yangwang825 commited on
Commit
e5dae4e
·
verified ·
1 Parent(s): 688891e

Training in progress, epoch 1

Browse files
README.md CHANGED
@@ -1,199 +1,86 @@
1
  ---
2
  library_name: transformers
3
- tags: []
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
- This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
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]
 
1
  ---
2
  library_name: transformers
3
+ tags:
4
+ - audio-classification
5
+ - generated_from_trainer
6
+ datasets:
7
+ - voxceleb
8
+ metrics:
9
+ - accuracy
10
+ model-index:
11
+ - name: ce-len3-bs256-lr1e-3
12
+ results:
13
+ - task:
14
+ name: Audio Classification
15
+ type: audio-classification
16
+ dataset:
17
+ name: confit/voxceleb
18
+ type: voxceleb
19
+ config: verification
20
+ split: train
21
+ args: verification
22
+ metrics:
23
+ - name: Accuracy
24
+ type: accuracy
25
+ value: 0.9410023545240498
26
  ---
27
 
28
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
29
+ should probably proofread and complete it, then remove this comment. -->
30
 
31
+ # ce-len3-bs256-lr1e-3
32
 
33
+ This model is a fine-tuned version of [](https://huggingface.co/) on the confit/voxceleb dataset.
34
+ It achieves the following results on the evaluation set:
35
+ - Loss: 0.2946
36
+ - Accuracy: 0.9410
37
 
38
+ ## Model description
39
 
40
+ More information needed
41
 
42
+ ## Intended uses & limitations
43
 
44
+ More information needed
45
 
46
+ ## Training and evaluation data
47
 
48
+ More information needed
 
 
 
 
 
 
49
 
50
+ ## Training procedure
51
 
52
+ ### Training hyperparameters
53
 
54
+ The following hyperparameters were used during training:
55
+ - learning_rate: 0.001
56
+ - train_batch_size: 256
57
+ - eval_batch_size: 1
58
+ - seed: 914
59
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
60
+ - lr_scheduler_type: linear
61
+ - lr_scheduler_warmup_ratio: 0.1
62
+ - num_epochs: 10.0
63
+ - mixed_precision_training: Native AMP
64
 
65
+ ### Training results
66
 
67
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
68
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
69
+ | 4.6728 | 1.0 | 523 | 4.3456 | 0.1504 |
70
+ | 3.224 | 2.0 | 1046 | 2.2589 | 0.5141 |
71
+ | 2.3964 | 3.0 | 1569 | 1.4663 | 0.6836 |
72
+ | 1.8474 | 4.0 | 2092 | 0.9548 | 0.7927 |
73
+ | 1.5275 | 5.0 | 2615 | 0.6698 | 0.8571 |
74
+ | 1.248 | 6.0 | 3138 | 0.5270 | 0.8899 |
75
+ | 1.0991 | 7.0 | 3661 | 0.4500 | 0.9037 |
76
+ | 0.9221 | 8.0 | 4184 | 0.3572 | 0.9267 |
77
+ | 0.7997 | 9.0 | 4707 | 0.3138 | 0.9353 |
78
+ | 0.7603 | 10.0 | 5230 | 0.2946 | 0.9410 |
79
 
