Upload folder using huggingface_hub
Browse files- checkpoint-450/README.md +202 -0
- checkpoint-450/adapter_config.json +37 -0
- checkpoint-450/adapter_model.safetensors +3 -0
- checkpoint-450/optimizer.pt +3 -0
- checkpoint-450/rng_state.pth +3 -0
- checkpoint-450/scheduler.pt +3 -0
- checkpoint-450/trainer_state.json +3183 -0
- checkpoint-450/training_args.bin +3 -0
checkpoint-450/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: unsloth/gemma-2-27b-bnb-4bit
|
| 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 0.11.1
|
checkpoint-450/adapter_config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "unsloth/gemma-2-27b-bnb-4bit",
|
| 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": "unsloth",
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"up_proj",
|
| 27 |
+
"k_proj",
|
| 28 |
+
"gate_proj",
|
| 29 |
+
"down_proj",
|
| 30 |
+
"v_proj",
|
| 31 |
+
"q_proj",
|
| 32 |
+
"o_proj"
|
| 33 |
+
],
|
| 34 |
+
"task_type": "CAUSAL_LM",
|
| 35 |
+
"use_dora": false,
|
| 36 |
+
"use_rslora": true
|
| 37 |
+
}
|
checkpoint-450/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:40d35e93a4b2e62b10e52da33e46076826a2c5da4458fee7bf2dadc664943da3
|
| 3 |
+
size 27247459432
|
checkpoint-450/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82e4b056f6fa0b39383ebf9fc7c3ce0288ca0ef7cf30583e4778e1d937bb5f93
|
| 3 |
+
size 6559156728
|
checkpoint-450/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eb8e13a761d59b362762d6ee820fea31ce0b95fd5dc88a4c162745b069a81dba
|
| 3 |
+
size 14244
|
checkpoint-450/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:19dd60884b3934527554a6ea0f306ef52c6e394e8bba106a3a35fe806204c2d7
|
| 3 |
+
size 1064
|
checkpoint-450/trainer_state.json
ADDED
|
@@ -0,0 +1,3183 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9054325955734407,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 450,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.002012072434607646,
|
| 13 |
+
"grad_norm": 46.70955276489258,
|
| 14 |
+
"learning_rate": 6.5e-06,
|
| 15 |
+
"loss": 1.4376,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004024144869215292,
|
| 20 |
+
"grad_norm": 25.710079193115234,
|
| 21 |
+
"learning_rate": 1.3e-05,
|
| 22 |
+
"loss": 1.2619,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006036217303822937,
|
| 27 |
+
"grad_norm": 26.04344940185547,
|
| 28 |
+
"learning_rate": 1.9499999999999996e-05,
|
| 29 |
+
"loss": 1.2208,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008048289738430584,
|
| 34 |
+
"grad_norm": 8.251016616821289,
|
| 35 |
+
"learning_rate": 2.6e-05,
|
| 36 |
+
"loss": 1.0171,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01006036217303823,
|
| 41 |
+
"grad_norm": 4.199310302734375,
|
| 42 |
+
"learning_rate": 3.25e-05,
|
| 43 |
+
"loss": 0.9699,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.012072434607645875,
|
| 48 |
+
"grad_norm": 6.134476661682129,
|
| 49 |
+
"learning_rate": 3.899999999999999e-05,
|
| 50 |
+
"loss": 0.9899,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.014084507042253521,
|
| 55 |
+
"grad_norm": 4.348274230957031,
|
| 56 |
+
"learning_rate": 4.5499999999999995e-05,
|
| 57 |
+
"loss": 0.9959,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.01609657947686117,
|
| 62 |
+
"grad_norm": 3.9475340843200684,
|
| 63 |
+
"learning_rate": 5.2e-05,
|
| 64 |
+
"loss": 0.8857,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.018108651911468814,
|
| 69 |
+
"grad_norm": 3.1918065547943115,
|
| 70 |
+
"learning_rate": 5.85e-05,
|
| 71 |
+
"loss": 0.9008,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02012072434607646,
|
| 76 |
+
"grad_norm": 2.7590606212615967,
|
| 77 |
+
"learning_rate": 6.5e-05,
|
| 78 |
+
"loss": 0.8703,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.022132796780684104,
|
| 83 |
+
"grad_norm": 3.0539116859436035,
|
| 84 |
+
"learning_rate": 6.499932377117017e-05,
|
| 85 |
+
"loss": 0.8258,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.02414486921529175,
|
| 90 |
+
"grad_norm": 2.821988344192505,
|
| 91 |
+
"learning_rate": 6.499729511282135e-05,
|
| 92 |
+
"loss": 0.9071,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.026156941649899398,
|
| 97 |
+
"grad_norm": 2.776759147644043,
|
| 98 |
+
"learning_rate": 6.499391410937425e-05,
|
| 99 |
+
"loss": 0.8496,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.028169014084507043,
|
| 104 |
+
"grad_norm": 2.897883176803589,
|
| 105 |
+
"learning_rate": 6.498918090152627e-05,
|
| 106 |
+
"loss": 0.8339,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.030181086519114688,
|
| 111 |
+
"grad_norm": 2.511650562286377,
|
| 112 |
+
"learning_rate": 6.498309568624549e-05,
|
| 113 |
+
"loss": 0.8398,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.03219315895372234,
|
| 118 |
+
"grad_norm": 2.69217848777771,
|
| 119 |
+
"learning_rate": 6.497565871676255e-05,
|
| 120 |
+
"loss": 0.8373,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.03420523138832998,
|
| 125 |
+
"grad_norm": 2.5202300548553467,
|
| 126 |
+
"learning_rate": 6.496687030256012e-05,
|
| 127 |
+
"loss": 0.8418,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.03621730382293763,
|
| 132 |
+
"grad_norm": 2.4313321113586426,
|
| 133 |
+
"learning_rate": 6.495673080935999e-05,
|
| 134 |
+
"loss": 0.8482,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.03822937625754527,
|
| 139 |
+
"grad_norm": 2.278684139251709,
|
| 140 |
+
"learning_rate": 6.494524065910784e-05,
|
| 141 |
+
"loss": 0.8059,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.04024144869215292,
|
| 146 |
+
"grad_norm": 2.2294907569885254,
|
| 147 |
+
"learning_rate": 6.493240032995573e-05,
|
| 148 |
+
"loss": 0.8085,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.04225352112676056,
|
| 153 |
+
"grad_norm": 2.446382999420166,
|
| 154 |
+
"learning_rate": 6.491821035624218e-05,
|
| 155 |
+
"loss": 0.8298,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.04426559356136821,
|
| 160 |
+
"grad_norm": 2.4150474071502686,
|
| 161 |
+
"learning_rate": 6.490267132846992e-05,
|
| 162 |
+
"loss": 0.9265,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.04627766599597585,
|
| 167 |
+
"grad_norm": 2.170037269592285,
|
| 168 |
+
"learning_rate": 6.488578389328129e-05,
|
| 169 |
+
"loss": 0.8323,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0482897384305835,
|
| 174 |
+
"grad_norm": 2.010956287384033,
|
| 175 |
+
"learning_rate": 6.486754875343145e-05,
|
| 176 |
+
"loss": 0.8646,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.05030181086519115,
|
| 181 |
+
"grad_norm": 1.8752851486206055,
|
| 182 |
+
"learning_rate": 6.484796666775895e-05,
|
| 183 |
+
"loss": 0.8402,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.052313883299798795,
|
| 188 |
+
"grad_norm": 1.8610150814056396,
|
| 189 |
+
"learning_rate": 6.482703845115434e-05,
|
| 190 |
+
"loss": 0.8256,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.05432595573440644,
|
| 195 |
+
"grad_norm": 1.8047019243240356,
|
| 196 |
+
"learning_rate": 6.480476497452613e-05,
|
| 197 |
+
"loss": 0.819,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.056338028169014086,
|
| 202 |
+
"grad_norm": 1.6434446573257446,
|
| 203 |
+
"learning_rate": 6.47811471647646e-05,
|
| 204 |
+
"loss": 0.8671,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.05835010060362173,
|
| 209 |
+
"grad_norm": 1.5787513256072998,
|
| 210 |
+
"learning_rate": 6.475618600470322e-05,
|
| 211 |
+
"loss": 0.7992,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.060362173038229376,
|
| 216 |
+
"grad_norm": 1.4350961446762085,
|
| 217 |
+
"learning_rate": 6.472988253307773e-05,
|
| 218 |
+
"loss": 0.7524,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.06237424547283702,
|
| 223 |
+
"grad_norm": 1.391878604888916,
|
| 224 |
+
"learning_rate": 6.470223784448298e-05,
|
| 225 |
+
"loss": 0.7541,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.06438631790744467,
|
| 230 |
+
"grad_norm": 1.4117164611816406,
|
| 231 |
+
"learning_rate": 6.467325308932728e-05,
|
| 232 |
+
"loss": 0.7899,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.06639839034205232,
|
| 237 |
+
"grad_norm": 1.3123915195465088,
|
| 238 |
+
"learning_rate": 6.464292947378462e-05,
|
| 239 |
+
"loss": 0.7402,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.06841046277665996,
|
| 244 |
+
"grad_norm": 1.4149091243743896,
|
| 245 |
+
"learning_rate": 6.46112682597444e-05,
|
| 246 |
+
"loss": 0.8195,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.07042253521126761,
|
| 251 |
+
"grad_norm": 1.2951886653900146,
|
| 252 |
+
"learning_rate": 6.457827076475899e-05,
|
| 253 |
+
"loss": 0.7179,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.07243460764587525,
|
| 258 |
+
"grad_norm": 1.3390341997146606,
|
| 259 |
+
"learning_rate": 6.454393836198883e-05,
|
| 260 |
+
"loss": 0.7973,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.0744466800804829,
|
| 265 |
+
"grad_norm": 1.307084560394287,
|
| 266 |
+
"learning_rate": 6.450827248014535e-05,
|
| 267 |
+
"loss": 0.7591,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.07645875251509054,
|
| 272 |
+
"grad_norm": 1.3605775833129883,
|
| 273 |
+
"learning_rate": 6.447127460343148e-05,
|
| 274 |
+
"loss": 0.8085,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.07847082494969819,
|
| 279 |
+
"grad_norm": 1.4956183433532715,
|
| 280 |
+
"learning_rate": 6.443294627147988e-05,
|
| 281 |
+
"loss": 0.735,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.08048289738430583,
|
| 286 |
+
"grad_norm": 1.371940016746521,
|
| 287 |
+
"learning_rate": 6.439328907928889e-05,
|
| 288 |
+
"loss": 0.8025,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.08249496981891348,
|
| 293 |
+
"grad_norm": 1.4716073274612427,
|
| 294 |
+
"learning_rate": 6.435230467715615e-05,
|
| 295 |
+
"loss": 0.7381,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.08450704225352113,
|
| 300 |
+
"grad_norm": 1.3676444292068481,
|
| 301 |
+
"learning_rate": 6.430999477060994e-05,
|
| 302 |
+
"loss": 0.7601,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.08651911468812877,
|
| 307 |
+
"grad_norm": 1.5072672367095947,
|
| 308 |
+
"learning_rate": 6.426636112033818e-05,
|
| 309 |
+
"loss": 0.8209,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.08853118712273642,
|
| 314 |
+
"grad_norm": 1.3186477422714233,
|
| 315 |
+
"learning_rate": 6.422140554211511e-05,
|
| 316 |
+
"loss": 0.777,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.09054325955734406,
|
| 321 |
+
"grad_norm": 1.4689198732376099,
|
| 322 |
+
"learning_rate": 6.417512990672591e-05,
|
| 323 |
+
"loss": 0.871,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.0925553319919517,
|
| 328 |
+
"grad_norm": 1.287394642829895,
|
| 329 |
+
"learning_rate": 6.412753613988862e-05,
|
| 330 |
+
"loss": 0.7281,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.09456740442655935,
|
| 335 |
+
"grad_norm": 1.1629823446273804,
|
| 336 |
+
"learning_rate": 6.407862622217414e-05,
|
| 337 |
+
"loss": 0.715,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.096579476861167,
|
| 342 |
+
"grad_norm": 1.2146556377410889,
|
| 343 |
+
"learning_rate": 6.402840218892381e-05,
|
| 344 |
+
"loss": 0.6985,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.09859154929577464,
|
| 349 |
+
"grad_norm": 1.3280084133148193,
|
| 350 |
+
"learning_rate": 6.397686613016466e-05,
|
| 351 |
+
"loss": 0.8195,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.1006036217303823,
|
| 356 |
+
"grad_norm": 1.157867193222046,
|
| 357 |
+
"learning_rate": 6.392402019052244e-05,
|
| 358 |
+
"loss": 0.7529,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.10261569416498995,
|
| 363 |
+
"grad_norm": 1.1449801921844482,
|
| 364 |
+
