|
| 2932 |
+
"grad_norm": 1.1529548168182373,
|
| 2933 |
+
"learning_rate": 1.3021505216514017e-07,
|
| 2934 |
+
"loss": 0.7629,
|
| 2935 |
+
"step": 418
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.9739655973965597,
|
| 2939 |
+
"grad_norm": 1.207676649093628,
|
| 2940 |
+
"learning_rate": 1.0942849466543308e-07,
|
| 2941 |
+
"loss": 0.8672,
|
| 2942 |
+
"step": 419
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.9762900976290098,
|
| 2946 |
+
"grad_norm": 1.264219045639038,
|
| 2947 |
+
"learning_rate": 9.044558668043461e-08,
|
| 2948 |
+
"loss": 0.9103,
|
| 2949 |
+
"step": 420
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.9786145978614598,
|
| 2953 |
+
"grad_norm": 1.2143738269805908,
|
| 2954 |
+
"learning_rate": 7.32673852619689e-08,
|
| 2955 |
+
"loss": 0.8604,
|
| 2956 |
+
"step": 421
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.9809390980939098,
|
| 2960 |
+
"grad_norm": 1.172698974609375,
|
| 2961 |
+
"learning_rate": 5.7894846967866115e-08,
|
| 2962 |
+
"loss": 0.823,
|
| 2963 |
+
"step": 422
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.9832635983263598,
|
| 2967 |
+
"grad_norm": 1.194075584411621,
|
| 2968 |
+
"learning_rate": 4.432882780868341e-08,
|
| 2969 |
+
"loss": 0.8139,
|
| 2970 |
+
"step": 423
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.9855880985588099,
|
| 2974 |
+
"grad_norm": 1.191976547241211,
|
| 2975 |
+
"learning_rate": 3.257008320005111e-08,
|
| 2976 |
+
"loss": 0.8117,
|
| 2977 |
+
"step": 424
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.9879125987912599,
|
| 2981 |
+
"grad_norm": 1.202752709388733,
|
| 2982 |
+
"learning_rate": 2.261926792058999e-08,
|
| 2983 |
+
"loss": 0.8479,
|
| 2984 |
+
"step": 425
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.9902370990237099,
|
| 2988 |
+
"grad_norm": 1.2659246921539307,
|
| 2989 |
+
"learning_rate": 1.4476936075475397e-08,
|
| 2990 |
+
"loss": 0.8472,
|
| 2991 |
+
"step": 426
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.9925615992561599,
|
| 2995 |
+
"grad_norm": 1.1615605354309082,
|
| 2996 |
+
"learning_rate": 8.143541065568914e-09,
|
| 2997 |
+
"loss": 0.778,
|
| 2998 |
+
"step": 427
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.99488609948861,
|
| 3002 |
+
"grad_norm": 1.23662531375885,
|
| 3003 |
+
"learning_rate": 3.619435562157158e-09,
|
| 3004 |
+
"loss": 0.8868,
|
| 3005 |
+
"step": 428
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.99721059972106,
|
| 3009 |
+
"grad_norm": 1.181993007659912,
|
| 3010 |
+
"learning_rate": 9.048714873446938e-10,
|
| 3011 |
+
"loss": 0.8409,
|
| 3012 |
+
"step": 429
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.99953509995351,
|
| 3016 |
+
"grad_norm": 1.1682720184326172,
|
| 3017 |
+
"learning_rate": 0.0,
|
| 3018 |
+
"loss": 0.8589,
|
| 3019 |
+
"step": 430
|
| 3020 |
+
}
|
| 3021 |
+
],
|
| 3022 |
+
"logging_steps": 1,
|
| 3023 |
+
"max_steps": 430,
|
| 3024 |
+
"num_input_tokens_seen": 0,
|
| 3025 |
+
"num_train_epochs": 1,
|
| 3026 |
+
"save_steps": 500,
|
| 3027 |
+
"stateful_callbacks": {
|
| 3028 |
+
"TrainerControl": {
|
| 3029 |
+
"args": {
|
| 3030 |
+
"should_epoch_stop": false,
|
| 3031 |
+
"should_evaluate": false,
|
| 3032 |
+
"should_log": false,
|
| 3033 |
+
"should_save": true,
|
| 3034 |
+
"should_training_stop": true
|
| 3035 |
+
},
|
| 3036 |
+
"attributes": {}
|
| 3037 |
+
}
|
| 3038 |
+
},
|
| 3039 |
+
"total_flos": 1.6061243889342874e+17,
|
| 3040 |
+
"train_batch_size": 6,
|
| 3041 |
+
"trial_name": null,
|
| 3042 |
+
"trial_params": null
|
| 3043 |
+
}
|
checkpoint-430/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4debe46d465ccb686ed9be1290948389157ed41684e17b1236c73d822a8644c6
|
| 3 |
+
size 5496
|