 
80
 
81
+ ### Framework versions
82
 
83
+ - Transformers 4.48.3
84
+ - Pytorch 2.0.0+cu117
85
+ - Datasets 3.2.0
86
+ - Tokenizers 0.21.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.9410023545240498,
4
+ "eval_loss": 0.29460111260414124,
5
+ "eval_runtime": 49.9476,
6
+ "eval_samples_per_second": 297.612,
7
+ "eval_steps_per_second": 297.612,
8
+ "total_flos": 1.96318398191328e+18,
9
+ "train_loss": 2.1099621225725396,
10
+ "train_runtime": 19361.582,
11
+ "train_samples_per_second": 69.094,
12
+ "train_steps_per_second": 0.27
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.9410023545240498,
4
+ "eval_loss": 0.29460111260414124,
5
+ "eval_runtime": 49.9476,
6
+ "eval_samples_per_second": 297.612,
7
+ "eval_steps_per_second": 297.612
8
+ }
log_history.json ADDED
@@ -0,0 +1,1928 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "loss": 7.1456,
4
+ "grad_norm": 4.7170538902282715,
5
+ "learning_rate": 3.824091778202677e-05,
6
+ "epoch": 0.03824091778202677,
7
+ "step": 20
8
+ },
9
+ {
10
+ "loss": 7.1158,
11
+ "grad_norm": 4.5134358406066895,
12
+ "learning_rate": 7.648183556405354e-05,
13
+ "epoch": 0.07648183556405354,
14
+ "step": 40
15
+ },
16
+ {
17
+ "loss": 7.0575,
18
+ "grad_norm": 3.9098806381225586,
19
+ "learning_rate": 0.0001147227533460803,
20
+ "epoch": 0.1147227533460803,
21
+ "step": 60
22
+ },
23
+ {
24
+ "loss": 6.9743,
25
+ "grad_norm": 3.2200050354003906,
26
+ "learning_rate": 0.00015296367112810707,
27
+ "epoch": 0.15296367112810708,
28
+ "step": 80
29
+ },
30
+ {
31
+ "loss": 6.874,
32
+ "grad_norm": 2.5479934215545654,
33
+ "learning_rate": 0.00019120458891013384,
34
+ "epoch": 0.19120458891013384,
35
+ "step": 100
36
+ },
37
+ {
38
+ "loss": 6.7278,
39
+ "grad_norm": 2.168301820755005,
40
+ "learning_rate": 0.0002294455066921606,
41
+ "epoch": 0.2294455066921606,
42
+ "step": 120
43
+ },
44
+ {
45
+ "loss": 6.5778,
46
+ "grad_norm": 2.0083394050598145,
47
+ "learning_rate": 0.0002676864244741874,
48
+ "epoch": 0.2676864244741874,
49
+ "step": 140
50
+ },
51
+ {
52
+ "loss": 6.4192,
53
+ "grad_norm": 1.8298897743225098,
54
+ "learning_rate": 0.00030592734225621415,
55
+ "epoch": 0.30592734225621415,
56
+ "step": 160
57
+ },
58
+ {
59
+ "loss": 6.2416,
60
+ "grad_norm": 1.782423734664917,
61
+ "learning_rate": 0.00034416826003824094,
62
+ "epoch": 0.3441682600382409,
63
+ "step": 180
64
+ },
65
+ {
66
+ "loss": 6.0624,
67
+ "grad_norm": 1.8139146566390991,
68
+ "learning_rate": 0.0003824091778202677,
69
+ "epoch": 0.3824091778202677,
70
+ "step": 200
71
+ },
72
+ {
73
+ "loss": 5.9686,
74
+ "grad_norm": 1.7659958600997925,
75
+ "learning_rate": 0.0004206500956022944,
76
+ "epoch": 0.42065009560229444,
77
+ "step": 220
78
+ },
79
+ {
80
+ "loss": 5.8142,
81
+ "grad_norm": 1.8660094738006592,
82
+ "learning_rate": 0.0004588910133843212,
83
+ "epoch": 0.4588910133843212,
84
+ "step": 240
85
+ },
86
+ {
87
+ "loss": 5.6944,
88
+ "grad_norm": 1.831566333770752,
89
+ "learning_rate": 0.0004971319311663481,
90
+ "epoch": 0.497131931166348,
91
+ "step": 260
92
+ },
93
+ {
94
+ "loss": 5.6101,
95
+ "grad_norm": 1.7546241283416748,
96
+ "learning_rate": 0.0005353728489483748,
97
+ "epoch": 0.5353728489483748,
98
+ "step": 280
99
+ },
100
+ {
101
+ "loss": 5.5192,
102
+ "grad_norm": 1.8890600204467773,
103
+ "learning_rate": 0.0005736137667304016,
104
+ "epoch": 0.5736137667304015,
105
+ "step": 300
106
+ },
107
+ {
108
+ "loss": 5.3822,
109
+ "grad_norm": 1.7542874813079834,
110
+ "learning_rate": 0.0006118546845124283,
111
+ "epoch": 0.6118546845124283,
112
+ "step": 320
113
+ },
114
+ {
115
+ "loss": 5.3236,
116
+ "grad_norm": 1.8762731552124023,
117
+ "learning_rate": 0.000650095602294455,
118
+ "epoch": 0.6500956022944551,
119
+ "step": 340
120
+ },
121
+ {
122
+ "loss": 5.2483,
123
+ "grad_norm": 1.886903166770935,
124
+ "learning_rate": 0.0006883365200764819,
125
+ "epoch": 0.6883365200764818,
126
+ "step": 360
127
+ },
128
+ {
129
+ "loss": 5.1245,
130
+ "grad_norm": 1.9873583316802979,
131
+ "learning_rate": 0.0007265774378585086,
132
+ "epoch": 0.7265774378585086,
133
+ "step": 380
134
+ },
135
+ {
136
+ "loss": 5.0771,
137
+ "grad_norm": 1.953506350517273,
138
+ "learning_rate": 0.0007648183556405354,
139
+ "epoch": 0.7648183556405354,
140
+ "step": 400
141
+ },
142
+ {
143
+ "loss": 5.0354,
144
+ "grad_norm": 1.851192831993103,
145
+ "learning_rate": 0.0008030592734225621,
146
+ "epoch": 0.8030592734225621,
147
+ "step": 420
148
+ },
149
+ {
150
+ "loss": 4.9532,
151
+ "grad_norm": 1.861971139907837,
152
+ "learning_rate": 0.0008413001912045888,
153
+ "epoch": 0.8413001912045889,
154
+ "step": 440
155
+ },
156
+ {
157
+ "loss": 4.8698,
158
+ "grad_norm": 1.9388970136642456,
159
+ "learning_rate": 0.0008795411089866157,
160
+ "epoch": 0.8795411089866156,
161
+ "step": 460
162
+ },
163
+ {
164
+ "loss": 4.847,
165
+ "grad_norm": 1.919184684753418,
166
+ "learning_rate": 0.0009177820267686424,
167
+ "epoch": 0.9177820267686424,
168
+ "step": 480
169
+ },
170
+ {
171
+ "loss": 4.7754,
172
+ "grad_norm": 1.8886795043945312,
173
+ "learning_rate": 0.0009560229445506692,
174
+ "epoch": 0.9560229445506692,
175
+ "step": 500
176
+ },
177
+ {
178
+ "loss": 4.6728,
179
+ "grad_norm": 1.9832813739776611,
180
+ "learning_rate": 0.0009942638623326961,
181
+ "epoch": 0.994263862332696,
182
+ "step": 520
183
+ },
184
+ {
185
+ "eval_loss": 4.345639228820801,
186
+ "eval_accuracy": 0.15042045072317525,
187
+ "eval_runtime": 211.2136,
188
+ "eval_samples_per_second": 70.379,
189
+ "eval_steps_per_second": 70.379,
190
+ "epoch": 1.0,
191
+ "step": 523
192
+ },
193
+ {
194
+ "loss": 4.5657,
195
+ "grad_norm": 1.831260323524475,
196
+ "learning_rate": 0.0009963883577650309,
197
+ "epoch": 1.0325047801147227,
198
+ "step": 540
199
+ },
200
+ {
201
+ "loss": 4.5047,
202
+ "grad_norm": 1.934414029121399,
203
+ "learning_rate": 0.0009921393669003612,
204
+ "epoch": 1.0707456978967496,
205
+ "step": 560
206
+ },
207
+ {
208
+ "loss": 4.4165,
209
+ "grad_norm": 1.7718721628189087,
210
+ "learning_rate": 0.0009878903760356915,
211
+ "epoch": 1.1089866156787762,
212
+ "step": 580
213
+ },
214
+ {
215
+ "loss": 4.3933,
216
+ "grad_norm": 1.741455078125,
217
+ "learning_rate": 0.0009836413851710218,
218
+ "epoch": 1.147227533460803,
219
+ "step": 600
220
+ },
221
+ {
222
+ "loss": 4.3249,
223
+ "grad_norm": 1.8857481479644775,
224
+ "learning_rate": 0.0009793923943063523,
225
+ "epoch": 1.1854684512428297,
226
+ "step": 620
227
+ },
228
+ {
229
+ "loss": 4.2147,
230
+ "grad_norm": 1.8325748443603516,
231
+ "learning_rate": 0.0009751434034416827,
232
+ "epoch": 1.2237093690248566,
233
+ "step": 640
234
+ },
235
+ {
236
+ "loss": 4.1569,
237
+ "grad_norm": 1.8758591413497925,
238
+ "learning_rate": 0.000970894412577013,
239
+ "epoch": 1.2619502868068833,
240
+ "step": 660
241
+ },
242
+ {
243
+ "loss": 4.131,
244
+ "grad_norm": 1.899542212486267,
245
+ "learning_rate": 0.0009666454217123433,
246
+ "epoch": 1.3001912045889101,
247
+ "step": 680
248
+ },
249
+ {
250
+ "loss": 4.0467,
251
+ "grad_norm": 1.8188538551330566,
252
+ "learning_rate": 0.0009623964308476737,
253
+ "epoch": 1.338432122370937,
254
+ "step": 700
255
+ },
256
+ {
257
+ "loss": 3.9904,
258
+ "grad_norm": 1.7679705619812012,
259
+ "learning_rate": 0.000958147439983004,
260
+ "epoch": 1.3766730401529637,
261
+ "step": 720
262
+ },
263
+ {
264
+ "loss": 3.9464,
265
+ "grad_norm": 1.849482774734497,
266
+ "learning_rate": 0.0009538984491183344,
267
+ "epoch": 1.4149139579349903,
268
+ "step": 740
269
+ },
270
+ {
271
+ "loss": 3.9104,
272
+ "grad_norm": 1.8237632513046265,
273
+ "learning_rate": 0.0009496494582536647,
274