"learning_rate": 6.386986656913242e-05,
|
| 365 |
+
"loss": 0.7834,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.10462776659959759,
|
| 370 |
+
"grad_norm": 1.2190605401992798,
|
| 371 |
+
"learning_rate": 6.381440751954782e-05,
|
| 372 |
+
"loss": 0.8203,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.10663983903420524,
|
| 377 |
+
"grad_norm": 1.2562769651412964,
|
| 378 |
+
"learning_rate": 6.375764534964608e-05,
|
| 379 |
+
"loss": 0.7338,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.10865191146881288,
|
| 384 |
+
"grad_norm": 1.1834882497787476,
|
| 385 |
+
"learning_rate": 6.369958242153278e-05,
|
| 386 |
+
"loss": 0.7489,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.11066398390342053,
|
| 391 |
+
"grad_norm": 1.2381786108016968,
|
| 392 |
+
"learning_rate": 6.364022115144337e-05,
|
| 393 |
+
"loss": 0.7537,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.11267605633802817,
|
| 398 |
+
"grad_norm": 1.256866693496704,
|
| 399 |
+
"learning_rate": 6.357956400964259e-05,
|
| 400 |
+
"loss": 0.7725,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.11468812877263582,
|
| 405 |
+
"grad_norm": 1.2332978248596191,
|
| 406 |
+
"learning_rate": 6.351761352032171e-05,
|
| 407 |
+
"loss": 0.774,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.11670020120724346,
|
| 412 |
+
"grad_norm": 1.3557363748550415,
|
| 413 |
+
"learning_rate": 6.345437226149345e-05,
|
| 414 |
+
"loss": 0.7877,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.11871227364185111,
|
| 419 |
+
"grad_norm": 1.1381844282150269,
|
| 420 |
+
"learning_rate": 6.338984286488472e-05,
|
| 421 |
+
"loss": 0.7152,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.12072434607645875,
|
| 426 |
+
"grad_norm": 1.147226095199585,
|
| 427 |
+
"learning_rate": 6.332402801582716e-05,
|
| 428 |
+
"loss": 0.7713,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.1227364185110664,
|
| 433 |
+
"grad_norm": 1.0942986011505127,
|
| 434 |
+
"learning_rate": 6.325693045314524e-05,
|
| 435 |
+
"loss": 0.7446,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.12474849094567404,
|
| 440 |
+
"grad_norm": 1.233619213104248,
|
| 441 |
+
"learning_rate": 6.318855296904246e-05,
|
| 442 |
+
"loss": 0.8107,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.1267605633802817,
|
| 447 |
+
"grad_norm": 1.1866984367370605,
|
| 448 |
+
"learning_rate": 6.311889840898498e-05,
|
| 449 |
+
"loss": 0.8206,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.12877263581488935,
|
| 454 |
+
"grad_norm": 1.176274061203003,
|
| 455 |
+
"learning_rate": 6.304796967158343e-05,
|
| 456 |
+
"loss": 0.7813,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.13078470824949698,
|
| 461 |
+
"grad_norm": 1.1214070320129395,
|
| 462 |
+
"learning_rate": 6.297576970847205e-05,
|
| 463 |
+
"loss": 0.6908,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.13279678068410464,
|
| 468 |
+
"grad_norm": 1.0725229978561401,
|
| 469 |
+
"learning_rate": 6.290230152418603e-05,
|
| 470 |
+
"loss": 0.756,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.13480885311871227,
|
| 475 |
+
"grad_norm": 1.1194369792938232,
|
| 476 |
+
"learning_rate": 6.28275681760364e-05,
|
| 477 |
+
"loss": 0.7766,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.13682092555331993,
|
| 482 |
+
"grad_norm": 1.090072512626648,
|
| 483 |
+
"learning_rate": 6.275157277398282e-05,
|
| 484 |
+
"loss": 0.7856,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.13883299798792756,
|
| 489 |
+
"grad_norm": 1.1602994203567505,
|
| 490 |
+
"learning_rate": 6.267431848050418e-05,
|
| 491 |
+
"loss": 0.7597,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.14084507042253522,
|
| 496 |
+
"grad_norm": 1.0984241962432861,
|
| 497 |
+
"learning_rate": 6.259580851046697e-05,
|
| 498 |
+
"loss": 0.7318,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.14285714285714285,
|
| 503 |
+
"grad_norm": 1.0639097690582275,
|
| 504 |
+
"learning_rate": 6.251604613099149e-05,
|
| 505 |
+
"loss": 0.7265,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.1448692152917505,
|
| 510 |
+
"grad_norm": 1.0951672792434692,
|
| 511 |
+
"learning_rate": 6.243503466131597e-05,
|
| 512 |
+
"loss": 0.7312,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.14688128772635814,
|
| 517 |
+
"grad_norm": 1.092551827430725,
|
| 518 |
+
"learning_rate": 6.235277747265828e-05,
|
| 519 |
+
"loss": 0.7556,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.1488933601609658,
|
| 524 |
+
"grad_norm": 1.1054296493530273,
|
| 525 |
+
"learning_rate": 6.226927798807585e-05,
|
| 526 |
+
"loss": 0.8489,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.15090543259557343,
|
| 531 |
+
"grad_norm": 1.0174497365951538,
|
| 532 |
+
"learning_rate": 6.218453968232303e-05,
|
| 533 |
+
"loss": 0.7104,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.1529175050301811,
|
| 538 |
+
"grad_norm": 1.146209716796875,
|
| 539 |
+
"learning_rate": 6.209856608170663e-05,
|
| 540 |
+
"loss": 0.7276,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.15492957746478872,
|
| 545 |
+
"grad_norm": 1.0379316806793213,
|
| 546 |
+
"learning_rate": 6.201136076393911e-05,
|
| 547 |
+
"loss": 0.7243,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.15694164989939638,
|
| 552 |
+
"grad_norm": 1.0034286975860596,
|
| 553 |
+
"learning_rate": 6.192292735798967e-05,
|
| 554 |
+
"loss": 0.7085,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.158953722334004,
|
| 559 |
+
"grad_norm": 1.0072968006134033,
|
| 560 |
+
"learning_rate": 6.183326954393332e-05,
|
| 561 |
+
"loss": 0.6865,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.16096579476861167,
|
| 566 |
+
"grad_norm": 1.0766922235488892,
|
| 567 |
+
"learning_rate": 6.174239105279768e-05,
|
| 568 |
+
"loss": 0.7628,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.16297786720321933,
|
| 573 |
+
"grad_norm": 1.014816403388977,
|
| 574 |
+
"learning_rate": 6.16502956664077e-05,
|
| 575 |
+
"loss": 0.7356,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.16498993963782696,
|
| 580 |
+
"grad_norm": 1.055881142616272,
|
| 581 |
+
"learning_rate": 6.155698721722834e-05,
|
| 582 |
+
"loss": 0.7763,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.16700201207243462,
|
| 587 |
+
"grad_norm": 1.0576510429382324,
|
| 588 |
+
"learning_rate": 6.146246958820503e-05,
|
| 589 |
+
"loss": 0.6852,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.16901408450704225,
|
| 594 |
+
"grad_norm": 1.2542794942855835,
|
| 595 |
+
"learning_rate": 6.136674671260213e-05,
|
| 596 |
+
"loss": 0.7956,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.1710261569416499,
|
| 601 |
+
"grad_norm": 1.0826421976089478,
|
| 602 |
+
"learning_rate": 6.12698225738392e-05,
|
| 603 |
+
"loss": 0.7285,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.17303822937625754,
|
| 608 |
+
"grad_norm": 1.0286874771118164,
|
| 609 |
+
"learning_rate": 6.11717012053253e-05,
|
| 610 |
+
"loss": 0.7363,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.1750503018108652,
|
| 615 |
+
"grad_norm": 1.1285161972045898,
|
| 616 |
+
"learning_rate": 6.107238669029108e-05,
|
| 617 |
+
"loss": 0.7807,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.17706237424547283,
|
| 622 |
+
"grad_norm": 2.394617795944214,
|
| 623 |
+
"learning_rate": 6.0971883161618906e-05,
|
| 624 |
+
"loss": 0.663,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.1790744466800805,
|
| 629 |
+
"grad_norm": 1.0971828699111938,
|
| 630 |
+
"learning_rate": 6.0870194801670825e-05,
|
| 631 |
+
"loss": 0.7331,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.18108651911468812,
|
| 636 |
+
"grad_norm": 1.0546624660491943,
|
| 637 |
+
"learning_rate": 6.076732584211459e-05,
|
| 638 |
+
"loss": 0.7244,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.18309859154929578,
|
| 643 |
+
"grad_norm": 1.0969500541687012,
|
| 644 |
+
"learning_rate": 6.06632805637475e-05,
|
| 645 |
+
"loss": 0.8122,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.1851106639839034,
|
| 650 |
+
"grad_norm": 1.0038318634033203,
|
| 651 |
+
"learning_rate": 6.055806329631827e-05,
|
| 652 |
+
"loss": 0.752,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.18712273641851107,
|
| 657 |
+
"grad_norm": 1.017544150352478,
|
| 658 |
+
"learning_rate": 6.045167841834689e-05,
|
| 659 |
+
"loss": 0.7629,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.1891348088531187,
|
| 664 |
+
"grad_norm": 1.0679768323898315,
|
| 665 |
+
"learning_rate": 6.034413035694237e-05,
|
| 666 |
+
"loss": 0.7533,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.19114688128772636,
|
| 671 |
+
"grad_norm": 1.1002590656280518,
|
| 672 |
+
"learning_rate": 6.023542358761855e-05,
|
| 673 |
+
"loss": 0.7566,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.193158953722334,
|
| 678 |
+
"grad_norm": 1.0083142518997192,
|
| 679 |
+
"learning_rate": 6.01255626341078e-05,
|
| 680 |
+
"loss": 0.6873,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.19517102615694165,
|
| 685 |
+
"grad_norm": 1.897566318511963,
|
| 686 |
+
"learning_rate": 6.0014552068172866e-05,
|
| 687 |
+
"loss": 0.8056,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.19718309859154928,
|
| 692 |
+
"grad_norm": 0.995782732963562,
|
| 693 |
+
"learning_rate": 5.990239650941649e-05,
|
| 694 |
+
"loss": 0.7062,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.19919517102615694,
|
| 699 |
+
"grad_norm": 0.9380747079849243,
|
| 700 |
+
"learning_rate": 5.97891006250893e-05,
|
| 701 |
+
"loss": 0.6674,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2012072434607646,
|
| 706 |
+
"grad_norm": 1.0377370119094849,
|
| 707 |
+
"learning_rate": 5.967466912989547e-05,
|
| 708 |
+
"loss": 0.7425,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.20321931589537223,
|
| 713 |
+
"grad_norm": 0.9492895603179932,
|
| 714 |
+
"learning_rate": 5.955910678579663e-05,
|
| 715 |
+
"loss": 0.7232,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.2052313883299799,
|
| 720 |
+
"grad_norm": 0.9299039244651794,
|
| 721 |
+
"learning_rate": 5.9442418401813614e-05,
|
| 722 |
+
"loss": 0.6996,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.20724346076458752,
|
| 727 |
+
"grad_norm": 1.0433120727539062,
|
| 728 |
+
"learning_rate": 5.932460883382638e-05,
|
| 729 |
+
"loss": 0.7504,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.20925553319919518,
|
| 734 |
+
"grad_norm": 1.02601158618927,
|
| 735 |
+
"learning_rate": 5.920568298437193e-05,
|
| 736 |
+
"loss": 0.7691,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.2112676056338028,
|
| 741 |
+
"grad_norm": 1.0401028394699097,
|
| 742 |
+
"learning_rate": 5.90856458024403e-05,
|
| 743 |
+
"loss": 0.7038,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.21327967806841047,
|
| 748 |
+
"grad_norm": 0.9615976214408875,
|
| 749 |
+
"learning_rate": 5.8964502283268614e-05,
|
| 750 |
+
"loss": 0.6894,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2152917505030181,
|
| 755 |
+
"grad_norm": 1.0715394020080566,
|
| 756 |
+
"learning_rate": 5.884225746813317e-05,
|
| 757 |
+
"loss": 0.762,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.21730382293762576,
|
| 762 |
+
"grad_norm": 1.0785269737243652,
|
| 763 |
+
"learning_rate": 5.871891644413972e-05,
|
| 764 |
+
"loss": 0.7538,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.2193158953722334,
|
| 769 |
+
"grad_norm": 0.9964181780815125,
|
| 770 |
+
"learning_rate": 5.8594484344011724e-05,
|
| 771 |
+
"loss": 0.7205,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.22132796780684105,
|
| 776 |
+
"grad_norm": 1.0187814235687256,
|
| 777 |
+
"learning_rate": 5.84689663458768e-05,
|
| 778 |
+
"loss": 0.6946,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.22334004024144868,