+ "epoch": 1.4531548757170172,
275
+ "step": 760
276
+ },
277
+ {
278
+ "loss": 3.8441,
279
+ "grad_norm": 1.8175936937332153,
280
+ "learning_rate": 0.0009454004673889951,
281
+ "epoch": 1.491395793499044,
282
+ "step": 780
283
+ },
284
+ {
285
+ "loss": 3.7898,
286
+ "grad_norm": 1.7967997789382935,
287
+ "learning_rate": 0.0009411514765243255,
288
+ "epoch": 1.5296367112810707,
289
+ "step": 800
290
+ },
291
+ {
292
+ "loss": 3.6894,
293
+ "grad_norm": 1.7681634426116943,
294
+ "learning_rate": 0.0009369024856596558,
295
+ "epoch": 1.5678776290630974,
296
+ "step": 820
297
+ },
298
+ {
299
+ "loss": 3.6798,
300
+ "grad_norm": 1.8655925989151,
301
+ "learning_rate": 0.0009326534947949862,
302
+ "epoch": 1.6061185468451242,
303
+ "step": 840
304
+ },
305
+ {
306
+ "loss": 3.6297,
307
+ "grad_norm": 1.853769302368164,
308
+ "learning_rate": 0.0009284045039303166,
309
+ "epoch": 1.644359464627151,
310
+ "step": 860
311
+ },
312
+ {
313
+ "loss": 3.5592,
314
+ "grad_norm": 1.8198288679122925,
315
+ "learning_rate": 0.0009241555130656469,
316
+ "epoch": 1.682600382409178,
317
+ "step": 880
318
+ },
319
+ {
320
+ "loss": 3.5056,
321
+ "grad_norm": 1.7744460105895996,
322
+ "learning_rate": 0.0009199065222009773,
323
+ "epoch": 1.7208413001912046,
324
+ "step": 900
325
+ },
326
+ {
327
+ "loss": 3.4635,
328
+ "grad_norm": 1.797914981842041,
329
+ "learning_rate": 0.0009156575313363077,
330
+ "epoch": 1.7590822179732313,
331
+ "step": 920
332
+ },
333
+ {
334
+ "loss": 3.4434,
335
+ "grad_norm": 1.8479169607162476,
336
+ "learning_rate": 0.000911408540471638,
337
+ "epoch": 1.7973231357552581,
338
+ "step": 940
339
+ },
340
+ {
341
+ "loss": 3.441,
342
+ "grad_norm": 1.818405032157898,
343
+ "learning_rate": 0.0009071595496069684,
344
+ "epoch": 1.835564053537285,
345
+ "step": 960
346
+ },
347
+ {
348
+ "loss": 3.3934,
349
+ "grad_norm": 1.7609572410583496,
350
+ "learning_rate": 0.0009029105587422988,
351
+ "epoch": 1.8738049713193117,
352
+ "step": 980
353
+ },
354
+ {
355
+ "loss": 3.2961,
356
+ "grad_norm": 1.7228211164474487,
357
+ "learning_rate": 0.0008986615678776291,
358
+ "epoch": 1.9120458891013383,
359
+ "step": 1000
360
+ },
361
+ {
362
+ "loss": 3.2611,
363
+ "grad_norm": 1.8148291110992432,
364
+ "learning_rate": 0.0008944125770129595,
365
+ "epoch": 1.9502868068833652,
366
+ "step": 1020
367
+ },
368
+ {
369
+ "loss": 3.224,
370
+ "grad_norm": 1.933300495147705,
371
+ "learning_rate": 0.0008901635861482899,
372
+ "epoch": 1.988527724665392,
373
+ "step": 1040
374
+ },
375
+ {
376
+ "eval_loss": 2.258894205093384,
377
+ "eval_accuracy": 0.5140935082408342,
378
+ "eval_runtime": 203.9361,
379
+ "eval_samples_per_second": 72.89,
380
+ "eval_steps_per_second": 72.89,
381
+ "epoch": 2.0,
382
+ "step": 1046
383
+ },
384
+ {
385
+ "loss": 3.1667,
386
+ "grad_norm": 1.819346308708191,
387
+ "learning_rate": 0.0008859145952836202,
388
+ "epoch": 2.026768642447419,
389
+ "step": 1060
390
+ },
391
+ {
392
+ "loss": 3.0232,
393
+ "grad_norm": 1.7024896144866943,
394
+ "learning_rate": 0.0008816656044189504,
395
+ "epoch": 2.0650095602294454,
396
+ "step": 1080
397
+ },
398
+ {
399
+ "loss": 3.0489,
400
+ "grad_norm": 1.7023948431015015,
401
+ "learning_rate": 0.000877416613554281,
402
+ "epoch": 2.1032504780114722,
403
+ "step": 1100
404
+ },
405
+ {
406
+ "loss": 2.9732,
407
+ "grad_norm": 1.804140329360962,
408
+ "learning_rate": 0.0008731676226896112,
409
+ "epoch": 2.141491395793499,
410
+ "step": 1120
411
+ },
412
+ {
413
+ "loss": 2.9562,
414
+ "grad_norm": 1.7260992527008057,
415
+ "learning_rate": 0.0008689186318249415,
416
+ "epoch": 2.179732313575526,
417
+ "step": 1140
418
+ },
419
+ {
420
+ "loss": 2.8875,
421
+ "grad_norm": 1.7970356941223145,
422
+ "learning_rate": 0.000864669640960272,
423
+ "epoch": 2.2179732313575524,
424
+ "step": 1160
425
+ },
426
+ {
427
+ "loss": 2.916,
428
+ "grad_norm": 1.8579261302947998,
429
+ "learning_rate": 0.0008604206500956023,
430
+ "epoch": 2.2562141491395793,
431
+ "step": 1180
432
+ },
433
+ {
434
+ "loss": 2.8963,
435
+ "grad_norm": 1.852342128753662,
436
+ "learning_rate": 0.0008561716592309326,
437
+ "epoch": 2.294455066921606,
438
+ "step": 1200
439
+ },
440
+ {
441
+ "loss": 2.8029,
442
+ "grad_norm": 1.8845752477645874,
443
+ "learning_rate": 0.000851922668366263,
444
+ "epoch": 2.332695984703633,
445
+ "step": 1220
446
+ },
447
+ {
448
+ "loss": 2.8237,
449
+ "grad_norm": 1.883952260017395,
450
+ "learning_rate": 0.0008476736775015934,
451
+ "epoch": 2.3709369024856595,
452
+ "step": 1240
453
+ },
454
+ {
455
+ "loss": 2.8473,
456
+ "grad_norm": 1.8383756875991821,
457
+ "learning_rate": 0.0008434246866369237,
458
+ "epoch": 2.4091778202676863,
459
+ "step": 1260
460
+ },
461
+ {
462
+ "loss": 2.7722,
463
+ "grad_norm": 1.8900470733642578,
464
+ "learning_rate": 0.0008391756957722541,
465
+ "epoch": 2.447418738049713,
466
+ "step": 1280
467
+ },
468
+ {
469
+ "loss": 2.7584,
470
+ "grad_norm": 1.8097845315933228,
471
+ "learning_rate": 0.0008349267049075845,
472
+ "epoch": 2.48565965583174,
473
+ "step": 1300
474
+ },
475
+ {
476
+ "loss": 2.7134,
477
+ "grad_norm": 1.7215895652770996,
478
+ "learning_rate": 0.0008306777140429148,
479
+ "epoch": 2.5239005736137665,
480
+ "step": 1320
481
+ },
482
+ {
483
+ "loss": 2.6531,
484
+ "grad_norm": 1.8249051570892334,
485
+ "learning_rate": 0.0008264287231782451,
486
+ "epoch": 2.5621414913957934,
487
+ "step": 1340
488
+ },
489
+ {
490
+ "loss": 2.6675,
491
+ "grad_norm": 1.8082237243652344,
492
+ "learning_rate": 0.0008221797323135756,
493
+ "epoch": 2.6003824091778203,
494
+ "step": 1360
495
+ },
496
+ {
497
+ "loss": 2.5702,
498
+ "grad_norm": 1.7981261014938354,
499
+ "learning_rate": 0.0008179307414489059,
500
+ "epoch": 2.638623326959847,
501
+ "step": 1380
502
+ },
503
+ {
504
+ "loss": 2.6339,
505
+ "grad_norm": 1.6964036226272583,
506
+ "learning_rate": 0.0008136817505842362,
507
+ "epoch": 2.676864244741874,
508
+ "step": 1400
509
+ },
510
+ {
511
+ "loss": 2.5489,
512
+ "grad_norm": 1.755050778388977,
513
+ "learning_rate": 0.0008094327597195667,
514
+ "epoch": 2.7151051625239004,
515
+ "step": 1420
516
+ },
517
+ {
518
+ "loss": 2.5908,
519
+ "grad_norm": 1.7242581844329834,
520
+ "learning_rate": 0.000805183768854897,
521
+ "epoch": 2.7533460803059273,
522
+ "step": 1440
523
+ },
524
+ {
525
+ "loss": 2.5143,
526
+ "grad_norm": 1.819612741470337,
527
+ "learning_rate": 0.0008009347779902273,
528
+ "epoch": 2.791586998087954,
529
+ "step": 1460
530
+ },
531
+ {
532
+ "loss": 2.4662,
533
+ "grad_norm": 1.7033363580703735,
534
+ "learning_rate": 0.0007966857871255578,
535
+ "epoch": 2.8298279158699806,
536
+ "step": 1480
537
+ },
538
+ {
539
+ "loss": 2.4044,
540
+ "grad_norm": 1.7662159204483032,
541
+ "learning_rate": 0.000792436796260888,
542
+ "epoch": 2.8680688336520075,
543
+ "step": 1500
544
+ },
545
+ {
546
+ "loss": 2.4636,
547
+ "grad_norm": 1.7460269927978516,
548
+ "learning_rate": 0.0007881878053962183,
549
+ "epoch": 2.9063097514340344,
550
+ "step": 1520
551
+ },
552
+ {
553
+ "loss": 2.3955,
554
+ "grad_norm": 1.8268380165100098,
555
+ "learning_rate": 0.0007839388145315488,
556
+ "epoch": 2.9445506692160612,
557
+ "step": 1540
558
+ },
559
+ {
560
+ "loss": 2.3964,
561
+ "grad_norm": 1.796981930732727,
562
+ "learning_rate": 0.0007796898236668791,
563
+ "epoch": 2.982791586998088,
564
+ "step": 1560
565
+ },
566
+ {
567
+ "eval_loss": 1.4662528038024902,
568
+ "eval_accuracy": 0.6835519677093844,
569
+ "eval_runtime": 419.2044,
570
+ "eval_samples_per_second": 35.46,
571
+ "eval_steps_per_second": 35.46,
572
+ "epoch": 3.0,
573
+ "step": 1569
574
+ },
575
+ {
576