|
| 783 |
+
"grad_norm": 0.949118435382843,
|
| 784 |
+
"learning_rate": 5.834236767305117e-05,
|
| 785 |
+
"loss": 0.6963,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.22535211267605634,
|
| 790 |
+
"grad_norm": 0.9602085947990417,
|
| 791 |
+
"learning_rate": 5.821469359382237e-05,
|
| 792 |
+
"loss": 0.7319,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.22736418511066397,
|
| 797 |
+
"grad_norm": 0.9620065689086914,
|
| 798 |
+
"learning_rate": 5.808594942122999e-05,
|
| 799 |
+
"loss": 0.7442,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.22937625754527163,
|
| 804 |
+
"grad_norm": 0.9228153228759766,
|
| 805 |
+
"learning_rate": 5.795614051284457e-05,
|
| 806 |
+
"loss": 0.7212,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.23138832997987926,
|
| 811 |
+
"grad_norm": 1.029982328414917,
|
| 812 |
+
"learning_rate": 5.782527227054462e-05,
|
| 813 |
+
"loss": 0.7445,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.23340040241448692,
|
| 818 |
+
"grad_norm": 1.0322256088256836,
|
| 819 |
+
"learning_rate": 5.769335014029191e-05,
|
| 820 |
+
"loss": 0.6763,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.23541247484909456,
|
| 825 |
+
"grad_norm": 1.071828007698059,
|
| 826 |
+
"learning_rate": 5.7560379611904754e-05,
|
| 827 |
+
"loss": 0.7681,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.23742454728370221,
|
| 832 |
+
"grad_norm": 1.0592682361602783,
|
| 833 |
+
"learning_rate": 5.7426366218829596e-05,
|
| 834 |
+
"loss": 0.7557,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.23943661971830985,
|
| 839 |
+
"grad_norm": 1.019776701927185,
|
| 840 |
+
"learning_rate": 5.729131553791076e-05,
|
| 841 |
+
"loss": 0.7524,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.2414486921529175,
|
| 846 |
+
"grad_norm": 1.0034253597259521,
|
| 847 |
+
"learning_rate": 5.715523318915832e-05,
|
| 848 |
+
"loss": 0.7844,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.24346076458752516,
|
| 853 |
+
"grad_norm": 1.0709611177444458,
|
| 854 |
+
"learning_rate": 5.701812483551428e-05,
|
| 855 |
+
"loss": 0.7771,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.2454728370221328,
|
| 860 |
+
"grad_norm": 1.089322805404663,
|
| 861 |
+
"learning_rate": 5.687999618261688e-05,
|
| 862 |
+
"loss": 0.7913,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.24748490945674045,
|
| 867 |
+
"grad_norm": 0.981561005115509,
|
| 868 |
+
"learning_rate": 5.6740852978563204e-05,
|
| 869 |
+
"loss": 0.744,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.24949698189134809,
|
| 874 |
+
"grad_norm": 0.9540185928344727,
|
| 875 |
+
"learning_rate": 5.66007010136699e-05,
|
| 876 |
+
"loss": 0.7201,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.2515090543259557,
|
| 881 |
+
"grad_norm": 1.0497649908065796,
|
| 882 |
+
"learning_rate": 5.645954612023233e-05,
|
| 883 |
+
"loss": 0.7942,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.2535211267605634,
|
| 888 |
+
"grad_norm": 0.9719651341438293,
|
| 889 |
+
"learning_rate": 5.631739417228178e-05,
|
| 890 |
+
"loss": 0.7133,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.25553319919517103,
|
| 895 |
+
"grad_norm": 0.9591907858848572,
|
| 896 |
+
"learning_rate": 5.6174251085341044e-05,
|
| 897 |
+
"loss": 0.7118,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.2575452716297787,
|
| 902 |
+
"grad_norm": 0.999717652797699,
|
| 903 |
+
"learning_rate": 5.603012281617824e-05,
|
| 904 |
+
"loss": 0.745,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.2595573440643863,
|
| 909 |
+
"grad_norm": 0.984555184841156,
|
| 910 |
+
"learning_rate": 5.5885015362558974e-05,
|
| 911 |
+
"loss": 0.7398,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.26156941649899396,
|
| 916 |
+
"grad_norm": 1.0207607746124268,
|
| 917 |
+
"learning_rate": 5.573893476299669e-05,
|
| 918 |
+
"loss": 0.7432,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.2635814889336016,
|
| 923 |
+
"grad_norm": 1.0601558685302734,
|
| 924 |
+
"learning_rate": 5.55918870965014e-05,
|
| 925 |
+
"loss": 0.7713,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.2655935613682093,
|
| 930 |
+
"grad_norm": 1.0741450786590576,
|
| 931 |
+
"learning_rate": 5.544387848232677e-05,
|
| 932 |
+
"loss": 0.7851,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.2676056338028169,
|
| 937 |
+
"grad_norm": 1.0342241525650024,
|
| 938 |
+
"learning_rate": 5.5294915079715334e-05,
|
| 939 |
+
"loss": 0.7593,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.26961770623742454,
|
| 944 |
+
"grad_norm": 0.983668863773346,
|
| 945 |
+
"learning_rate": 5.514500308764235e-05,
|
| 946 |
+
"loss": 0.7426,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.2716297786720322,
|
| 951 |
+
"grad_norm": 0.9781654477119446,
|
| 952 |
+
"learning_rate": 5.499414874455771e-05,
|
| 953 |
+
"loss": 0.7306,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.27364185110663986,
|
| 958 |
+
"grad_norm": 1.0362467765808105,
|
| 959 |
+
"learning_rate": 5.48423583281264e-05,
|
| 960 |
+
"loss": 0.7742,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.27565392354124746,
|
| 965 |
+
"grad_norm": 0.9543638229370117,
|
| 966 |
+
"learning_rate": 5.4689638154967235e-05,
|
| 967 |
+
"loss": 0.6956,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.2776659959758551,
|
| 972 |
+
"grad_norm": 1.0143051147460938,
|
| 973 |
+
"learning_rate": 5.453599458038998e-05,
|
| 974 |
+
"loss": 0.7562,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.2796780684104628,
|
| 979 |
+
"grad_norm": 0.9799973368644714,
|
| 980 |
+
"learning_rate": 5.438143399813095e-05,
|
| 981 |
+
"loss": 0.7326,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.28169014084507044,
|
| 986 |
+
"grad_norm": 0.9217185974121094,
|
| 987 |
+
"learning_rate": 5.422596284008683e-05,
|
| 988 |
+
"loss": 0.7464,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.2837022132796781,
|
| 993 |
+
"grad_norm": 0.8847222328186035,
|
| 994 |
+
"learning_rate": 5.4069587576047144e-05,
|
| 995 |
+
"loss": 0.7148,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.2857142857142857,
|
| 1000 |
+
"grad_norm": 0.8999807834625244,
|
| 1001 |
+
"learning_rate": 5.39123147134249e-05,
|
| 1002 |
+
"loss": 0.7207,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.28772635814889336,
|
| 1007 |
+
"grad_norm": 0.9449793100357056,
|
| 1008 |
+
"learning_rate": 5.3754150796985895e-05,
|
| 1009 |
+
"loss": 0.738,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.289738430583501,
|
| 1014 |
+
"grad_norm": 1.1088876724243164,
|
| 1015 |
+
"learning_rate": 5.359510240857628e-05,
|
| 1016 |
+
"loss": 0.7431,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.2917505030181087,
|
| 1021 |
+
"grad_norm": 0.931364119052887,
|
| 1022 |
+
"learning_rate": 5.343517616684872e-05,
|
| 1023 |
+
"loss": 0.7132,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.2937625754527163,
|
| 1028 |
+
"grad_norm": 1.0146406888961792,
|
| 1029 |
+
"learning_rate": 5.327437872698691e-05,
|
| 1030 |
+
"loss": 0.7916,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.29577464788732394,
|
| 1035 |
+
"grad_norm": 1.0176244974136353,
|
| 1036 |
+
"learning_rate": 5.3112716780428674e-05,
|
| 1037 |
+
"loss": 0.7131,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.2977867203219316,
|
| 1042 |
+
"grad_norm": 0.9973041415214539,
|
| 1043 |
+
"learning_rate": 5.295019705458749e-05,
|
| 1044 |
+
"loss": 0.7689,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.29979879275653926,
|
| 1049 |
+
"grad_norm": 1.0221103429794312,
|
| 1050 |
+
"learning_rate": 5.278682631257252e-05,
|
| 1051 |
+
"loss": 0.7762,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.30181086519114686,
|
| 1056 |
+
"grad_norm": 0.9501760601997375,
|
| 1057 |
+
"learning_rate": 5.2622611352907214e-05,
|
| 1058 |
+
"loss": 0.7103,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.3038229376257545,
|
| 1063 |
+
"grad_norm": 0.9289485216140747,
|
| 1064 |
+
"learning_rate": 5.245755900924631e-05,
|
| 1065 |
+
"loss": 0.6982,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.3058350100603622,
|
| 1070 |
+
"grad_norm": 1.4150779247283936,
|
| 1071 |
+
"learning_rate": 5.2291676150091576e-05,
|
| 1072 |
+
"loss": 0.7035,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.30784708249496984,
|
| 1077 |
+
"grad_norm": 0.9871881008148193,
|
| 1078 |
+
"learning_rate": 5.2124969678505864e-05,
|
| 1079 |
+
"loss": 0.7388,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.30985915492957744,
|
| 1084 |
+
"grad_norm": 0.9766415357589722,
|
| 1085 |
+
"learning_rate": 5.195744653182594e-05,
|
| 1086 |
+
"loss": 0.7142,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.3118712273641851,
|
| 1091 |
+
"grad_norm": 0.9838590621948242,
|
| 1092 |
+
"learning_rate": 5.178911368137373e-05,
|
| 1093 |
+
"loss": 0.6978,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.31388329979879276,
|
| 1098 |
+
"grad_norm": 0.9251114726066589,
|
| 1099 |
+
"learning_rate": 5.1619978132166254e-05,
|
| 1100 |
+
"loss": 0.6737,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.3158953722334004,
|
| 1105 |
+
"grad_norm": 0.9593889117240906,
|
| 1106 |
+
"learning_rate": 5.145004692262411e-05,
|
| 1107 |
+
"loss": 0.7233,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.317907444668008,
|
| 1112 |
+
"grad_norm": 0.9678305387496948,
|
| 1113 |
+
"learning_rate": 5.127932712427851e-05,
|
| 1114 |
+
"loss": 0.702,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.3199195171026157,
|
| 1119 |
+
"grad_norm": 0.9605986475944519,
|
| 1120 |
+
"learning_rate": 5.110782584147718e-05,
|
| 1121 |
+
"loss": 0.7678,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.32193158953722334,
|
| 1126 |
+
"grad_norm": 0.9980550408363342,
|
| 1127 |
+
"learning_rate": 5.093555021108849e-05,
|
| 1128 |
+
"loss": 0.7299,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.323943661971831,
|
| 1133 |
+
"grad_norm": 0.9018529057502747,
|
| 1134 |
+
"learning_rate": 5.0762507402204637e-05,
|
| 1135 |
+
"loss": 0.7109,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.32595573440643866,
|
| 1140 |
+
"grad_norm": 1.1389586925506592,
|
| 1141 |
+
"learning_rate": 5.058870461584322e-05,
|
| 1142 |
+
"loss": 0.7371,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.32796780684104626,
|
| 1147 |
+
"grad_norm": 1.0409871339797974,
|
| 1148 |
+
"learning_rate": 5.0414149084647644e-05,
|
| 1149 |
+
"loss": 0.8062,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.3299798792756539,
|
| 1154 |
+
"grad_norm": 0.9527642130851746,
|
| 1155 |
+
"learning_rate": 5.0238848072586037e-05,
|
| 1156 |
+
"loss": 0.7692,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.3319919517102616,
|
| 1161 |
+
"grad_norm": 0.9563398361206055,
|
| 1162 |
+
"learning_rate": 5.0062808874649074e-05,
|
| 1163 |
+
"loss": 0.737,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.33400402414486924,
|
| 1168 |
+
"grad_norm": 1.0413488149642944,
|
| 1169 |
+
"learning_rate": 4.9886038816546355e-05,
|
| 1170 |
+
"loss": 0.7503,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.33601609657947684,
|
| 1175 |
+
"grad_norm": 0.9624356031417847,
|
| 1176 |
+
"learning_rate": 4.970854525440152e-05,
|
| 1177 |
+
"loss": 0.7034,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.3380281690140845,
|
| 1182 |
+
"grad_norm": 0.9548975229263306,
|
| 1183 |
+
"learning_rate": 4.9530335574446196e-05,
|
| 1184 |
+
"loss": 0.7213,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.34004024144869216,
|
| 1189 |
+
"grad_norm": 0.8841491937637329,
|
| 1190 |
+
"learning_rate": 4.9351417192712596e-05,
|
| 1191 |
+
"loss": 0.685,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.3420523138832998,