+ "loss": 2.3174,
577
+ "grad_norm": 1.7852272987365723,
578
+ "learning_rate": 0.0007754408328022094,
579
+ "epoch": 3.0210325047801145,
580
+ "step": 1580
581
+ },
582
+ {
583
+ "loss": 2.2913,
584
+ "grad_norm": 1.8464534282684326,
585
+ "learning_rate": 0.0007711918419375399,
586
+ "epoch": 3.0592734225621414,
587
+ "step": 1600
588
+ },
589
+ {
590
+ "loss": 2.2856,
591
+ "grad_norm": 1.7783145904541016,
592
+ "learning_rate": 0.0007669428510728702,
593
+ "epoch": 3.0975143403441683,
594
+ "step": 1620
595
+ },
596
+ {
597
+ "loss": 2.2099,
598
+ "grad_norm": 1.744454264640808,
599
+ "learning_rate": 0.0007626938602082005,
600
+ "epoch": 3.135755258126195,
601
+ "step": 1640
602
+ },
603
+ {
604
+ "loss": 2.23,
605
+ "grad_norm": 1.8276797533035278,
606
+ "learning_rate": 0.0007584448693435309,
607
+ "epoch": 3.173996175908222,
608
+ "step": 1660
609
+ },
610
+ {
611
+ "loss": 2.1912,
612
+ "grad_norm": 1.8144315481185913,
613
+ "learning_rate": 0.0007541958784788613,
614
+ "epoch": 3.2122370936902485,
615
+ "step": 1680
616
+ },
617
+ {
618
+ "loss": 2.1818,
619
+ "grad_norm": 1.8499830961227417,
620
+ "learning_rate": 0.0007499468876141916,
621
+ "epoch": 3.2504780114722753,
622
+ "step": 1700
623
+ },
624
+ {
625
+ "loss": 2.1349,
626
+ "grad_norm": 1.7623099088668823,
627
+ "learning_rate": 0.000745697896749522,
628
+ "epoch": 3.288718929254302,
629
+ "step": 1720
630
+ },
631
+ {
632
+ "loss": 2.1055,
633
+ "grad_norm": 1.8180640935897827,
634
+ "learning_rate": 0.0007414489058848524,
635
+ "epoch": 3.3269598470363286,
636
+ "step": 1740
637
+ },
638
+ {
639
+ "loss": 2.1077,
640
+ "grad_norm": 1.8159964084625244,
641
+ "learning_rate": 0.0007371999150201827,
642
+ "epoch": 3.3652007648183555,
643
+ "step": 1760
644
+ },
645
+ {
646
+ "loss": 2.0999,
647
+ "grad_norm": 1.7902129888534546,
648
+ "learning_rate": 0.0007329509241555131,
649
+ "epoch": 3.4034416826003824,
650
+ "step": 1780
651
+ },
652
+ {
653
+ "loss": 2.1188,
654
+ "grad_norm": 1.7685898542404175,
655
+ "learning_rate": 0.0007287019332908435,
656
+ "epoch": 3.4416826003824093,
657
+ "step": 1800
658
+ },
659
+ {
660
+ "loss": 2.0956,
661
+ "grad_norm": 1.758325219154358,
662
+ "learning_rate": 0.0007244529424261738,
663
+ "epoch": 3.479923518164436,
664
+ "step": 1820
665
+ },
666
+ {
667
+ "loss": 2.0488,
668
+ "grad_norm": 1.7802537679672241,
669
+ "learning_rate": 0.0007202039515615042,
670
+ "epoch": 3.5181644359464626,
671
+ "step": 1840
672
+ },
673
+ {
674
+ "loss": 2.0776,
675
+ "grad_norm": 1.8220280408859253,
676
+ "learning_rate": 0.0007159549606968346,
677
+ "epoch": 3.5564053537284894,
678
+ "step": 1860
679
+ },
680
+ {
681
+ "loss": 2.0257,
682
+ "grad_norm": 1.8494378328323364,
683
+ "learning_rate": 0.0007117059698321649,
684
+ "epoch": 3.5946462715105163,
685
+ "step": 1880
686
+ },
687
+ {
688
+ "loss": 2.0332,
689
+ "grad_norm": 1.719109296798706,
690
+ "learning_rate": 0.0007074569789674953,
691
+ "epoch": 3.632887189292543,
692
+ "step": 1900
693
+ },
694
+ {
695
+ "loss": 2.0052,
696
+ "grad_norm": 1.9517509937286377,
697
+ "learning_rate": 0.0007032079881028257,
698
+ "epoch": 3.67112810707457,
699
+ "step": 1920
700
+ },
701
+ {
702
+ "loss": 1.9901,
703
+ "grad_norm": 1.7318940162658691,
704
+ "learning_rate": 0.0006989589972381559,
705
+ "epoch": 3.7093690248565965,
706
+ "step": 1940
707
+ },
708
+ {
709
+ "loss": 1.9411,
710
+ "grad_norm": 1.767015814781189,
711
+ "learning_rate": 0.0006947100063734863,
712
+ "epoch": 3.7476099426386233,
713
+ "step": 1960
714
+ },
715
+ {
716
+ "loss": 2.0048,
717
+ "grad_norm": 1.761806607246399,
718
+ "learning_rate": 0.0006904610155088166,
719
+ "epoch": 3.78585086042065,
720
+ "step": 1980
721
+ },
722
+ {
723
+ "loss": 1.9312,
724
+ "grad_norm": 1.7126002311706543,
725
+ "learning_rate": 0.000686212024644147,
726
+ "epoch": 3.8240917782026767,
727
+ "step": 2000
728
+ },
729
+ {
730
+ "loss": 1.9443,
731
+ "grad_norm": 1.7167437076568604,
732
+ "learning_rate": 0.0006819630337794774,
733
+ "epoch": 3.8623326959847035,
734
+ "step": 2020
735
+ },
736
+ {
737
+ "loss": 1.893,
738
+ "grad_norm": 1.749881386756897,
739
+ "learning_rate": 0.0006777140429148077,
740
+ "epoch": 3.9005736137667304,
741
+ "step": 2040
742
+ },
743
+ {
744
+ "loss": 1.8876,
745
+ "grad_norm": 1.6846592426300049,
746
+ "learning_rate": 0.0006734650520501381,
747
+ "epoch": 3.9388145315487573,
748
+ "step": 2060
749
+ },
750
+ {
751
+ "loss": 1.8474,
752
+ "grad_norm": 1.8149057626724243,
753
+ "learning_rate": 0.0006692160611854685,
754
+ "epoch": 3.977055449330784,
755
+ "step": 2080
756
+ },
757
+ {
758
+ "eval_loss": 0.9547563195228577,
759
+ "eval_accuracy": 0.7926673393878237,
760
+ "eval_runtime": 183.3001,
761
+ "eval_samples_per_second": 81.097,
762
+ "eval_steps_per_second": 81.097,
763
+ "epoch": 4.0,
764
+ "step": 2092
765
+ },
766
+ {
767
+ "loss": 1.813,
768
+ "grad_norm": 1.6455098390579224,
769
+ "learning_rate": 0.0006649670703207988,
770
+ "epoch": 4.015296367112811,
771
+ "step": 2100
772
+ },
773
+ {
774
+ "loss": 1.7354,
775
+ "grad_norm": 1.6958200931549072,
776
+ "learning_rate": 0.0006607180794561292,
777
+ "epoch": 4.053537284894838,
778
+ "step": 2120
779
+ },
780
+ {
781
+ "loss": 1.7479,
782
+ "grad_norm": 1.7456037998199463,
783
+ "learning_rate": 0.0006564690885914596,
784
+ "epoch": 4.091778202676864,
785
+ "step": 2140
786
+ },
787
+ {
788
+ "loss": 1.7138,
789
+ "grad_norm": 1.7887734174728394,
790
+ "learning_rate": 0.0006522200977267899,
791
+ "epoch": 4.130019120458891,
792
+ "step": 2160
793
+ },
794
+ {
795
+ "loss": 1.7023,
796
+ "grad_norm": 1.7080284357070923,
797
+ "learning_rate": 0.0006479711068621203,
798
+ "epoch": 4.168260038240918,
799
+ "step": 2180
800
+ },
801
+ {
802
+ "loss": 1.7526,
803
+ "grad_norm": 1.8061983585357666,
804
+ "learning_rate": 0.0006437221159974506,
805
+ "epoch": 4.2065009560229445,
806
+ "step": 2200
807
+ },
808
+ {
809
+ "loss": 1.7474,
810
+ "grad_norm": 1.7831811904907227,
811
+ "learning_rate": 0.000639473125132781,
812
+ "epoch": 4.244741873804971,
813
+ "step": 2220
814
+ },
815
+ {
816
+ "loss": 1.6688,
817
+ "grad_norm": 1.752357840538025,
818
+ "learning_rate": 0.0006352241342681113,
819
+ "epoch": 4.282982791586998,
820
+ "step": 2240
821
+ },
822
+ {
823
+ "loss": 1.7009,
824
+ "grad_norm": 1.7843034267425537,
825
+ "learning_rate": 0.0006309751434034417,
826
+ "epoch": 4.321223709369025,
827
+ "step": 2260
828
+ },
829
+ {
830
+ "loss": 1.6727,
831
+ "grad_norm": 1.7608367204666138,
832
+ "learning_rate": 0.0006267261525387721,
833
+ "epoch": 4.359464627151052,
834
+ "step": 2280
835
+ },
836
+ {
837
+ "loss": 1.6801,
838
+ "grad_norm": 1.6877254247665405,
839
+ "learning_rate": 0.0006224771616741024,
840
+ "epoch": 4.397705544933078,
841
+ "step": 2300
842
+ },
843
+ {
844
+ "loss": 1.7108,
845
+ "grad_norm": 1.7891350984573364,
846
+ "learning_rate": 0.0006182281708094328,
847
+ "epoch": 4.435946462715105,
848
+ "step": 2320
849
+ },
850
+ {
851
+ "loss": 1.6442,
852
+ "grad_norm": 1.7104123830795288,
853
+ "learning_rate": 0.0006139791799447631,
854
+ "epoch": 4.474187380497132,
855
+ "step": 2340
856
+ },
857
+ {
858
+ "loss": 1.6531,
859
+ "grad_norm": 1.7026969194412231,
860
+ "learning_rate": 0.0006097301890800934,
861
+ "epoch": 4.512428298279159,
862
+ "step": 2360
863
+ },
864
+ {
865
+ "loss": 1.6539,
866
+ "grad_norm": 1.7890552282333374,
867
+ "learning_rate": 0.0006054811982154238,
868
+ "epoch": 4.550669216061186,
869
+ "step": 2380
870
+ },
871
+ {
872
+ "loss": 1.6681,
873
+ "grad_norm": 1.8423861265182495,
874
+ "learning_rate": 0.0006012322073507542,
875
+ "epoch": 4.588910133843212,