|
| 1196 |
+
"grad_norm": 0.989604115486145,
|
| 1197 |
+
"learning_rate": 4.917179755472491e-05,
|
| 1198 |
+
"loss": 0.6952,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.3440643863179074,
|
| 1203 |
+
"grad_norm": 0.8883060216903687,
|
| 1204 |
+
"learning_rate": 4.899148413518942e-05,
|
| 1205 |
+
"loss": 0.697,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.3460764587525151,
|
| 1210 |
+
"grad_norm": 0.8740846514701843,
|
| 1211 |
+
"learning_rate": 4.881048443768356e-05,
|
| 1212 |
+
"loss": 0.7045,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.34808853118712274,
|
| 1217 |
+
"grad_norm": 0.8820964097976685,
|
| 1218 |
+
"learning_rate": 4.862880599434353e-05,
|
| 1219 |
+
"loss": 0.7007,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.3501006036217304,
|
| 1224 |
+
"grad_norm": 0.882949709892273,
|
| 1225 |
+
"learning_rate": 4.844645636555094e-05,
|
| 1226 |
+
"loss": 0.693,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.352112676056338,
|
| 1231 |
+
"grad_norm": 0.952804684638977,
|
| 1232 |
+
"learning_rate": 4.8263443139618176e-05,
|
| 1233 |
+
"loss": 0.7319,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.35412474849094566,
|
| 1238 |
+
"grad_norm": 0.9323515295982361,
|
| 1239 |
+
"learning_rate": 4.8079773932472584e-05,
|
| 1240 |
+
"loss": 0.7206,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.3561368209255533,
|
| 1245 |
+
"grad_norm": 1.0918467044830322,
|
| 1246 |
+
"learning_rate": 4.78954563873396e-05,
|
| 1247 |
+
"loss": 0.7152,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.358148893360161,
|
| 1252 |
+
"grad_norm": 2.0716049671173096,
|
| 1253 |
+
"learning_rate": 4.7710498174424614e-05,
|
| 1254 |
+
"loss": 0.7769,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.36016096579476864,
|
| 1259 |
+
"grad_norm": 0.9245615601539612,
|
| 1260 |
+
"learning_rate": 4.7524906990593844e-05,
|
| 1261 |
+
"loss": 0.7085,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.36217303822937624,
|
| 1266 |
+
"grad_norm": 1.0692616701126099,
|
| 1267 |
+
"learning_rate": 4.7338690559054004e-05,
|
| 1268 |
+
"loss": 0.6948,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.3641851106639839,
|
| 1273 |
+
"grad_norm": 0.8987394571304321,
|
| 1274 |
+
"learning_rate": 4.71518566290309e-05,
|
| 1275 |
+
"loss": 0.7167,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.36619718309859156,
|
| 1280 |
+
"grad_norm": 0.946850061416626,
|
| 1281 |
+
"learning_rate": 4.6964412975447e-05,
|
| 1282 |
+
"loss": 0.7227,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.3682092555331992,
|
| 1287 |
+
"grad_norm": 0.9121927618980408,
|
| 1288 |
+
"learning_rate": 4.6776367398597846e-05,
|
| 1289 |
+
"loss": 0.6868,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.3702213279678068,
|
| 1294 |
+
"grad_norm": 0.8747940063476562,
|
| 1295 |
+
"learning_rate": 4.658772772382744e-05,
|
| 1296 |
+
"loss": 0.6797,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.3722334004024145,
|
| 1301 |
+
"grad_norm": 0.9101600646972656,
|
| 1302 |
+
"learning_rate": 4.639850180120268e-05,
|
| 1303 |
+
"loss": 0.7636,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.37424547283702214,
|
| 1308 |
+
"grad_norm": 0.9515472054481506,
|
| 1309 |
+
"learning_rate": 4.620869750518656e-05,
|
| 1310 |
+
"loss": 0.7357,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.3762575452716298,
|
| 1315 |
+
"grad_norm": 0.9573916792869568,
|
| 1316 |
+
"learning_rate": 4.60183227343106e-05,
|
| 1317 |
+
"loss": 0.7303,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.3782696177062374,
|
| 1322 |
+
"grad_norm": 0.942021369934082,
|
| 1323 |
+
"learning_rate": 4.582738541084612e-05,
|
| 1324 |
+
"loss": 0.709,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.38028169014084506,
|
| 1329 |
+
"grad_norm": 0.8716509342193604,
|
| 1330 |
+
"learning_rate": 4.5635893480474474e-05,
|
| 1331 |
+
"loss": 0.6681,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.3822937625754527,
|
| 1336 |
+
"grad_norm": 0.8702856302261353,
|
| 1337 |
+
"learning_rate": 4.544385491195658e-05,
|
| 1338 |
+
"loss": 0.6902,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.3843058350100604,
|
| 1343 |
+
"grad_norm": 0.9312735795974731,
|
| 1344 |
+
"learning_rate": 4.525127769680109e-05,
|
| 1345 |
+
"loss": 0.6895,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.386317907444668,
|
| 1350 |
+
"grad_norm": 0.9693159461021423,
|
| 1351 |
+
"learning_rate": 4.505816984893203e-05,
|
| 1352 |
+
"loss": 0.7286,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.38832997987927564,
|
| 1357 |
+
"grad_norm": 1.0244489908218384,
|
| 1358 |
+
"learning_rate": 4.486453940435518e-05,
|
| 1359 |
+
"loss": 0.7899,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.3903420523138833,
|
| 1364 |
+
"grad_norm": 0.970856249332428,
|
| 1365 |
+
"learning_rate": 4.46703944208237e-05,
|
| 1366 |
+
"loss": 0.7176,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.39235412474849096,
|
| 1371 |
+
"grad_norm": 0.9749726057052612,
|
| 1372 |
+
"learning_rate": 4.447574297750281e-05,
|
| 1373 |
+
"loss": 0.6914,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.39436619718309857,
|
| 1378 |
+
"grad_norm": 0.9390575885772705,
|
| 1379 |
+
"learning_rate": 4.428059317463363e-05,
|
| 1380 |
+
"loss": 0.7125,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.3963782696177062,
|
| 1385 |
+
"grad_norm": 0.9480944871902466,
|
| 1386 |
+
"learning_rate": 4.4084953133196e-05,
|
| 1387 |
+
"loss": 0.681,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.3983903420523139,
|
| 1392 |
+
"grad_norm": 0.9078746438026428,
|
| 1393 |
+
"learning_rate": 4.388883099457064e-05,
|
| 1394 |
+
"loss": 0.6917,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.40040241448692154,
|
| 1399 |
+
"grad_norm": 0.9508323073387146,
|
| 1400 |
+
"learning_rate": 4.369223492020026e-05,
|
| 1401 |
+
"loss": 0.7082,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.4024144869215292,
|
| 1406 |
+
"grad_norm": 0.8641968965530396,
|
| 1407 |
+
"learning_rate": 4.3495173091249974e-05,
|
| 1408 |
+
"loss": 0.6343,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.4044265593561368,
|
| 1413 |
+
"grad_norm": 0.9805442094802856,
|
| 1414 |
+
"learning_rate": 4.329765370826689e-05,
|
| 1415 |
+
"loss": 0.7091,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.40643863179074446,
|
| 1420 |
+
"grad_norm": 0.9265369772911072,
|
| 1421 |
+
"learning_rate": 4.309968499083876e-05,
|
| 1422 |
+
"loss": 0.6812,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.4084507042253521,
|
| 1427 |
+
"grad_norm": 0.9390007257461548,
|
| 1428 |
+
"learning_rate": 4.290127517725199e-05,
|
| 1429 |
+
"loss": 0.7208,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.4104627766599598,
|
| 1434 |
+
"grad_norm": 0.918609619140625,
|
| 1435 |
+
"learning_rate": 4.27024325241488e-05,
|
| 1436 |
+
"loss": 0.6829,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.4124748490945674,
|
| 1441 |
+
"grad_norm": 0.9491612315177917,
|
| 1442 |
+
"learning_rate": 4.250316530618363e-05,
|
| 1443 |
+
"loss": 0.749,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.41448692152917505,
|
| 1448 |
+
"grad_norm": 0.9603286385536194,
|
| 1449 |
+
"learning_rate": 4.23034818156788e-05,
|
| 1450 |
+
"loss": 0.7192,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.4164989939637827,
|
| 1455 |
+
"grad_norm": 0.8992831110954285,
|
| 1456 |
+
"learning_rate": 4.210339036227942e-05,
|
| 1457 |
+
"loss": 0.6715,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.41851106639839036,
|
| 1462 |
+
"grad_norm": 0.9181328415870667,
|
| 1463 |
+
"learning_rate": 4.1902899272607605e-05,
|
| 1464 |
+
"loss": 0.7586,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.42052313883299797,
|
| 1469 |
+
"grad_norm": 0.9009472131729126,
|
| 1470 |
+
"learning_rate": 4.1702016889915974e-05,
|
| 1471 |
+
"loss": 0.7139,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.4225352112676056,
|
| 1476 |
+
"grad_norm": 0.9172364473342896,
|
| 1477 |
+
"learning_rate": 4.150075157374043e-05,
|
| 1478 |
+
"loss": 0.7403,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.4245472837022133,
|
| 1483 |
+
"grad_norm": 0.9123634099960327,
|
| 1484 |
+
"learning_rate": 4.1299111699552316e-05,
|
| 1485 |
+
"loss": 0.7326,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.42655935613682094,
|
| 1490 |
+
"grad_norm": 0.8299218416213989,
|
| 1491 |
+
"learning_rate": 4.109710565840987e-05,
|
| 1492 |
+
"loss": 0.6788,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.42857142857142855,
|
| 1497 |
+
"grad_norm": 0.9061017632484436,
|
| 1498 |
+
"learning_rate": 4.0894741856608986e-05,
|
| 1499 |
+
"loss": 0.6942,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.4305835010060362,
|
| 1504 |
+
"grad_norm": 0.8914936184883118,
|
| 1505 |
+
"learning_rate": 4.069202871533352e-05,
|
| 1506 |
+
"loss": 0.6633,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.43259557344064387,
|
| 1511 |
+
"grad_norm": 0.8976927995681763,
|
| 1512 |
+
"learning_rate": 4.048897467030468e-05,
|
| 1513 |
+
"loss": 0.6815,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.4346076458752515,
|
| 1518 |
+
"grad_norm": 0.869555652141571,
|
| 1519 |
+
"learning_rate": 4.028558817143013e-05,
|
| 1520 |
+
"loss": 0.6925,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.43661971830985913,
|
| 1525 |
+
"grad_norm": 0.8817659020423889,
|
| 1526 |
+
"learning_rate": 4.008187768245229e-05,
|
| 1527 |
+
"loss": 0.6621,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.4386317907444668,
|
| 1532 |
+
"grad_norm": 0.9075917601585388,
|
| 1533 |
+
"learning_rate": 3.98778516805961e-05,
|
| 1534 |
+
"loss": 0.6926,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.44064386317907445,
|
| 1539 |
+
"grad_norm": 0.8682200312614441,
|
| 1540 |
+
"learning_rate": 3.967351865621631e-05,
|
| 1541 |
+
"loss": 0.6613,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.4426559356136821,
|
| 1546 |
+
"grad_norm": 0.8848491311073303,
|
| 1547 |
+
"learning_rate": 3.94688871124441e-05,
|
| 1548 |
+
"loss": 0.7014,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.44466800804828976,
|
| 1553 |
+
"grad_norm": 0.8899165391921997,
|
| 1554 |
+
"learning_rate": 3.9263965564833295e-05,
|
| 1555 |
+
"loss": 0.6684,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.44668008048289737,
|
| 1560 |
+
"grad_norm": 0.8962348699569702,
|
| 1561 |
+
"learning_rate": 3.905876254100594e-05,
|
| 1562 |
+
"loss": 0.7285,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.448692152917505,
|
| 1567 |
+
"grad_norm": 0.8958956599235535,
|
| 1568 |
+
"learning_rate": 3.885328658029748e-05,
|
| 1569 |
+
"loss": 0.7128,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.4507042253521127,
|
| 1574 |
+
"grad_norm": 0.9015375375747681,
|
| 1575 |
+
"learning_rate": 3.8647546233401336e-05,
|
| 1576 |
+
"loss": 0.6767,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.45271629778672035,
|
| 1581 |
+
"grad_norm": 0.9416881203651428,
|
| 1582 |
+
"learning_rate": 3.8441550062013143e-05,
|
| 1583 |
+
"loss": 0.7186,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.45472837022132795,
|
| 1588 |
+
"grad_norm": 0.9649824500083923,
|
| 1589 |
+
"learning_rate": 3.8235306638474455e-05,
|
| 1590 |
+
"loss": 0.7352,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.4567404426559356,
|
| 1595 |
+
"grad_norm": 0.8740935325622559,
|
| 1596 |
+
"learning_rate": 3.802882454541594e-05,
|
| 1597 |
+
"loss": 0.6633,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.45875251509054327,
|
| 1602 |
+
"grad_norm": 0.9129350185394287,
|
| 1603 |
+
"learning_rate": 3.782211237540036e-05,
|
| 1604 |
+