876
+ "step": 2400
877
+ },
878
+ {
879
+ "loss": 1.5935,
880
+ "grad_norm": 1.6434499025344849,
881
+ "learning_rate": 0.0005969832164860845,
882
+ "epoch": 4.627151051625239,
883
+ "step": 2420
884
+ },
885
+ {
886
+ "loss": 1.6273,
887
+ "grad_norm": 1.7261130809783936,
888
+ "learning_rate": 0.0005927342256214149,
889
+ "epoch": 4.665391969407266,
890
+ "step": 2440
891
+ },
892
+ {
893
+ "loss": 1.6181,
894
+ "grad_norm": 1.7288273572921753,
895
+ "learning_rate": 0.0005884852347567453,
896
+ "epoch": 4.7036328871892925,
897
+ "step": 2460
898
+ },
899
+ {
900
+ "loss": 1.5719,
901
+ "grad_norm": 1.773258924484253,
902
+ "learning_rate": 0.0005842362438920756,
903
+ "epoch": 4.741873804971319,
904
+ "step": 2480
905
+ },
906
+ {
907
+ "loss": 1.578,
908
+ "grad_norm": 1.7676658630371094,
909
+ "learning_rate": 0.000579987253027406,
910
+ "epoch": 4.780114722753346,
911
+ "step": 2500
912
+ },
913
+ {
914
+ "loss": 1.535,
915
+ "grad_norm": 1.8115794658660889,
916
+ "learning_rate": 0.0005757382621627364,
917
+ "epoch": 4.818355640535373,
918
+ "step": 2520
919
+ },
920
+ {
921
+ "loss": 1.5493,
922
+ "grad_norm": 1.7989414930343628,
923
+ "learning_rate": 0.0005714892712980667,
924
+ "epoch": 4.8565965583174,
925
+ "step": 2540
926
+ },
927
+ {
928
+ "loss": 1.5489,
929
+ "grad_norm": 1.6607849597930908,
930
+ "learning_rate": 0.000567240280433397,
931
+ "epoch": 4.894837476099426,
932
+ "step": 2560
933
+ },
934
+ {
935
+ "loss": 1.5091,
936
+ "grad_norm": 1.630257487297058,
937
+ "learning_rate": 0.0005629912895687275,
938
+ "epoch": 4.933078393881453,
939
+ "step": 2580
940
+ },
941
+ {
942
+ "loss": 1.5275,
943
+ "grad_norm": 1.7995944023132324,
944
+ "learning_rate": 0.0005587422987040578,
945
+ "epoch": 4.97131931166348,
946
+ "step": 2600
947
+ },
948
+ {
949
+ "eval_loss": 0.6697778105735779,
950
+ "eval_accuracy": 0.8571140262361251,
951
+ "eval_runtime": 181.3784,
952
+ "eval_samples_per_second": 81.956,
953
+ "eval_steps_per_second": 81.956,
954
+ "epoch": 5.0,
955
+ "step": 2615
956
+ },
957
+ {
958
+ "loss": 1.4774,
959
+ "grad_norm": 1.7868553400039673,
960
+ "learning_rate": 0.0005544933078393881,
961
+ "epoch": 5.009560229445507,
962
+ "step": 2620
963
+ },
964
+ {
965
+ "loss": 1.3955,
966
+ "grad_norm": 1.6380654573440552,
967
+ "learning_rate": 0.0005502443169747186,
968
+ "epoch": 5.047801147227533,
969
+ "step": 2640
970
+ },
971
+ {
972
+ "loss": 1.4414,
973
+ "grad_norm": 1.7844533920288086,
974
+ "learning_rate": 0.0005459953261100489,
975
+ "epoch": 5.08604206500956,
976
+ "step": 2660
977
+ },
978
+ {
979
+ "loss": 1.3782,
980
+ "grad_norm": 1.779080867767334,
981
+ "learning_rate": 0.0005417463352453792,
982
+ "epoch": 5.124282982791587,
983
+ "step": 2680
984
+ },
985
+ {
986
+ "loss": 1.4152,
987
+ "grad_norm": 1.741326928138733,
988
+ "learning_rate": 0.0005374973443807097,
989
+ "epoch": 5.162523900573614,
990
+ "step": 2700
991
+ },
992
+ {
993
+ "loss": 1.3996,
994
+ "grad_norm": 1.7447401285171509,
995
+ "learning_rate": 0.00053324835351604,
996
+ "epoch": 5.2007648183556405,
997
+ "step": 2720
998
+ },
999
+ {
1000
+ "loss": 1.4137,
1001
+ "grad_norm": 1.8067736625671387,
1002
+ "learning_rate": 0.0005289993626513702,
1003
+ "epoch": 5.239005736137667,
1004
+ "step": 2740
1005
+ },
1006
+ {
1007
+ "loss": 1.3937,
1008
+ "grad_norm": 1.7393046617507935,
1009
+ "learning_rate": 0.0005247503717867008,
1010
+ "epoch": 5.277246653919694,
1011
+ "step": 2760
1012
+ },
1013
+ {
1014
+ "loss": 1.3912,
1015
+ "grad_norm": 1.756184458732605,
1016
+ "learning_rate": 0.000520501380922031,
1017
+ "epoch": 5.315487571701721,
1018
+ "step": 2780
1019
+ },
1020
+ {
1021
+ "loss": 1.387,
1022
+ "grad_norm": 1.7133733034133911,
1023
+ "learning_rate": 0.0005162523900573613,
1024
+ "epoch": 5.353728489483748,
1025
+ "step": 2800
1026
+ },
1027
+ {
1028
+ "loss": 1.3551,
1029
+ "grad_norm": 1.6597713232040405,
1030
+ "learning_rate": 0.0005120033991926918,
1031
+ "epoch": 5.3919694072657744,
1032
+ "step": 2820
1033
+ },
1034
+ {
1035
+ "loss": 1.3557,
1036
+ "grad_norm": 1.8462845087051392,
1037
+ "learning_rate": 0.0005077544083280221,
1038
+ "epoch": 5.430210325047801,
1039
+ "step": 2840
1040
+ },
1041
+ {
1042
+ "loss": 1.3495,
1043
+ "grad_norm": 1.6737143993377686,
1044
+ "learning_rate": 0.0005035054174633524,
1045
+ "epoch": 5.468451242829828,
1046
+ "step": 2860
1047
+ },
1048
+ {
1049
+ "loss": 1.394,
1050
+ "grad_norm": 1.7071157693862915,
1051
+ "learning_rate": 0.0004992564265986828,
1052
+ "epoch": 5.506692160611855,
1053
+ "step": 2880
1054
+ },
1055
+ {
1056
+ "loss": 1.3263,
1057
+ "grad_norm": 1.663072943687439,
1058
+ "learning_rate": 0.0004950074357340132,
1059
+ "epoch": 5.544933078393882,
1060
+ "step": 2900
1061
+ },
1062
+ {
1063
+ "loss": 1.3474,
1064
+ "grad_norm": 1.640093207359314,
1065
+ "learning_rate": 0.0004907584448693436,
1066
+ "epoch": 5.583173996175908,
1067
+ "step": 2920
1068
+ },
1069
+ {
1070
+ "loss": 1.3375,
1071
+ "grad_norm": 1.762568712234497,
1072
+ "learning_rate": 0.0004865094540046739,
1073
+ "epoch": 5.621414913957935,
1074
+ "step": 2940
1075
+ },
1076
+ {
1077
+ "loss": 1.3218,
1078
+ "grad_norm": 1.6714434623718262,
1079
+ "learning_rate": 0.00048226046314000425,
1080
+ "epoch": 5.659655831739962,
1081
+ "step": 2960
1082
+ },
1083
+ {
1084
+ "loss": 1.3008,
1085
+ "grad_norm": 1.7594107389450073,
1086
+ "learning_rate": 0.0004780114722753346,
1087
+ "epoch": 5.6978967495219885,
1088
+ "step": 2980
1089
+ },
1090
+ {
1091
+ "loss": 1.3331,
1092
+ "grad_norm": 1.6483973264694214,
1093
+ "learning_rate": 0.000473762481410665,
1094
+ "epoch": 5.736137667304015,
1095
+ "step": 3000
1096
+ },
1097
+ {
1098
+ "loss": 1.2775,
1099
+ "grad_norm": 1.7252651453018188,
1100
+ "learning_rate": 0.00046951349054599533,
1101
+ "epoch": 5.774378585086042,
1102
+ "step": 3020
1103
+ },
1104
+ {
1105
+ "loss": 1.2747,
1106
+ "grad_norm": 1.7860745191574097,
1107
+ "learning_rate": 0.0004652644996813257,
1108
+ "epoch": 5.812619502868069,
1109
+ "step": 3040
1110
+ },
1111
+ {
1112
+ "loss": 1.2946,
1113
+ "grad_norm": 1.749874234199524,
1114
+ "learning_rate": 0.0004610155088166561,
1115
+ "epoch": 5.850860420650095,
1116
+ "step": 3060
1117
+ },
1118
+ {
1119
+ "loss": 1.2849,
1120
+ "grad_norm": 1.7197644710540771,
1121
+ "learning_rate": 0.0004567665179519864,
1122
+ "epoch": 5.8891013384321225,
1123
+ "step": 3080
1124
+ },
1125
+ {
1126
+ "loss": 1.2544,
1127
+ "grad_norm": 1.6396132707595825,
1128
+ "learning_rate": 0.00045251752708731676,
1129
+ "epoch": 5.927342256214149,
1130
+ "step": 3100
1131
+ },
1132
+ {
1133
+ "loss": 1.248,
1134
+ "grad_norm": 1.720376968383789,
1135
+ "learning_rate": 0.0004482685362226471,
1136
+ "epoch": 5.965583173996176,
1137
+ "step": 3120
1138
+ },
1139
+ {
1140
+ "eval_loss": 0.5270123481750488,
1141
+ "eval_accuracy": 0.8899428187016482,
1142
+ "eval_runtime": 183.7621,
1143
+ "eval_samples_per_second": 80.893,
1144
+ "eval_steps_per_second": 80.893,
1145
+ "epoch": 6.0,
1146
+ "step": 3138
1147
+ },
1148
+ {
1149
+ "loss": 1.2398,
1150
+ "grad_norm": 1.5206599235534668,
1151
+ "learning_rate": 0.0004440195453579775,
1152
+ "epoch": 6.003824091778203,
1153
+ "step": 3140
1154
+ },
1155
+ {
1156
+ "loss": 1.2097,
1157
+ "grad_norm": 1.7172082662582397,
1158
+ "learning_rate": 0.00043977055449330785,
1159
+ "epoch": 6.042065009560229,
1160
+ "step": 3160
1161
+ },
1162
+ {
1163
+ "loss": 1.1669,
1164
+ "grad_norm": 1.5570909976959229,
1165
+ "learning_rate": 0.0004355215636286382,
1166
+ "epoch": 6.080305927342256,
1167
+ "step": 3180
1168
+ },
1169
+ {
1170
+ "loss": 1.1647,
1171