"loss": 0.7051,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.4607645875251509,
|
| 1609 |
+
"grad_norm": 0.8774951100349426,
|
| 1610 |
+
"learning_rate": 3.7615178730564845e-05,
|
| 1611 |
+
"loss": 0.7249,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.46277665995975853,
|
| 1616 |
+
"grad_norm": 0.9072405099868774,
|
| 1617 |
+
"learning_rate": 3.740803222226304e-05,
|
| 1618 |
+
"loss": 0.6891,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.4647887323943662,
|
| 1623 |
+
"grad_norm": 1.0077128410339355,
|
| 1624 |
+
"learning_rate": 3.720068147070668e-05,
|
| 1625 |
+
"loss": 0.724,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.46680080482897385,
|
| 1630 |
+
"grad_norm": 0.9419925808906555,
|
| 1631 |
+
"learning_rate": 3.6993135104606915e-05,
|
| 1632 |
+
"loss": 0.7281,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.4688128772635815,
|
| 1637 |
+
"grad_norm": 0.9003535509109497,
|
| 1638 |
+
"learning_rate": 3.678540176081521e-05,
|
| 1639 |
+
"loss": 0.6508,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.4708249496981891,
|
| 1644 |
+
"grad_norm": 0.9552720785140991,
|
| 1645 |
+
"learning_rate": 3.657749008396394e-05,
|
| 1646 |
+
"loss": 0.7385,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.47283702213279677,
|
| 1651 |
+
"grad_norm": 0.9708763957023621,
|
| 1652 |
+
"learning_rate": 3.636940872610662e-05,
|
| 1653 |
+
"loss": 0.762,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.47484909456740443,
|
| 1658 |
+
"grad_norm": 0.8927114605903625,
|
| 1659 |
+
"learning_rate": 3.6161166346357934e-05,
|
| 1660 |
+
"loss": 0.7123,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.4768611670020121,
|
| 1665 |
+
"grad_norm": 0.869623601436615,
|
| 1666 |
+
"learning_rate": 3.595277161053328e-05,
|
| 1667 |
+
"loss": 0.6986,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.4788732394366197,
|
| 1672 |
+
"grad_norm": 0.8514843583106995,
|
| 1673 |
+
"learning_rate": 3.574423319078829e-05,
|
| 1674 |
+
"loss": 0.6689,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.48088531187122735,
|
| 1679 |
+
"grad_norm": 0.88021320104599,
|
| 1680 |
+
"learning_rate": 3.5535559765257785e-05,
|
| 1681 |
+
"loss": 0.6539,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.482897384305835,
|
| 1686 |
+
"grad_norm": 0.869594931602478,
|
| 1687 |
+
"learning_rate": 3.5326760017694795e-05,
|
| 1688 |
+
"loss": 0.7198,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.48490945674044267,
|
| 1693 |
+
"grad_norm": 0.9412367939949036,
|
| 1694 |
+
"learning_rate": 3.511784263710911e-05,
|
| 1695 |
+
"loss": 0.73,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.4869215291750503,
|
| 1700 |
+
"grad_norm": 0.8942079544067383,
|
| 1701 |
+
"learning_rate": 3.4908816317405686e-05,
|
| 1702 |
+
"loss": 0.7009,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.48893360160965793,
|
| 1707 |
+
"grad_norm": 0.901075541973114,
|
| 1708 |
+
"learning_rate": 3.46996897570229e-05,
|
| 1709 |
+
"loss": 0.7044,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.4909456740442656,
|
| 1714 |
+
"grad_norm": 0.9331154227256775,
|
| 1715 |
+
"learning_rate": 3.4490471658570554e-05,
|
| 1716 |
+
"loss": 0.7393,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.49295774647887325,
|
| 1721 |
+
"grad_norm": 0.9193384647369385,
|
| 1722 |
+
"learning_rate": 3.428117072846772e-05,
|
| 1723 |
+
"loss": 0.6808,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.4949698189134809,
|
| 1728 |
+
"grad_norm": 0.9850835800170898,
|
| 1729 |
+
"learning_rate": 3.407179567658043e-05,
|
| 1730 |
+
"loss": 0.7291,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.4969818913480885,
|
| 1735 |
+
"grad_norm": 0.9162485003471375,
|
| 1736 |
+
"learning_rate": 3.3862355215859225e-05,
|
| 1737 |
+
"loss": 0.7291,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.49899396378269617,
|
| 1742 |
+
"grad_norm": 0.8797596096992493,
|
| 1743 |
+
"learning_rate": 3.3652858061976584e-05,
|
| 1744 |
+
"loss": 0.7091,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.5010060362173038,
|
| 1749 |
+
"grad_norm": 0.8453248739242554,
|
| 1750 |
+
"learning_rate": 3.3443312932964194e-05,
|
| 1751 |
+
"loss": 0.6448,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.5030181086519114,
|
| 1756 |
+
"grad_norm": 0.9353291392326355,
|
| 1757 |
+
"learning_rate": 3.323372854885023e-05,
|
| 1758 |
+
"loss": 0.7358,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.5050301810865191,
|
| 1763 |
+
"grad_norm": 0.8059361577033997,
|
| 1764 |
+
"learning_rate": 3.3024113631296376e-05,
|
| 1765 |
+
"loss": 0.6646,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.5070422535211268,
|
| 1770 |
+
"grad_norm": 1.006386160850525,
|
| 1771 |
+
"learning_rate": 3.281447690323499e-05,
|
| 1772 |
+
"loss": 0.709,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.5090543259557344,
|
| 1777 |
+
"grad_norm": 1.0037897825241089,
|
| 1778 |
+
"learning_rate": 3.260482708850601e-05,
|
| 1779 |
+
"loss": 0.7314,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.5110663983903421,
|
| 1784 |
+
"grad_norm": 0.857748806476593,
|
| 1785 |
+
"learning_rate": 3.239517291149399e-05,
|
| 1786 |
+
"loss": 0.6773,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.5130784708249497,
|
| 1791 |
+
"grad_norm": 0.8884151577949524,
|
| 1792 |
+
"learning_rate": 3.218552309676501e-05,
|
| 1793 |
+
"loss": 0.6952,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.5150905432595574,
|
| 1798 |
+
"grad_norm": 0.8516953587532043,
|
| 1799 |
+
"learning_rate": 3.197588636870362e-05,
|
| 1800 |
+
"loss": 0.6202,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.5171026156941649,
|
| 1805 |
+
"grad_norm": 0.8813678026199341,
|
| 1806 |
+
"learning_rate": 3.176627145114977e-05,
|
| 1807 |
+
"loss": 0.697,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.5191146881287726,
|
| 1812 |
+
"grad_norm": 0.8686804175376892,
|
| 1813 |
+
"learning_rate": 3.15566870670358e-05,
|
| 1814 |
+
"loss": 0.6814,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.5211267605633803,
|
| 1819 |
+
"grad_norm": 0.8625949025154114,
|
| 1820 |
+
"learning_rate": 3.134714193802342e-05,
|
| 1821 |
+
"loss": 0.6673,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.5231388329979879,
|
| 1826 |
+
"grad_norm": 0.8244426250457764,
|
| 1827 |
+
"learning_rate": 3.113764478414077e-05,
|
| 1828 |
+
"loss": 0.631,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.5251509054325956,
|
| 1833 |
+
"grad_norm": 0.8436176180839539,
|
| 1834 |
+
"learning_rate": 3.092820432341956e-05,
|
| 1835 |
+
"loss": 0.6734,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.5271629778672032,
|
| 1840 |
+
"grad_norm": 0.8252584934234619,
|
| 1841 |
+
"learning_rate": 3.0718829271532275e-05,
|
| 1842 |
+
"loss": 0.6562,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.5291750503018109,
|
| 1847 |
+
"grad_norm": 0.845025897026062,
|
| 1848 |
+
"learning_rate": 3.0509528341429443e-05,
|
| 1849 |
+
"loss": 0.7029,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.5311871227364185,
|
| 1854 |
+
"grad_norm": 0.8654035925865173,
|
| 1855 |
+
"learning_rate": 3.03003102429771e-05,
|
| 1856 |
+
"loss": 0.6654,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.5331991951710262,
|
| 1861 |
+
"grad_norm": 0.8877770304679871,
|
| 1862 |
+
"learning_rate": 3.009118368259431e-05,
|
| 1863 |
+
"loss": 0.6565,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.5352112676056338,
|
| 1868 |
+
"grad_norm": 0.8893894553184509,
|
| 1869 |
+
"learning_rate": 2.9882157362890886e-05,
|
| 1870 |
+
"loss": 0.6163,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.5372233400402414,
|
| 1875 |
+
"grad_norm": 0.9666234254837036,
|
| 1876 |
+
"learning_rate": 2.9673239982305202e-05,
|
| 1877 |
+
"loss": 0.6895,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.5392354124748491,
|
| 1882 |
+
"grad_norm": 0.9029472470283508,
|
| 1883 |
+
"learning_rate": 2.9464440234742216e-05,
|
| 1884 |
+
"loss": 0.6978,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.5412474849094567,
|
| 1889 |
+
"grad_norm": 0.9182049036026001,
|
| 1890 |
+
"learning_rate": 2.9255766809211712e-05,
|
| 1891 |
+
"loss": 0.6729,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.5432595573440644,
|
| 1896 |
+
"grad_norm": 0.9030729532241821,
|
| 1897 |
+
"learning_rate": 2.9047228389466707e-05,
|
| 1898 |
+
"loss": 0.7138,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.545271629778672,
|
| 1903 |
+
"grad_norm": 0.9141706228256226,
|
| 1904 |
+
"learning_rate": 2.8838833653642064e-05,
|
| 1905 |
+
"loss": 0.7274,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.5472837022132797,
|
| 1910 |
+
"grad_norm": 0.9512321949005127,
|
| 1911 |
+
"learning_rate": 2.8630591273893377e-05,
|
| 1912 |
+
"loss": 0.7714,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.5492957746478874,
|
| 1917 |
+
"grad_norm": 0.972109854221344,
|
| 1918 |
+
"learning_rate": 2.842250991603607e-05,
|
| 1919 |
+
"loss": 0.6968,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.5513078470824949,
|
| 1924 |
+
"grad_norm": 0.9172069430351257,
|
| 1925 |
+
"learning_rate": 2.8214598239184785e-05,
|
| 1926 |
+
"loss": 0.7382,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.5533199195171026,
|
| 1931 |
+
"grad_norm": 0.8818020820617676,
|
| 1932 |
+
"learning_rate": 2.800686489539308e-05,
|
| 1933 |
+
"loss": 0.6968,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.5553319919517102,
|
| 1938 |
+
"grad_norm": 0.8782933354377747,
|
| 1939 |
+
"learning_rate": 2.7799318529293317e-05,
|
| 1940 |
+
"loss": 0.662,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.5573440643863179,
|
| 1945 |
+
"grad_norm": 0.8515357375144958,
|
| 1946 |
+
"learning_rate": 2.7591967777736968e-05,
|
| 1947 |
+
"loss": 0.6968,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.5593561368209256,
|
| 1952 |
+
"grad_norm": 0.9061254858970642,
|
| 1953 |
+
"learning_rate": 2.7384821269435143e-05,
|
| 1954 |
+
"loss": 0.723,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.5613682092555332,
|
| 1959 |
+
"grad_norm": 0.90815269947052,
|
| 1960 |
+
"learning_rate": 2.7177887624599636e-05,
|
| 1961 |
+
"loss": 0.7091,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.5633802816901409,
|
| 1966 |
+
"grad_norm": 0.8476876616477966,
|
| 1967 |
+
"learning_rate": 2.6971175454584048e-05,
|
| 1968 |
+
"loss": 0.6615,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.5653923541247485,
|
| 1973 |
+
"grad_norm": 0.8543688058853149,
|
| 1974 |
+
"learning_rate": 2.676469336152555e-05,
|
| 1975 |
+
"loss": 0.665,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.5674044265593562,
|
| 1980 |
+
"grad_norm": 0.8241449594497681,
|
| 1981 |
+
"learning_rate": 2.6558449937986858e-05,
|
| 1982 |
+
"loss": 0.6665,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.5694164989939637,
|
| 1987 |
+
"grad_norm": 0.9415947794914246,
|
| 1988 |
+
"learning_rate": 2.6352453766598655e-05,
|
| 1989 |
+
"loss": 0.6802,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.5714285714285714,
|
| 1994 |
+
"grad_norm": 0.8769863843917847,
|
| 1995 |
+
"learning_rate": 2.614671341970252e-05,
|
| 1996 |
+
"loss": 0.7162,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.5734406438631791,
|
| 2001 |
+
"grad_norm": 0.8706234097480774,
|
| 2002 |
+
"learning_rate": 2.5941237458994048e-05,
|
| 2003 |
+
"loss": 0.6951,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.5754527162977867,
|
| 2008 |
+
"grad_norm": 0.807435929775238,
|
| 2009 |
+
"learning_rate": 2.5736034435166702e-05,
|
| 2010 |
+
"loss": 0.636,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.5774647887323944,
|
| 2015 |
+
"grad_norm": 0.8227114081382751,
|
| 2016 |
+