+ "grad_norm": 1.7044614553451538,
1172
+ "learning_rate": 0.0004312725727639686,
1173
+ "epoch": 6.118546845124283,
1174
+ "step": 3200
1175
+ },
1176
+ {
1177
+ "loss": 1.1627,
1178
+ "grad_norm": 1.5819571018218994,
1179
+ "learning_rate": 0.0004270235818992989,
1180
+ "epoch": 6.15678776290631,
1181
+ "step": 3220
1182
+ },
1183
+ {
1184
+ "loss": 1.1728,
1185
+ "grad_norm": 1.7076871395111084,
1186
+ "learning_rate": 0.0004227745910346293,
1187
+ "epoch": 6.195028680688337,
1188
+ "step": 3240
1189
+ },
1190
+ {
1191
+ "loss": 1.1459,
1192
+ "grad_norm": 1.7301490306854248,
1193
+ "learning_rate": 0.0004185256001699597,
1194
+ "epoch": 6.233269598470363,
1195
+ "step": 3260
1196
+ },
1197
+ {
1198
+ "loss": 1.1676,
1199
+ "grad_norm": 1.7135626077651978,
1200
+ "learning_rate": 0.00041427660930528997,
1201
+ "epoch": 6.27151051625239,
1202
+ "step": 3280
1203
+ },
1204
+ {
1205
+ "loss": 1.1488,
1206
+ "grad_norm": 1.602142572402954,
1207
+ "learning_rate": 0.00041002761844062037,
1208
+ "epoch": 6.309751434034417,
1209
+ "step": 3300
1210
+ },
1211
+ {
1212
+ "loss": 1.1395,
1213
+ "grad_norm": 1.755293846130371,
1214
+ "learning_rate": 0.00040577862757595076,
1215
+ "epoch": 6.347992351816444,
1216
+ "step": 3320
1217
+ },
1218
+ {
1219
+ "loss": 1.1321,
1220
+ "grad_norm": 1.663662314414978,
1221
+ "learning_rate": 0.00040152963671128105,
1222
+ "epoch": 6.3862332695984705,
1223
+ "step": 3340
1224
+ },
1225
+ {
1226
+ "loss": 1.1317,
1227
+ "grad_norm": 1.7366993427276611,
1228
+ "learning_rate": 0.00039728064584661145,
1229
+ "epoch": 6.424474187380497,
1230
+ "step": 3360
1231
+ },
1232
+ {
1233
+ "loss": 1.1449,
1234
+ "grad_norm": 1.7560149431228638,
1235
+ "learning_rate": 0.0003930316549819418,
1236
+ "epoch": 6.462715105162524,
1237
+ "step": 3380
1238
+ },
1239
+ {
1240
+ "loss": 1.13,
1241
+ "grad_norm": 1.7576582431793213,
1242
+ "learning_rate": 0.00038878266411727214,
1243
+ "epoch": 6.500956022944551,
1244
+ "step": 3400
1245
+ },
1246
+ {
1247
+ "loss": 1.1419,
1248
+ "grad_norm": 1.7916873693466187,
1249
+ "learning_rate": 0.00038453367325260254,
1250
+ "epoch": 6.539196940726577,
1251
+ "step": 3420
1252
+ },
1253
+ {
1254
+ "loss": 1.1107,
1255
+ "grad_norm": 1.5987508296966553,
1256
+ "learning_rate": 0.0003802846823879329,
1257
+ "epoch": 6.577437858508604,
1258
+ "step": 3440
1259
+ },
1260
+ {
1261
+ "loss": 1.1162,
1262
+ "grad_norm": 1.8192518949508667,
1263
+ "learning_rate": 0.0003760356915232632,
1264
+ "epoch": 6.615678776290631,
1265
+ "step": 3460
1266
+ },
1267
+ {
1268
+ "loss": 1.1255,
1269
+ "grad_norm": 1.7236486673355103,
1270
+ "learning_rate": 0.0003717867006585936,
1271
+ "epoch": 6.653919694072657,
1272
+ "step": 3480
1273
+ },
1274
+ {
1275
+ "loss": 1.0629,
1276
+ "grad_norm": 1.8209389448165894,
1277
+ "learning_rate": 0.0003675377097939239,
1278
+ "epoch": 6.692160611854685,
1279
+ "step": 3500
1280
+ },
1281
+ {
1282
+ "loss": 1.0809,
1283
+ "grad_norm": 1.652782678604126,
1284
+ "learning_rate": 0.0003632887189292543,
1285
+ "epoch": 6.730401529636711,
1286
+ "step": 3520
1287
+ },
1288
+ {
1289
+ "loss": 1.1286,
1290
+ "grad_norm": 1.6148645877838135,
1291
+ "learning_rate": 0.00035903972806458466,
1292
+ "epoch": 6.768642447418738,
1293
+ "step": 3540
1294
+ },
1295
+ {
1296
+ "loss": 1.1069,
1297
+ "grad_norm": 1.6869423389434814,
1298
+ "learning_rate": 0.000354790737199915,
1299
+ "epoch": 6.806883365200765,
1300
+ "step": 3560
1301
+ },
1302
+ {
1303
+ "loss": 1.0911,
1304
+ "grad_norm": 1.6373172998428345,
1305
+ "learning_rate": 0.0003505417463352454,
1306
+ "epoch": 6.845124282982791,
1307
+ "step": 3580
1308
+ },
1309
+ {
1310
+ "loss": 1.0808,
1311
+ "grad_norm": 1.6761549711227417,
1312
+ "learning_rate": 0.00034629275547057574,
1313
+ "epoch": 6.8833652007648185,
1314
+ "step": 3600
1315
+ },
1316
+ {
1317
+ "loss": 1.0809,
1318
+ "grad_norm": 1.6510460376739502,
1319
+ "learning_rate": 0.0003420437646059061,
1320
+ "epoch": 6.921606118546845,
1321
+ "step": 3620
1322
+ },
1323
+ {
1324
+ "loss": 1.0912,
1325
+ "grad_norm": 1.7351855039596558,
1326
+ "learning_rate": 0.0003377947737412365,
1327
+ "epoch": 6.959847036328872,
1328
+ "step": 3640
1329
+ },
1330
+ {
1331
+ "loss": 1.0991,
1332
+ "grad_norm": 1.7165274620056152,
1333
+ "learning_rate": 0.00033354578287656683,
1334
+ "epoch": 6.998087954110899,
1335
+ "step": 3660
1336
+ },
1337
+ {
1338
+ "eval_loss": 0.44995447993278503,
1339
+ "eval_accuracy": 0.9037336024217961,
1340
+ "eval_runtime": 189.5634,
1341
+ "eval_samples_per_second": 78.417,
1342
+ "eval_steps_per_second": 78.417,
1343
+ "epoch": 7.0,
1344
+ "step": 3661
1345
+ },
1346
+ {
1347
+ "loss": 1.0154,
1348
+ "grad_norm": 1.6468501091003418,
1349
+ "learning_rate": 0.0003292967920118972,
1350
+ "epoch": 7.036328871892925,
1351
+ "step": 3680
1352
+ },
1353
+ {
1354
+ "loss": 1.0378,
1355
+ "grad_norm": 1.79421067237854,
1356
+ "learning_rate": 0.0003250478011472275,
1357
+ "epoch": 7.074569789674952,
1358
+ "step": 3700
1359
+ },
1360
+ {
1361
+ "loss": 1.0145,
1362
+ "grad_norm": 1.7234885692596436,
1363
+ "learning_rate": 0.0003207988102825579,
1364
+ "epoch": 7.112810707456979,
1365
+ "step": 3720
1366
+ },
1367
+ {
1368
+ "loss": 1.0012,
1369
+ "grad_norm": 1.6947157382965088,
1370
+ "learning_rate": 0.00031654981941788826,
1371
+ "epoch": 7.151051625239006,
1372
+ "step": 3740
1373
+ },
1374
+ {
1375
+ "loss": 1.0191,
1376
+ "grad_norm": 1.6818758249282837,
1377
+ "learning_rate": 0.0003123008285532186,
1378
+ "epoch": 7.189292543021033,
1379
+ "step": 3760
1380
+ },
1381
+ {
1382
+ "loss": 1.0437,
1383
+ "grad_norm": 1.557080864906311,
1384
+ "learning_rate": 0.000308051837688549,
1385
+ "epoch": 7.227533460803059,
1386
+ "step": 3780
1387
+ },
1388
+ {
1389
+ "loss": 1.0079,
1390
+ "grad_norm": 1.6532793045043945,
1391
+ "learning_rate": 0.00030380284682387935,
1392
+ "epoch": 7.265774378585086,
1393
+ "step": 3800
1394
+ },
1395
+ {
1396
+ "loss": 0.9904,
1397
+ "grad_norm": 1.646686315536499,
1398
+ "learning_rate": 0.0002995538559592097,
1399
+ "epoch": 7.304015296367113,
1400
+ "step": 3820
1401
+ },
1402
+ {
1403
+ "loss": 1.0002,
1404
+ "grad_norm": 1.6772829294204712,
1405
+ "learning_rate": 0.0002953048650945401,
1406
+ "epoch": 7.342256214149139,
1407
+ "step": 3840
1408
+ },
1409
+ {
1410
+ "loss": 0.9674,
1411
+ "grad_norm": 1.6452054977416992,
1412
+ "learning_rate": 0.0002910558742298704,
1413
+ "epoch": 7.3804971319311665,
1414
+ "step": 3860
1415
+ },
1416
+ {
1417
+ "loss": 0.9642,
1418
+ "grad_norm": 1.592207908630371,
1419
+ "learning_rate": 0.0002868068833652008,
1420
+ "epoch": 7.418738049713193,
1421
+ "step": 3880
1422
+ },
1423
+ {
1424
+ "loss": 0.973,
1425
+ "grad_norm": 1.7015941143035889,
1426
+ "learning_rate": 0.0002825578925005312,
1427
+ "epoch": 7.45697896749522,
1428
+ "step": 3900
1429
+ },
1430
+ {
1431
+ "loss": 0.9803,
1432
+ "grad_norm": 1.6589232683181763,
1433
+ "learning_rate": 0.00027830890163586146,
1434
+ "epoch": 7.495219885277247,
1435
+ "step": 3920
1436
+ },
1437
+ {
1438
+ "loss": 0.9702,
1439
+ "grad_norm": 1.660190463066101,
1440
+ "learning_rate": 0.00027405991077119186,
1441
+ "epoch": 7.533460803059273,
1442
+ "step": 3940
1443
+ },
1444
+ {
1445
+ "loss": 0.9919,
1446
+ "grad_norm": 1.7052509784698486,
1447
+ "learning_rate": 0.00026981091990652226,
1448
+ "epoch": 7.5717017208413,
1449
+ "step": 3960
1450
+ },
1451
+ {
1452
+ "loss": 0.938,
1453
+ "grad_norm": 1.6874445676803589,
1454
+ "learning_rate": 0.00026556192904185255,
1455
+ "epoch": 7.609942638623327,
1456
+ "step": 3980
1457
+ },
1458
+ {
1459
+ "loss": 0.9564,
1460
+ "grad_norm": 1.811640739440918,