"learning_rate": 2.5531112887555894e-05,
|
| 2017 |
+
"loss": 0.6926,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.579476861167002,
|
| 2022 |
+
"grad_norm": 0.8422671556472778,
|
| 2023 |
+
"learning_rate": 2.532648134378369e-05,
|
| 2024 |
+
"loss": 0.672,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.5814889336016097,
|
| 2029 |
+
"grad_norm": 0.913819432258606,
|
| 2030 |
+
"learning_rate": 2.5122148319403903e-05,
|
| 2031 |
+
"loss": 0.7599,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.5835010060362174,
|
| 2036 |
+
"grad_norm": 0.848904013633728,
|
| 2037 |
+
"learning_rate": 2.491812231754771e-05,
|
| 2038 |
+
"loss": 0.7112,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.5855130784708249,
|
| 2043 |
+
"grad_norm": 0.857905924320221,
|
| 2044 |
+
"learning_rate": 2.4714411828569864e-05,
|
| 2045 |
+
"loss": 0.6655,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.5875251509054326,
|
| 2050 |
+
"grad_norm": 0.8949502110481262,
|
| 2051 |
+
"learning_rate": 2.451102532969531e-05,
|
| 2052 |
+
"loss": 0.7178,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.5895372233400402,
|
| 2057 |
+
"grad_norm": 0.8601780533790588,
|
| 2058 |
+
"learning_rate": 2.4307971284666478e-05,
|
| 2059 |
+
"loss": 0.6953,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.5915492957746479,
|
| 2064 |
+
"grad_norm": 0.8672630786895752,
|
| 2065 |
+
"learning_rate": 2.4105258143391005e-05,
|
| 2066 |
+
"loss": 0.6621,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.5935613682092555,
|
| 2071 |
+
"grad_norm": 0.8478452563285828,
|
| 2072 |
+
"learning_rate": 2.3902894341590137e-05,
|
| 2073 |
+
"loss": 0.6971,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.5955734406438632,
|
| 2078 |
+
"grad_norm": 0.8357344269752502,
|
| 2079 |
+
"learning_rate": 2.3700888300447675e-05,
|
| 2080 |
+
"loss": 0.6876,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.5975855130784709,
|
| 2085 |
+
"grad_norm": 0.8916183710098267,
|
| 2086 |
+
"learning_rate": 2.3499248426259564e-05,
|
| 2087 |
+
"loss": 0.6754,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.5995975855130785,
|
| 2092 |
+
"grad_norm": 0.9368688464164734,
|
| 2093 |
+
"learning_rate": 2.3297983110084027e-05,
|
| 2094 |
+
"loss": 0.757,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.6016096579476862,
|
| 2099 |
+
"grad_norm": 0.8134625554084778,
|
| 2100 |
+
"learning_rate": 2.3097100727392396e-05,
|
| 2101 |
+
"loss": 0.6631,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.6036217303822937,
|
| 2106 |
+
"grad_norm": 0.9809048771858215,
|
| 2107 |
+
"learning_rate": 2.2896609637720577e-05,
|
| 2108 |
+
"loss": 0.718,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.6056338028169014,
|
| 2113 |
+
"grad_norm": 0.8210386633872986,
|
| 2114 |
+
"learning_rate": 2.2696518184321198e-05,
|
| 2115 |
+
"loss": 0.6516,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.607645875251509,
|
| 2120 |
+
"grad_norm": 0.8537083268165588,
|
| 2121 |
+
"learning_rate": 2.2496834693816363e-05,
|
| 2122 |
+
"loss": 0.6541,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.6096579476861167,
|
| 2127 |
+
"grad_norm": 0.8941294550895691,
|
| 2128 |
+
"learning_rate": 2.2297567475851198e-05,
|
| 2129 |
+
"loss": 0.7086,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.6116700201207244,
|
| 2134 |
+
"grad_norm": 0.8543776869773865,
|
| 2135 |
+
"learning_rate": 2.2098724822748016e-05,
|
| 2136 |
+
"loss": 0.6528,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.613682092555332,
|
| 2141 |
+
"grad_norm": 0.9148063063621521,
|
| 2142 |
+
"learning_rate": 2.190031500916124e-05,
|
| 2143 |
+
"loss": 0.6844,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.6156941649899397,
|
| 2148 |
+
"grad_norm": 0.9270927309989929,
|
| 2149 |
+
"learning_rate": 2.1702346291733105e-05,
|
| 2150 |
+
"loss": 0.6898,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.6177062374245473,
|
| 2155 |
+
"grad_norm": 0.8864634037017822,
|
| 2156 |
+
"learning_rate": 2.1504826908750016e-05,
|
| 2157 |
+
"loss": 0.7,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.6197183098591549,
|
| 2162 |
+
"grad_norm": 0.8082477450370789,
|
| 2163 |
+
"learning_rate": 2.1307765079799747e-05,
|
| 2164 |
+
"loss": 0.6638,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.6217303822937625,
|
| 2169 |
+
"grad_norm": 0.9282078742980957,
|
| 2170 |
+
"learning_rate": 2.1111169005429353e-05,
|
| 2171 |
+
"loss": 0.7176,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.6237424547283702,
|
| 2176 |
+
"grad_norm": 0.8172498941421509,
|
| 2177 |
+
"learning_rate": 2.0915046866803992e-05,
|
| 2178 |
+
"loss": 0.6371,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.6257545271629779,
|
| 2183 |
+
"grad_norm": 0.939055860042572,
|
| 2184 |
+
"learning_rate": 2.0719406825366366e-05,
|
| 2185 |
+
"loss": 0.7575,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.6277665995975855,
|
| 2190 |
+
"grad_norm": 0.8269034028053284,
|
| 2191 |
+
"learning_rate": 2.0524257022497183e-05,
|
| 2192 |
+
"loss": 0.6912,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.6297786720321932,
|
| 2197 |
+
"grad_norm": 0.8468128442764282,
|
| 2198 |
+
"learning_rate": 2.0329605579176304e-05,
|
| 2199 |
+
"loss": 0.7046,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.6317907444668008,
|
| 2204 |
+
"grad_norm": 0.8607275485992432,
|
| 2205 |
+
"learning_rate": 2.0135460595644815e-05,
|
| 2206 |
+
"loss": 0.677,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.6338028169014085,
|
| 2211 |
+
"grad_norm": 0.830365777015686,
|
| 2212 |
+
"learning_rate": 1.9941830151067965e-05,
|
| 2213 |
+
"loss": 0.6652,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.635814889336016,
|
| 2218 |
+
"grad_norm": 0.9029973745346069,
|
| 2219 |
+
"learning_rate": 1.974872230319891e-05,
|
| 2220 |
+
"loss": 0.7056,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.6378269617706237,
|
| 2225 |
+
"grad_norm": 0.8448197245597839,
|
| 2226 |
+
"learning_rate": 1.9556145088043427e-05,
|
| 2227 |
+
"loss": 0.6836,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.6398390342052314,
|
| 2232 |
+
"grad_norm": 0.9112847447395325,
|
| 2233 |
+
"learning_rate": 1.936410651952551e-05,
|
| 2234 |
+
"loss": 0.7007,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.641851106639839,
|
| 2239 |
+
"grad_norm": 0.8929232358932495,
|
| 2240 |
+
"learning_rate": 1.9172614589153883e-05,
|
| 2241 |
+
"loss": 0.6927,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.6438631790744467,
|
| 2246 |
+
"grad_norm": 0.9945328831672668,
|
| 2247 |
+
"learning_rate": 1.8981677265689392e-05,
|
| 2248 |
+
"loss": 0.6732,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.6458752515090543,
|
| 2253 |
+
"grad_norm": 0.9466527700424194,
|
| 2254 |
+
"learning_rate": 1.8791302494813435e-05,
|
| 2255 |
+
"loss": 0.7199,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.647887323943662,
|
| 2260 |
+
"grad_norm": 0.8851666450500488,
|
| 2261 |
+
"learning_rate": 1.8601498198797338e-05,
|
| 2262 |
+
"loss": 0.6805,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.6498993963782697,
|
| 2267 |
+
"grad_norm": 0.8846597075462341,
|
| 2268 |
+
"learning_rate": 1.8412272276172555e-05,
|
| 2269 |
+
"loss": 0.6908,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.6519114688128773,
|
| 2274 |
+
"grad_norm": 0.9064490795135498,
|
| 2275 |
+
"learning_rate": 1.8223632601402155e-05,
|
| 2276 |
+
"loss": 0.6676,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.6539235412474849,
|
| 2281 |
+
"grad_norm": 0.7833627462387085,
|
| 2282 |
+
"learning_rate": 1.8035587024552997e-05,
|
| 2283 |
+
"loss": 0.6214,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.6559356136820925,
|
| 2288 |
+
"grad_norm": 0.8748745322227478,
|
| 2289 |
+
"learning_rate": 1.7848143370969098e-05,
|
| 2290 |
+
"loss": 0.6829,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.6579476861167002,
|
| 2295 |
+
"grad_norm": 0.8304067254066467,
|
| 2296 |
+
"learning_rate": 1.7661309440945997e-05,
|
| 2297 |
+
"loss": 0.6612,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.6599597585513078,
|
| 2302 |
+
"grad_norm": 0.9479273557662964,
|
| 2303 |
+
"learning_rate": 1.7475093009406154e-05,
|
| 2304 |
+
"loss": 0.7223,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.6619718309859155,
|
| 2309 |
+
"grad_norm": 0.8814991116523743,
|
| 2310 |
+
"learning_rate": 1.728950182557538e-05,
|
| 2311 |
+
"loss": 0.6951,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.6639839034205232,
|
| 2316 |
+
"grad_norm": 0.9388280510902405,
|
| 2317 |
+
"learning_rate": 1.7104543612660397e-05,
|
| 2318 |
+
"loss": 0.7415,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.6659959758551308,
|
| 2323 |
+
"grad_norm": 0.8191916942596436,
|
| 2324 |
+
"learning_rate": 1.692022606752741e-05,
|
| 2325 |
+
"loss": 0.6254,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.6680080482897385,
|
| 2330 |
+
"grad_norm": 0.9048965573310852,
|
| 2331 |
+
"learning_rate": 1.673655686038183e-05,
|
| 2332 |
+
"loss": 0.7146,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.670020120724346,
|
| 2337 |
+
"grad_norm": 0.9126314520835876,
|
| 2338 |
+
"learning_rate": 1.6553543634449058e-05,
|
| 2339 |
+
"loss": 0.7106,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.6720321931589537,
|
| 2344 |
+
"grad_norm": 0.9011045098304749,
|
| 2345 |
+
"learning_rate": 1.637119400565647e-05,
|
| 2346 |
+
"loss": 0.7553,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.6740442655935613,
|
| 2351 |
+
"grad_norm": 0.9251480102539062,
|
| 2352 |
+
"learning_rate": 1.618951556231644e-05,
|
| 2353 |
+
"loss": 0.6894,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.676056338028169,
|
| 2358 |
+
"grad_norm": 0.8758312463760376,
|
| 2359 |
+
"learning_rate": 1.600851586481057e-05,
|
| 2360 |
+
"loss": 0.6746,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.6780684104627767,
|
| 2365 |
+
"grad_norm": 0.8351629972457886,
|
| 2366 |
+
"learning_rate": 1.5828202445275094e-05,
|
| 2367 |
+
"loss": 0.634,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.6800804828973843,
|
| 2372 |
+
"grad_norm": 0.847764253616333,
|
| 2373 |
+
"learning_rate": 1.5648582807287395e-05,
|
| 2374 |
+
"loss": 0.64,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.682092555331992,
|
| 2379 |
+
"grad_norm": 0.8462485074996948,
|
| 2380 |
+
"learning_rate": 1.54696644255538e-05,
|
| 2381 |
+
"loss": 0.6642,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.6841046277665996,
|
| 2386 |
+
"grad_norm": 0.8721317052841187,
|
| 2387 |
+
"learning_rate": 1.529145474559848e-05,
|
| 2388 |
+
"loss": 0.7009,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.6861167002012073,
|
| 2393 |
+
"grad_norm": 0.8421750068664551,
|
| 2394 |
+
"learning_rate": 1.5113961183453648e-05,
|
| 2395 |
+
"loss": 0.6638,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.6881287726358148,
|
| 2400 |
+
"grad_norm": 0.9829575419425964,
|
| 2401 |
+
"learning_rate": 1.4937191125350919e-05,
|
| 2402 |
+
"loss": 0.7874,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.6901408450704225,
|
| 2407 |
+
"grad_norm": 0.8514189720153809,
|
| 2408 |
+
"learning_rate": 1.4761151927413958e-05,
|
| 2409 |
+
"loss": 0.6728,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.6921529175050302,
|
| 2414 |
+
"grad_norm": 0.8339893817901611,
|
| 2415 |
+
"learning_rate": 1.4585850915352355e-05,
|
| 2416 |
+
"loss": 0.6345,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.6941649899396378,
|
| 2421 |
+
"grad_norm": 0.8384912014007568,
|
| 2422 |
+
"learning_rate": 1.4411295384156769e-05,
|
| 2423 |
+
"loss": 0.642,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.6961770623742455,
|
| 2428 |
+
"grad_norm": 0.8968607187271118,
|
| 2429 |
+
"learning_rate": 1.4237492597795365e-05,
|
| 2430 |
+
"loss": 0.7056,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.6981891348088531,
|
| 2435 |
+
"grad_norm": 0.8969266414642334,
|
| 2436 |
+
"learning_rate": 1.4064449788911513e-05,
|
| 2437 |
+
"loss": 0.6912,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.7002012072434608,
|
| 2442 |
+
"grad_norm": 0.9129147529602051,
|
| 2443 |
+
"learning_rate": 1.3892174158522827e-05,
|
| 2444 |
+
"loss": 0.757,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.7022132796780685,
|
| 2449 |
+
"grad_norm": 0.8593868017196655,
|
| 2450 |
+
"learning_rate": 1.3720672875721471e-05,
|
| 2451 |
+
"loss": 0.6597,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.704225352112676,
|
| 2456 |
+
"grad_norm": 0.8287497758865356,
|
| 2457 |
+
"learning_rate": 1.3549953077375903e-05,
|
| 2458 |
+
"loss": 0.6758,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.7062374245472837,
|
| 2463 |
+
"grad_norm": 0.8732616901397705,
|
| 2464 |
+
"learning_rate": 1.3380021867833747e-05,
|
| 2465 |
+
"loss": 0.7015,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.7082494969818913,
|
| 2470 |
+
"grad_norm": 0.8855911493301392,
|
| 2471 |
+
"learning_rate": 1.3210886318626269e-05,
|
| 2472 |
+
"loss": 0.69,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.710261569416499,
|
| 2477 |
+
"grad_norm": 0.8534879088401794,
|
| 2478 |
+
"learning_rate": 1.304255346817406e-05,
|
| 2479 |
+
"loss": 0.6603,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.7122736418511066,
|
| 2484 |
+
"grad_norm": 0.9096892476081848,
|
| 2485 |
+
"learning_rate": 1.2875030321494125e-05,
|
| 2486 |
+
"loss": 0.7196,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.7142857142857143,
|
| 2491 |
+
"grad_norm": 0.8905626535415649,
|
| 2492 |
+
"learning_rate": 1.2708323849908427e-05,
|
| 2493 |
+
"loss": 0.7093,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.716297786720322,
|
| 2498 |
+
"grad_norm": 0.8651593327522278,
|
| 2499 |
+
"learning_rate": 1.2542440990753688e-05,
|
| 2500 |
+
"loss": 0.6445,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.7183098591549296,
|
| 2505 |
+
"grad_norm": 0.8822929263114929,
|
| 2506 |
+
"learning_rate": 1.2377388647092794e-05,
|
| 2507 |
+
"loss": 0.7181,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.7203219315895373,
|
| 2512 |
+
"grad_norm": 0.8250637054443359,
|
| 2513 |
+
"learning_rate": 1.2213173687427467e-05,
|
| 2514 |
+
"loss": 0.6881,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.7223340040241448,
|
| 2519 |
+
"grad_norm": 0.7754917144775391,
|
| 2520 |
+
"learning_rate": 1.2049802945412505e-05,
|
| 2521 |
+
"loss": 0.6295,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.7243460764587525,
|
| 2526 |
+
"grad_norm": 0.9656423926353455,
|
| 2527 |
+
"learning_rate": 1.188728321957133e-05,
|
| 2528 |
+
"loss": 0.7574,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.7263581488933601,
|
| 2533 |
+
"grad_norm": 0.8139829039573669,
|
| 2534 |
+
"learning_rate": 1.1725621273013094e-05,
|
| 2535 |
+
"loss": 0.6644,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.7283702213279678,
|
| 2540 |
+
"grad_norm": 0.9022115468978882,
|
| 2541 |
+
"learning_rate": 1.1564823833151273e-05,
|
| 2542 |
+
"loss": 0.6871,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.7303822937625755,
|
| 2547 |
+
"grad_norm": 0.9244035482406616,
|
| 2548 |
+
"learning_rate": 1.1404897591423706e-05,
|
| 2549 |
+
"loss": 0.7392,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.7323943661971831,
|
| 2554 |
+
"grad_norm": 0.827328622341156,
|
| 2555 |
+
"learning_rate": 1.1245849203014094e-05,
|
| 2556 |
+
"loss": 0.6342,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.7344064386317908,
|
| 2561 |
+
"grad_norm": 0.8412946462631226,
|
| 2562 |
+
"learning_rate": 1.1087685286575101e-05,
|
| 2563 |
+
"loss": 0.656,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.7364185110663984,
|
| 2568 |
+
"grad_norm": 0.8751910328865051,
|
| 2569 |
+
"learning_rate": 1.0930412423952864e-05,
|
| 2570 |
+
"loss": 0.6828,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.738430583501006,
|
| 2575 |
+
"grad_norm": 0.8002843260765076,
|
| 2576 |
+
"learning_rate": 1.0774037159913161e-05,
|
| 2577 |
+
"loss": 0.6336,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.7404426559356136,
|
| 2582 |
+
"grad_norm": 0.8765881061553955,
|
| 2583 |
+
"learning_rate": 1.0618566001869048e-05,
|
| 2584 |
+
"loss": 0.693,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.7424547283702213,
|
| 2589 |
+
"grad_norm": 0.8102418780326843,
|
| 2590 |
+
"learning_rate": 1.0464005419610007e-05,
|
| 2591 |
+
"loss": 0.632,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.744466800804829,
|
| 2596 |
+
"grad_norm": 0.842564582824707,
|
| 2597 |
+
"learning_rate": 1.031036184503277e-05,
|
| 2598 |
+
"loss": 0.6622,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.7464788732394366,
|
| 2603 |
+
"grad_norm": 0.8570309281349182,
|
| 2604 |
+
"learning_rate": 1.0157641671873589e-05,
|
| 2605 |
+
"loss": 0.656,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.7484909456740443,
|
| 2610 |
+
"grad_norm": 0.832196831703186,
|
| 2611 |
+
"learning_rate": 1.0005851255442282e-05,
|
| 2612 |
+
"loss": 0.6575,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.7505030181086519,
|
| 2617 |
+
"grad_norm": 0.8045952320098877,
|
| 2618 |
+
"learning_rate": 9.854996912357643e-06,
|
| 2619 |
+
"loss": 0.6114,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.7525150905432596,
|
| 2624 |
+
"grad_norm": 0.9697309136390686,
|
| 2625 |
+
"learning_rate": 9.705084920284659e-06,
|
| 2626 |
+
"loss": 0.7239,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.7545271629778671,
|
| 2631 |
+
"grad_norm": 0.9199627637863159,
|
| 2632 |
+
"learning_rate": 9.556121517673238e-06,
|
| 2633 |
+
"loss": 0.7238,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.7565392354124748,
|
| 2638 |
+
"grad_norm": 0.841739296913147,
|
| 2639 |
+
"learning_rate": 9.408112903498584e-06,
|
| 2640 |
+
"loss": 0.6588,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.7585513078470825,
|
| 2645 |
+
"grad_norm": 0.8698973059654236,
|
| 2646 |
+
"learning_rate": 9.261065237003308e-06,
|
| 2647 |
+
"loss": 0.6793,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.7605633802816901,
|
| 2652 |
+
"grad_norm": 0.9169113636016846,
|
| 2653 |
+
"learning_rate": 9.11498463744102e-06,
|
| 2654 |
+
"loss": 0.681,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.7625754527162978,
|
| 2659 |
+
"grad_norm": 0.8953713774681091,
|
| 2660 |
+
"learning_rate": 8.969877183821753e-06,
|
| 2661 |
+
"loss": 0.7116,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.7645875251509054,
|
| 2666 |
+
"grad_norm": 0.8802800178527832,
|
| 2667 |
+
"learning_rate": 8.825748914658955e-06,
|
| 2668 |
+
"loss": 0.7088,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.7665995975855131,
|
| 2673 |
+
"grad_norm": 0.8699951171875,
|
| 2674 |
+
"learning_rate": 8.682605827718215e-06,
|
| 2675 |
+
"loss": 0.6427,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.7686116700201208,
|
| 2680 |
+
"grad_norm": 0.9355603456497192,
|
| 2681 |
+
"learning_rate": 8.540453879767662e-06,
|
| 2682 |
+
"loss": 0.7048,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.7706237424547284,
|
| 2687 |
+
"grad_norm": 0.8565495014190674,
|
| 2688 |
+
"learning_rate": 8.399298986330093e-06,
|
| 2689 |
+
"loss": 0.6451,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.772635814889336,
|
| 2694 |
+
"grad_norm": 0.8059428930282593,
|
| 2695 |
+
"learning_rate": 8.259147021436798e-06,
|
| 2696 |
+
"loss": 0.6392,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.7746478873239436,
|
| 2701 |
+
"grad_norm": 0.8618172407150269,
|
| 2702 |
+
"learning_rate": 8.120003817383111e-06,
|
| 2703 |
+
"loss": 0.6967,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.7766599597585513,
|
| 2708 |
+
"grad_norm": 0.8753212690353394,
|
| 2709 |
+
"learning_rate": 7.981875164485717e-06,
|
| 2710 |
+
"loss": 0.6774,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.778672032193159,
|
| 2715 |
+
"grad_norm": 0.8684304356575012,
|
| 2716 |
+
"learning_rate": 7.844766810841674e-06,
|
| 2717 |
+
"loss": 0.6508,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.7806841046277666,
|
| 2722 |
+
"grad_norm": 0.916515052318573,
|
| 2723 |
+
"learning_rate": 7.708684462089232e-06,
|
| 2724 |
+
"loss": 0.7209,
|
| 2725 |
+
"step": 388
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.7826961770623743,
|
| 2729 |
+
"grad_norm": 0.8353291153907776,
|
| 2730 |
+
"learning_rate": 7.573633781170395e-06,
|
| 2731 |
+
"loss": 0.6533,
|
| 2732 |
+
"step": 389
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.7847082494969819,
|
| 2736 |
+
"grad_norm": 0.8614966869354248,
|
| 2737 |
+
"learning_rate": 7.439620388095243e-06,
|
| 2738 |
+
"loss": 0.6952,
|
| 2739 |
+
"step": 390
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.7867203219315896,
|
| 2743 |
+
"grad_norm": 0.8962644934654236,
|
| 2744 |
+
"learning_rate": 7.3066498597080875e-06,
|
| 2745 |
+
"loss": 0.6624,
|
| 2746 |
+
"step": 391
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.7887323943661971,
|
| 2750 |
+
"grad_norm": 0.8484448194503784,
|
| 2751 |
+
"learning_rate": 7.174727729455374e-06,
|
| 2752 |
+
"loss": 0.6643,
|
| 2753 |
+
"step": 392
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.7907444668008048,
|
| 2757 |
+
"grad_norm": 0.802003800868988,
|
| 2758 |
+
"learning_rate": 7.043859487155433e-06,
|
| 2759 |
+
"loss": 0.6685,
|
| 2760 |
+
"step": 393
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.7927565392354124,
|
| 2764 |
+
"grad_norm": 0.9428711533546448,
|
| 2765 |
+
"learning_rate": 6.914050578770002e-06,
|
| 2766 |
+
"loss": 0.6903,
|
| 2767 |
+
"step": 394
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.7947686116700201,
|
| 2771 |
+
"grad_norm": 0.8802878260612488,
|
| 2772 |
+
"learning_rate": 6.785306406177622e-06,
|
| 2773 |
+
"loss": 0.6904,
|
| 2774 |
+
"step": 395
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.7967806841046278,
|
| 2778 |
+
"grad_norm": 0.8296419978141785,
|
| 2779 |
+
"learning_rate": 6.657632326948827e-06,
|
| 2780 |
+
"loss": 0.6706,
|
| 2781 |
+
"step": 396
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.7987927565392354,
|
| 2785 |
+
"grad_norm": 0.8628143072128296,
|
| 2786 |
+
"learning_rate": 6.5310336541232e-06,
|
| 2787 |
+
"loss": 0.6381,
|
| 2788 |
+
"step": 397
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.8008048289738431,
|
| 2792 |
+
"grad_norm": 0.7924053072929382,
|
| 2793 |
+
"learning_rate": 6.405515655988267e-06,
|
| 2794 |
+
"loss": 0.6057,
|
| 2795 |
+
"step": 398
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.8028169014084507,
|
| 2799 |
+
"grad_norm": 0.8833969831466675,
|
| 2800 |
+
"learning_rate": 6.2810835558602785e-06,
|
| 2801 |
+
"loss": 0.6899,
|
| 2802 |
+
"step": 399
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.8048289738430584,
|
| 2806 |
+
"grad_norm": 0.8720591068267822,
|
| 2807 |
+
"learning_rate": 6.157742531866827e-06,
|
| 2808 |
+
"loss": 0.6513,
|
| 2809 |
+
"step": 400
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.806841046277666,
|
| 2813 |
+
"grad_norm": 0.8727748990058899,
|
| 2814 |
+
"learning_rate": 6.035497716731385e-06,
|
| 2815 |
+
"loss": 0.6646,
|
| 2816 |
+
"step": 401
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.8088531187122736,
|
| 2820 |
+
"grad_norm": 0.8175671696662903,
|
| 2821 |
+
"learning_rate": 5.9143541975596815e-06,
|
| 2822 |
+
"loss": 0.6625,
|
| 2823 |
+
"step": 402