1461
+ "learning_rate": 0.00026131293817718295,
1462
+ "epoch": 7.648183556405353,
1463
+ "step": 4000
1464
+ },
1465
+ {
1466
+ "loss": 0.9432,
1467
+ "grad_norm": 1.741968035697937,
1468
+ "learning_rate": 0.00025706394731251324,
1469
+ "epoch": 7.686424474187381,
1470
+ "step": 4020
1471
+ },
1472
+ {
1473
+ "loss": 0.9082,
1474
+ "grad_norm": 1.6731518507003784,
1475
+ "learning_rate": 0.00025281495644784364,
1476
+ "epoch": 7.724665391969407,
1477
+ "step": 4040
1478
+ },
1479
+ {
1480
+ "loss": 0.92,
1481
+ "grad_norm": 1.7399870157241821,
1482
+ "learning_rate": 0.00024856596558317403,
1483
+ "epoch": 7.762906309751434,
1484
+ "step": 4060
1485
+ },
1486
+ {
1487
+ "loss": 0.9433,
1488
+ "grad_norm": 1.580674171447754,
1489
+ "learning_rate": 0.0002443169747185044,
1490
+ "epoch": 7.801147227533461,
1491
+ "step": 4080
1492
+ },
1493
+ {
1494
+ "loss": 0.9584,
1495
+ "grad_norm": 1.5683550834655762,
1496
+ "learning_rate": 0.00024006798385383472,
1497
+ "epoch": 7.839388145315487,
1498
+ "step": 4100
1499
+ },
1500
+ {
1501
+ "loss": 0.9318,
1502
+ "grad_norm": 1.7664682865142822,
1503
+ "learning_rate": 0.00023581899298916507,
1504
+ "epoch": 7.8776290630975145,
1505
+ "step": 4120
1506
+ },
1507
+ {
1508
+ "loss": 0.8852,
1509
+ "grad_norm": 1.5522887706756592,
1510
+ "learning_rate": 0.00023157000212449544,
1511
+ "epoch": 7.915869980879541,
1512
+ "step": 4140
1513
+ },
1514
+ {
1515
+ "loss": 0.9121,
1516
+ "grad_norm": 1.626836895942688,
1517
+ "learning_rate": 0.00022732101125982578,
1518
+ "epoch": 7.954110898661568,
1519
+ "step": 4160
1520
+ },
1521
+ {
1522
+ "loss": 0.9221,
1523
+ "grad_norm": 1.8042898178100586,
1524
+ "learning_rate": 0.00022307202039515615,
1525
+ "epoch": 7.992351816443595,
1526
+ "step": 4180
1527
+ },
1528
+ {
1529
+ "eval_loss": 0.3572401702404022,
1530
+ "eval_accuracy": 0.9266733938782374,
1531
+ "eval_runtime": 168.0336,
1532
+ "eval_samples_per_second": 88.464,
1533
+ "eval_steps_per_second": 88.464,
1534
+ "epoch": 8.0,
1535
+ "step": 4184
1536
+ },
1537
+ {
1538
+ "loss": 0.8979,
1539
+ "grad_norm": 1.545024037361145,
1540
+ "learning_rate": 0.00021882302953048652,
1541
+ "epoch": 8.030592734225621,
1542
+ "step": 4200
1543
+ },
1544
+ {
1545
+ "loss": 0.8464,
1546
+ "grad_norm": 1.592607855796814,
1547
+ "learning_rate": 0.00021457403866581687,
1548
+ "epoch": 8.068833652007648,
1549
+ "step": 4220
1550
+ },
1551
+ {
1552
+ "loss": 0.8634,
1553
+ "grad_norm": 1.5347646474838257,
1554
+ "learning_rate": 0.0002103250478011472,
1555
+ "epoch": 8.107074569789676,
1556
+ "step": 4240
1557
+ },
1558
+ {
1559
+ "loss": 0.8425,
1560
+ "grad_norm": 1.621201515197754,
1561
+ "learning_rate": 0.0002060760569364776,
1562
+ "epoch": 8.145315487571702,
1563
+ "step": 4260
1564
+ },
1565
+ {
1566
+ "loss": 0.8776,
1567
+ "grad_norm": 1.7381062507629395,
1568
+ "learning_rate": 0.00020182706607180795,
1569
+ "epoch": 8.183556405353729,
1570
+ "step": 4280
1571
+ },
1572
+ {
1573
+ "loss": 0.854,
1574
+ "grad_norm": 1.5798373222351074,
1575
+ "learning_rate": 0.0001975780752071383,
1576
+ "epoch": 8.221797323135755,
1577
+ "step": 4300
1578
+ },
1579
+ {
1580
+ "loss": 0.8646,
1581
+ "grad_norm": 1.5751338005065918,
1582
+ "learning_rate": 0.00019332908434246867,
1583
+ "epoch": 8.260038240917781,
1584
+ "step": 4320
1585
+ },
1586
+ {
1587
+ "loss": 0.8521,
1588
+ "grad_norm": 1.570742130279541,
1589
+ "learning_rate": 0.00018908009347779904,
1590
+ "epoch": 8.29827915869981,
1591
+ "step": 4340
1592
+ },
1593
+ {
1594
+ "loss": 0.8858,
1595
+ "grad_norm": 1.7959846258163452,
1596
+ "learning_rate": 0.00018483110261312938,
1597
+ "epoch": 8.336520076481836,
1598
+ "step": 4360
1599
+ },
1600
+ {
1601
+ "loss": 0.8707,
1602
+ "grad_norm": 1.7537908554077148,
1603
+ "learning_rate": 0.00018058211174845973,
1604
+ "epoch": 8.374760994263863,
1605
+ "step": 4380
1606
+ },
1607
+ {
1608
+ "loss": 0.8375,
1609
+ "grad_norm": 1.635578989982605,
1610
+ "learning_rate": 0.0001763331208837901,
1611
+ "epoch": 8.413001912045889,
1612
+ "step": 4400
1613
+ },
1614
+ {
1615
+ "loss": 0.8631,
1616
+ "grad_norm": 1.5729222297668457,
1617
+ "learning_rate": 0.00017208413001912047,
1618
+ "epoch": 8.451242829827915,
1619
+ "step": 4420
1620
+ },
1621
+ {
1622
+ "loss": 0.8551,
1623
+ "grad_norm": 1.6586476564407349,
1624
+ "learning_rate": 0.00016783513915445082,
1625
+ "epoch": 8.489483747609942,
1626
+ "step": 4440
1627
+ },
1628
+ {
1629
+ "loss": 0.8637,
1630
+ "grad_norm": 1.6118619441986084,
1631
+ "learning_rate": 0.00016358614828978119,
1632
+ "epoch": 8.52772466539197,
1633
+ "step": 4460
1634
+ },
1635
+ {
1636
+ "loss": 0.8484,
1637
+ "grad_norm": 1.5538595914840698,
1638
+ "learning_rate": 0.00015933715742511153,
1639
+ "epoch": 8.565965583173996,
1640
+ "step": 4480
1641
+ },
1642
+ {
1643
+ "loss": 0.8433,
1644
+ "grad_norm": 1.5646642446517944,
1645
+ "learning_rate": 0.0001550881665604419,
1646
+ "epoch": 8.604206500956023,
1647
+ "step": 4500
1648
+ },
1649
+ {
1650
+ "loss": 0.8592,
1651
+ "grad_norm": 1.7190415859222412,
1652
+ "learning_rate": 0.00015083917569577227,
1653
+ "epoch": 8.64244741873805,
1654
+ "step": 4520
1655
+ },
1656
+ {
1657
+ "loss": 0.8236,
1658
+ "grad_norm": 1.4950307607650757,
1659
+ "learning_rate": 0.00014659018483110262,
1660
+ "epoch": 8.680688336520076,
1661
+ "step": 4540
1662
+ },
1663
+ {
1664
+ "loss": 0.8421,
1665
+ "grad_norm": 1.5117732286453247,
1666
+ "learning_rate": 0.00014234119396643296,
1667
+ "epoch": 8.718929254302104,
1668
+ "step": 4560
1669
+ },
1670
+ {
1671
+ "loss": 0.8287,
1672
+ "grad_norm": 1.5558750629425049,
1673
+ "learning_rate": 0.00013809220310176336,
1674
+ "epoch": 8.75717017208413,
1675
+ "step": 4580
1676
+ },
1677
+ {
1678
+ "loss": 0.8492,
1679
+ "grad_norm": 1.7955564260482788,
1680
+ "learning_rate": 0.0001338432122370937,
1681
+ "epoch": 8.795411089866157,
1682
+ "step": 4600
1683
+ },
1684
+ {
1685
+ "loss": 0.8419,
1686
+ "grad_norm": 1.6532599925994873,
1687
+ "learning_rate": 0.00012959422137242405,
1688
+ "epoch": 8.833652007648183,
1689
+ "step": 4620
1690
+ },
1691
+ {
1692
+ "loss": 0.8125,
1693
+ "grad_norm": 1.7040739059448242,
1694
+ "learning_rate": 0.0001253452305077544,
1695
+ "epoch": 8.87189292543021,
1696
+ "step": 4640
1697
+ },
1698
+ {
1699
+ "loss": 0.8187,
1700
+ "grad_norm": 1.7040703296661377,
1701
+ "learning_rate": 0.00012109623964308478,
1702
+ "epoch": 8.910133843212238,
1703
+ "step": 4660
1704
+ },
1705
+ {
1706
+ "loss": 0.8155,
1707
+ "grad_norm": 1.7090845108032227,
1708
+ "learning_rate": 0.00011684724877841513,
1709
+ "epoch": 8.948374760994264,
1710
+ "step": 4680
1711
+ },
1712
+ {
1713
+ "loss": 0.7997,
1714
+ "grad_norm": 1.6070616245269775,
1715
+ "learning_rate": 0.00011259825791374549,
1716
+ "epoch": 8.98661567877629,
1717
+ "step": 4700
1718
+ },
1719
+ {
1720
+ "eval_loss": 0.3138497769832611,
1721
+ "eval_accuracy": 0.9352842246888665,
1722
+ "eval_runtime": 177.561,
1723
+ "eval_samples_per_second": 83.718,
1724
+ "eval_steps_per_second": 83.718,
1725
+ "epoch": 9.0,
1726
+ "step": 4707
1727
+ },
1728
+ {
1729
+ "loss": 0.7906,
1730
+ "grad_norm": 1.5590825080871582,
1731
+ "learning_rate": 0.00010834926704907585,
1732
+ "epoch": 9.024856596558317,
1733
+ "step": 4720
1734
+ },
1735
+ {
1736
+ "loss": 0.7494,
1737
+ "grad_norm": 1.4745252132415771,
1738
+ "learning_rate": 0.0001041002761844062,
1739
+ "epoch": 9.063097514340344,
1740
+ "step": 4740
1741
+ },
1742
+ {
1743
+ "loss": 0.7854,
1744
+ "grad_norm": 1.61099112033844,
1745
+ "learning_rate": 9.985128531973658e-05,
1746
+ "epoch": 9.101338432122372,
1747
+ "step": 4760
1748
+ },