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.8108651911468813,
|
| 2827 |
+
"grad_norm": 0.8272060751914978,
|
| 2828 |
+
"learning_rate": 5.794317015628064e-06,
|
| 2829 |
+
"loss": 0.6658,
|
| 2830 |
+
"step": 403
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.8128772635814889,
|
| 2834 |
+
"grad_norm": 0.8657524585723877,
|
| 2835 |
+
"learning_rate": 5.6753911661736174e-06,
|
| 2836 |
+
"loss": 0.6762,
|
| 2837 |
+
"step": 404
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.8148893360160966,
|
| 2841 |
+
"grad_norm": 0.8350597023963928,
|
| 2842 |
+
"learning_rate": 5.557581598186384e-06,
|
| 2843 |
+
"loss": 0.6557,
|
| 2844 |
+
"step": 405
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.8169014084507042,
|
| 2848 |
+
"grad_norm": 0.8281211256980896,
|
| 2849 |
+
"learning_rate": 5.440893214203366e-06,
|
| 2850 |
+
"loss": 0.6741,
|
| 2851 |
+
"step": 406
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.8189134808853119,
|
| 2855 |
+
"grad_norm": 0.7790670990943909,
|
| 2856 |
+
"learning_rate": 5.3253308701045154e-06,
|
| 2857 |
+
"loss": 0.6132,
|
| 2858 |
+
"step": 407
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.8209255533199196,
|
| 2862 |
+
"grad_norm": 0.9283133149147034,
|
| 2863 |
+
"learning_rate": 5.210899374910702e-06,
|
| 2864 |
+
"loss": 0.7217,
|
| 2865 |
+
"step": 408
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.8229376257545271,
|
| 2869 |
+
"grad_norm": 0.8975178599357605,
|
| 2870 |
+
"learning_rate": 5.097603490583504e-06,
|
| 2871 |
+
"loss": 0.7183,
|
| 2872 |
+
"step": 409
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.8249496981891348,
|
| 2876 |
+
"grad_norm": 0.9391026496887207,
|
| 2877 |
+
"learning_rate": 4.985447931827132e-06,
|
| 2878 |
+
"loss": 0.7487,
|
| 2879 |
+
"step": 410
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.8269617706237424,
|
| 2883 |
+
"grad_norm": 0.837672770023346,
|
| 2884 |
+
"learning_rate": 4.874437365892185e-06,
|
| 2885 |
+
"loss": 0.6717,
|
| 2886 |
+
"step": 411
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.8289738430583501,
|
| 2890 |
+
"grad_norm": 0.8382728695869446,
|
| 2891 |
+
"learning_rate": 4.7645764123814465e-06,
|
| 2892 |
+
"loss": 0.7032,
|
| 2893 |
+
"step": 412
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.8309859154929577,
|
| 2897 |
+
"grad_norm": 0.854611337184906,
|
| 2898 |
+
"learning_rate": 4.655869643057628e-06,
|
| 2899 |
+
"loss": 0.6821,
|
| 2900 |
+
"step": 413
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.8329979879275654,
|
| 2904 |
+
"grad_norm": 0.8104320764541626,
|
| 2905 |
+
"learning_rate": 4.548321581653112e-06,
|
| 2906 |
+
"loss": 0.6471,
|
| 2907 |
+
"step": 414
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.8350100603621731,
|
| 2911 |
+
"grad_norm": 0.8787878155708313,
|
| 2912 |
+
"learning_rate": 4.441936703681727e-06,
|
| 2913 |
+
"loss": 0.6592,
|
| 2914 |
+
"step": 415
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.8370221327967807,
|
| 2918 |
+
"grad_norm": 0.8627755641937256,
|
| 2919 |
+
"learning_rate": 4.336719436252494e-06,
|
| 2920 |
+
"loss": 0.6862,
|
| 2921 |
+
"step": 416
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.8390342052313883,
|
| 2925 |
+
"grad_norm": 0.8459790349006653,
|
| 2926 |
+
"learning_rate": 4.232674157885398e-06,
|
| 2927 |
+
"loss": 0.6636,
|
| 2928 |
+
"step": 417
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.8410462776659959,
|
| 2932 |
+
"grad_norm": 0.9139977097511292,
|
| 2933 |
+
"learning_rate": 4.129805198329172e-06,
|
| 2934 |
+
"loss": 0.7162,
|
| 2935 |
+
"step": 418
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.8430583501006036,
|
| 2939 |
+
"grad_norm": 0.824047327041626,
|
| 2940 |
+
"learning_rate": 4.028116838381097e-06,
|
| 2941 |
+
"loss": 0.6488,
|
| 2942 |
+
"step": 419
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.8450704225352113,
|
| 2946 |
+
"grad_norm": 0.8267231583595276,
|
| 2947 |
+
"learning_rate": 3.9276133097089126e-06,
|
| 2948 |
+
"loss": 0.6324,
|
| 2949 |
+
"step": 420
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.8470824949698189,
|
| 2953 |
+
"grad_norm": 0.8386700749397278,
|
| 2954 |
+
"learning_rate": 3.828298794674692e-06,
|
| 2955 |
+
"loss": 0.6702,
|
| 2956 |
+
"step": 421
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.8490945674044266,
|
| 2960 |
+
"grad_norm": 0.9528822302818298,
|
| 2961 |
+
"learning_rate": 3.730177426160793e-06,
|
| 2962 |
+
"loss": 0.6559,
|
| 2963 |
+
"step": 422
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.8511066398390342,
|
| 2967 |
+
"grad_norm": 0.8409848809242249,
|
| 2968 |
+
"learning_rate": 3.633253287397875e-06,
|
| 2969 |
+
"loss": 0.6607,
|
| 2970 |
+
"step": 423
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.8531187122736419,
|
| 2974 |
+
"grad_norm": 0.8412972688674927,
|
| 2975 |
+
"learning_rate": 3.5375304117949623e-06,
|
| 2976 |
+
"loss": 0.6647,
|
| 2977 |
+
"step": 424
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.8551307847082495,
|
| 2981 |
+
"grad_norm": 0.8134218454360962,
|
| 2982 |
+
"learning_rate": 3.443012782771652e-06,
|
| 2983 |
+
"loss": 0.6433,
|
| 2984 |
+
"step": 425
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.8571428571428571,
|
| 2988 |
+
"grad_norm": 0.7878421545028687,
|
| 2989 |
+
"learning_rate": 3.3497043335922893e-06,
|
| 2990 |
+
"loss": 0.6367,
|
| 2991 |
+
"step": 426
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.8591549295774648,
|
| 2995 |
+
"grad_norm": 0.8618017435073853,
|
| 2996 |
+
"learning_rate": 3.257608947202313e-06,
|
| 2997 |
+
"loss": 0.6408,
|
| 2998 |
+
"step": 427
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.8611670020120724,
|
| 3002 |
+
"grad_norm": 0.8096737265586853,
|
| 3003 |
+
"learning_rate": 3.166730456066674e-06,
|
| 3004 |
+
"loss": 0.6534,
|
| 3005 |
+
"step": 428
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.8631790744466801,
|
| 3009 |
+
"grad_norm": 0.8893185257911682,
|
| 3010 |
+
"learning_rate": 3.077072642010324e-06,
|
| 3011 |
+
"loss": 0.6769,
|
| 3012 |
+
"step": 429
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.8651911468812877,
|
| 3016 |
+
"grad_norm": 0.8753596544265747,
|
| 3017 |
+
"learning_rate": 2.9886392360608895e-06,
|
| 3018 |
+
"loss": 0.6491,
|
| 3019 |
+
"step": 430
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.8672032193158954,
|
| 3023 |
+
"grad_norm": 0.807885468006134,
|
| 3024 |
+
"learning_rate": 2.9014339182933596e-06,
|
| 3025 |
+
"loss": 0.6391,
|
| 3026 |
+
"step": 431
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.869215291750503,
|
| 3030 |
+
"grad_norm": 0.9466260075569153,
|
| 3031 |
+
"learning_rate": 2.8154603176769635e-06,
|
| 3032 |
+
"loss": 0.7129,
|
| 3033 |
+
"step": 432
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.8712273641851107,
|
| 3037 |
+
"grad_norm": 0.8722304701805115,
|
| 3038 |
+
"learning_rate": 2.7307220119241518e-06,
|
| 3039 |
+
"loss": 0.7236,
|
| 3040 |
+
"step": 433
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.8732394366197183,
|
| 3044 |
+
"grad_norm": 0.8718979954719543,
|
| 3045 |
+
"learning_rate": 2.647222527341713e-06,
|
| 3046 |
+
"loss": 0.6502,
|
| 3047 |
+
"step": 434
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.8752515090543259,
|
| 3051 |
+
"grad_norm": 0.8858015537261963,
|
| 3052 |
+
"learning_rate": 2.564965338684033e-06,
|
| 3053 |
+
"loss": 0.6764,
|
| 3054 |
+
"step": 435
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 0.8772635814889336,
|
| 3058 |
+
"grad_norm": 0.8998034596443176,
|
| 3059 |
+
"learning_rate": 2.4839538690084976e-06,
|
| 3060 |
+
"loss": 0.6673,
|
| 3061 |
+
"step": 436
|
| 3062 |
+
},
|
| 3063 |
+
{
|
| 3064 |
+
"epoch": 0.8792756539235412,
|
| 3065 |
+
"grad_norm": 0.8517904281616211,
|
| 3066 |
+
"learning_rate": 2.404191489533028e-06,
|
| 3067 |
+
"loss": 0.6999,
|
| 3068 |
+
"step": 437
|
| 3069 |
+
},
|
| 3070 |
+
{
|
| 3071 |
+
"epoch": 0.8812877263581489,
|
| 3072 |
+
"grad_norm": 0.8171457648277283,
|
| 3073 |
+
"learning_rate": 2.3256815194958124e-06,
|
| 3074 |
+
"loss": 0.6525,
|
| 3075 |
+
"step": 438
|
| 3076 |
+
},
|
| 3077 |
+
{
|
| 3078 |
+
"epoch": 0.8832997987927566,
|
| 3079 |
+
"grad_norm": 0.8563843369483948,
|
| 3080 |
+
"learning_rate": 2.2484272260171726e-06,
|
| 3081 |
+
"loss": 0.6553,
|
| 3082 |
+
"step": 439
|
| 3083 |
+
},
|
| 3084 |
+
{
|
| 3085 |
+
"epoch": 0.8853118712273642,
|
| 3086 |
+
"grad_norm": 0.8232471942901611,
|
| 3087 |
+
"learning_rate": 2.1724318239635983e-06,
|
| 3088 |
+
"loss": 0.6453,
|
| 3089 |
+
"step": 440
|
| 3090 |
+
},
|
| 3091 |
+
{
|
| 3092 |
+
"epoch": 0.8873239436619719,
|
| 3093 |
+
"grad_norm": 0.8499601483345032,
|
| 3094 |
+
"learning_rate": 2.0976984758139677e-06,
|
| 3095 |
+
"loss": 0.6392,
|
| 3096 |
+
"step": 441
|
| 3097 |
+
},
|
| 3098 |
+
{
|
| 3099 |
+
"epoch": 0.8893360160965795,
|
| 3100 |
+
"grad_norm": 0.8528664708137512,
|
| 3101 |
+
"learning_rate": 2.0242302915279466e-06,
|
| 3102 |
+
"loss": 0.6339,
|
| 3103 |
+
"step": 442
|
| 3104 |
+
},
|
| 3105 |
+
{
|
| 3106 |
+
"epoch": 0.8913480885311871,
|
| 3107 |
+
"grad_norm": 0.8776962757110596,
|
| 3108 |
+
"learning_rate": 1.952030328416571e-06,
|
| 3109 |
+
"loss": 0.7032,
|
| 3110 |
+
"step": 443
|
| 3111 |
+
},
|
| 3112 |
+
{
|
| 3113 |
+
"epoch": 0.8933601609657947,
|
| 3114 |
+
"grad_norm": 0.8249021172523499,
|
| 3115 |
+
"learning_rate": 1.8811015910150092e-06,
|
| 3116 |
+
"loss": 0.6529,
|
| 3117 |
+
"step": 444
|
| 3118 |
+
},
|
| 3119 |
+
{
|
| 3120 |
+
"epoch": 0.8953722334004024,
|
| 3121 |
+
"grad_norm": 0.8518944382667542,
|
| 3122 |
+
"learning_rate": 1.8114470309575507e-06,
|
| 3123 |
+
"loss": 0.6333,
|
| 3124 |
+
"step": 445
|
| 3125 |
+
},
|
| 3126 |
+
{
|
| 3127 |
+
"epoch": 0.89738430583501,
|
| 3128 |
+
"grad_norm": 0.8814213871955872,
|
| 3129 |
+
"learning_rate": 1.7430695468547504e-06,
|
| 3130 |
+
"loss": 0.7178,
|
| 3131 |
+
"step": 446
|
| 3132 |
+
},
|
| 3133 |
+
{
|
| 3134 |
+
"epoch": 0.8993963782696177,
|
| 3135 |
+
"grad_norm": 0.8187962770462036,
|
| 3136 |
+
"learning_rate": 1.675971984172836e-06,
|
| 3137 |
+
"loss": 0.6341,
|
| 3138 |
+
"step": 447
|
| 3139 |
+
},
|
| 3140 |
+
{
|
| 3141 |
+
"epoch": 0.9014084507042254,
|
| 3142 |
+
"grad_norm": 0.8370924592018127,
|
| 3143 |
+
"learning_rate": 1.6101571351152637e-06,
|
| 3144 |
+
"loss": 0.6686,
|
| 3145 |
+
"step": 448
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 0.903420523138833,
|
| 3149 |
+
"grad_norm": 0.8962713479995728,
|
| 3150 |
+
"learning_rate": 1.5456277385065525e-06,
|
| 3151 |
+
"loss": 0.7155,
|
| 3152 |
+
"step": 449
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 0.9054325955734407,
|
| 3156 |
+
"grad_norm": 0.8398902416229248,
|
| 3157 |
+
"learning_rate": 1.4823864796782953e-06,
|
| 3158 |
+
"loss": 0.6699,
|
| 3159 |
+
"step": 450
|
| 3160 |
+
}
|
| 3161 |
+
],
|
| 3162 |
+
"logging_steps": 1,
|
| 3163 |
+
"max_steps": 497,
|
| 3164 |
+
"num_input_tokens_seen": 0,
|
| 3165 |
+
"num_train_epochs": 1,
|
| 3166 |
+
"save_steps": 50,
|
| 3167 |
+
"stateful_callbacks": {
|
| 3168 |
+
"TrainerControl": {
|
| 3169 |
+
"args": {
|
| 3170 |
+
"should_epoch_stop": false,
|
| 3171 |
+
"should_evaluate": false,
|
| 3172 |
+
"should_log": false,
|
| 3173 |
+
"should_save": true,
|
| 3174 |
+
"should_training_stop": false
|
| 3175 |
+
},
|
| 3176 |
+
"attributes": {}
|
| 3177 |
+
}
|
| 3178 |
+
},
|
| 3179 |
+
"total_flos": 2.061914791993344e+18,
|
| 3180 |
+
"train_batch_size": 15,
|
| 3181 |
+
"trial_name": null,
|
| 3182 |
+
"trial_params": null
|
| 3183 |
+
}
|
checkpoint-450/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05ae03ee70ba5876b3caf4bbc7280e0ee8c8b2ace354431732afbb1143dd28b5
|
| 3 |
+
size 5112
|