1749
+ {
1750
+ "loss": 0.7636,
1751
+ "grad_norm": 1.5839650630950928,
1752
+ "learning_rate": 9.560229445506692e-05,
1753
+ "epoch": 9.139579349904398,
1754
+ "step": 4780
1755
+ },
1756
+ {
1757
+ "loss": 0.776,
1758
+ "grad_norm": 1.7259138822555542,
1759
+ "learning_rate": 9.135330359039729e-05,
1760
+ "epoch": 9.177820267686425,
1761
+ "step": 4800
1762
+ },
1763
+ {
1764
+ "loss": 0.7444,
1765
+ "grad_norm": 1.5495970249176025,
1766
+ "learning_rate": 8.710431272572764e-05,
1767
+ "epoch": 9.216061185468451,
1768
+ "step": 4820
1769
+ },
1770
+ {
1771
+ "loss": 0.7603,
1772
+ "grad_norm": 1.5250838994979858,
1773
+ "learning_rate": 8.2855321861058e-05,
1774
+ "epoch": 9.254302103250478,
1775
+ "step": 4840
1776
+ },
1777
+ {
1778
+ "loss": 0.7561,
1779
+ "grad_norm": 1.6244220733642578,
1780
+ "learning_rate": 7.860633099638836e-05,
1781
+ "epoch": 9.292543021032504,
1782
+ "step": 4860
1783
+ },
1784
+ {
1785
+ "loss": 0.7908,
1786
+ "grad_norm": 1.6825993061065674,
1787
+ "learning_rate": 7.435734013171871e-05,
1788
+ "epoch": 9.330783938814532,
1789
+ "step": 4880
1790
+ },
1791
+ {
1792
+ "loss": 0.7517,
1793
+ "grad_norm": 1.563707947731018,
1794
+ "learning_rate": 7.010834926704908e-05,
1795
+ "epoch": 9.369024856596559,
1796
+ "step": 4900
1797
+ },
1798
+ {
1799
+ "loss": 0.7679,
1800
+ "grad_norm": 1.7463629245758057,
1801
+ "learning_rate": 6.585935840237942e-05,
1802
+ "epoch": 9.407265774378585,
1803
+ "step": 4920
1804
+ },
1805
+ {
1806
+ "loss": 0.7426,
1807
+ "grad_norm": 1.5689053535461426,
1808
+ "learning_rate": 6.16103675377098e-05,
1809
+ "epoch": 9.445506692160611,
1810
+ "step": 4940
1811
+ },
1812
+ {
1813
+ "loss": 0.7695,
1814
+ "grad_norm": 1.6512914896011353,
1815
+ "learning_rate": 5.736137667304015e-05,
1816
+ "epoch": 9.483747609942638,
1817
+ "step": 4960
1818
+ },
1819
+ {
1820
+ "loss": 0.7603,
1821
+ "grad_norm": 1.6542084217071533,
1822
+ "learning_rate": 5.311238580837052e-05,
1823
+ "epoch": 9.521988527724666,
1824
+ "step": 4980
1825
+ },
1826
+ {
1827
+ "loss": 0.754,
1828
+ "grad_norm": 1.6929945945739746,
1829
+ "learning_rate": 4.8863394943700874e-05,
1830
+ "epoch": 9.560229445506693,
1831
+ "step": 5000
1832
+ },
1833
+ {
1834
+ "loss": 0.7597,
1835
+ "grad_norm": 1.4880517721176147,
1836
+ "learning_rate": 4.461440407903123e-05,
1837
+ "epoch": 9.598470363288719,
1838
+ "step": 5020
1839
+ },
1840
+ {
1841
+ "loss": 0.7624,
1842
+ "grad_norm": 1.578971266746521,
1843
+ "learning_rate": 4.036541321436159e-05,
1844
+ "epoch": 9.636711281070745,
1845
+ "step": 5040
1846
+ },
1847
+ {
1848
+ "loss": 0.7653,
1849
+ "grad_norm": 1.616727352142334,
1850
+ "learning_rate": 3.6116422349691954e-05,
1851
+ "epoch": 9.674952198852772,
1852
+ "step": 5060
1853
+ },
1854
+ {
1855
+ "loss": 0.7352,
1856
+ "grad_norm": 1.6762784719467163,
1857
+ "learning_rate": 3.186743148502231e-05,
1858
+ "epoch": 9.7131931166348,
1859
+ "step": 5080
1860
+ },
1861
+ {
1862
+ "loss": 0.732,
1863
+ "grad_norm": 1.5666388273239136,
1864
+ "learning_rate": 2.7618440620352666e-05,
1865
+ "epoch": 9.751434034416826,
1866
+ "step": 5100
1867
+ },
1868
+ {
1869
+ "loss": 0.7631,
1870
+ "grad_norm": 1.641012191772461,
1871
+ "learning_rate": 2.3369449755683023e-05,
1872
+ "epoch": 9.789674952198853,
1873
+ "step": 5120
1874
+ },
1875
+ {
1876
+ "loss": 0.7153,
1877
+ "grad_norm": 1.7024327516555786,
1878
+ "learning_rate": 1.9120458891013384e-05,
1879
+ "epoch": 9.82791586998088,
1880
+ "step": 5140
1881
+ },
1882
+ {
1883
+ "loss": 0.7247,
1884
+ "grad_norm": 1.4840829372406006,
1885
+ "learning_rate": 1.4871468026343743e-05,
1886
+ "epoch": 9.866156787762906,
1887
+ "step": 5160
1888
+ },
1889
+ {
1890
+ "loss": 0.7289,
1891
+ "grad_norm": 1.627562165260315,
1892
+ "learning_rate": 1.0622477161674103e-05,
1893
+ "epoch": 9.904397705544934,
1894
+ "step": 5180
1895
+ },
1896
+ {
1897
+ "loss": 0.7563,
1898
+ "grad_norm": 1.6473079919815063,
1899
+ "learning_rate": 6.373486297004461e-06,
1900
+ "epoch": 9.94263862332696,
1901
+ "step": 5200
1902
+ },
1903
+ {
1904
+ "loss": 0.7603,
1905
+ "grad_norm": 1.577776312828064,
1906
+ "learning_rate": 2.1244954323348204e-06,
1907
+ "epoch": 9.980879541108987,
1908
+ "step": 5220
1909
+ },
1910
+ {
1911
+ "eval_loss": 0.29460111260414124,
1912
+ "eval_accuracy": 0.9410023545240498,
1913
+ "eval_runtime": 187.2975,
1914
+ "eval_samples_per_second": 79.366,
1915
+ "eval_steps_per_second": 79.366,
1916
+ "epoch": 10.0,
1917
+ "step": 5230
1918
+ },
1919
+ {
1920
+ "train_runtime": 19361.582,
1921
+ "train_samples_per_second": 69.094,
1922
+ "train_steps_per_second": 0.27,
1923
+ "total_flos": 1.96318398191328e+18,
1924
+ "train_loss": 2.1099621225725396,
1925
+ "epoch": 10.0,
1926
+ "step": 5230
1927
+ }
1928
+ ]
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c601d4d283f57652bdd3d990eb54fbc23f8a75f6f27a6e12c17e62505fcbb90e
3
  size 20450108
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6442360cd0b5265302135afde5e0c51d63a9ba0335457a66bb0aab644e8a6b72
3
  size 20450108
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 1.96318398191328e+18,
4
+ "train_loss": 2.1099621225725396,
5
+ "train_runtime": 19361.582,
6
+ "train_samples_per_second": 69.094,
7
+ "train_steps_per_second": 0.27
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1959 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9410023545240498,
3
+ "best_model_checkpoint": "/mnt/data4_HDD_14TB/yang/voxceleb-checkpoints/xvector/voxceleb1/finetune/ce-len3-bs256-lr1e-3/checkpoint-5230",
4
+ "epoch": 10.0,
5
+ "eval_steps": 500,
6
+ "global_step": 5230,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03824091778202677,
13
+ "grad_norm": 4.7170538902282715,
14
+ "learning_rate": 3.824091778202677e-05,
15
+ "loss": 7.1456,
16
+ "step": 20
17
+ },
18
+ {
19
+ "epoch": 0.07648183556405354,
20
+ "grad_norm": 4.5134358406066895,
21
+ "learning_rate": 7.648183556405354e-05,
22
+ "loss": 7.1158,
23
+ "step": 40
24
+ },
25
+ {
26
+ "epoch": 0.1147227533460803,
27
+ "grad_norm": 3.9098806381225586,
28
+ "learning_rate": 0.0001147227533460803,
29
+ "loss": 7.0575,
30
+ "step": 60
31
+ },
32
+ {
33
+ "epoch": 0.15296367112810708,
34
+ "grad_norm": 3.2200050354003906,
35
+ "learning_rate": 0.00015296367112810707,
36
+ "loss": 6.9743,
37
+ "step": 80
38
+ },
39
+ {
40
+ "epoch": 0.19120458891013384,
41
+ "grad_norm": 2.5479934215545654,
42
+ "learning_rate": 0.00019120458891013384,
43
+ "loss": 6.874,
44
+ "step": 100
45
+ },
46
+ {
47
+ "epoch": 0.2294455066921606,
48
+ "grad_norm": 2.168301820755005,
49
+ "learning_rate": 0.0002294455066921606,
50
+ "loss": 6.7278,
51
+ "step": 120
52
+ },
53
+ {
54
+ "epoch": 0.2676864244741874,
55
+ "grad_norm": 2.0083394050598145,
56
+ "learning_rate": 0.0002676864244741874,
57
+ "loss": 6.5778,
58
+ "step": 140
59
+ },
60
+ {
61
+ "epoch": 0.30592734225621415,
62
+ "grad_norm": 1.8298897743225098,
63
+ "learning_rate": 0.00030592734225621415,
64
+ "loss": 6.4192,
65
+ "step": 160
66
+ },
67
+ {
68
+ "epoch": 0.3441682600382409,
69
+ "grad_norm": 1.782423734664917,
70
+ "learning_rate": 0.00034416826003824094,
71
+ "loss": 6.2416,
72
+ "step": 180
73
+ },
74
+ {
75
+ "epoch": 0.3824091778202677,
76
+ "grad_norm": 1.8139146566390991,
77
+ "learning_rate": 0.0003824091778202677,
78
+ "loss": 6.0624,
79
+ "step": 200
80
+ },
81
+ {
82
+ "epoch": 0.42065009560229444,
83
+ "grad_norm": 1.7659958600997925,
84
+ "learning_rate": 0.0004206500956022944,
85
+ "loss": 5.9686,
86
+ "step": 220
87
+ },
88
+ {
89
+ "epoch": 0.4588910133843212,
90
+ "grad_norm": 1.8660094738006592,
91
+ "learning_rate": 0.0004588910133843212,
92
+ "loss": 5.8142,
93
+ "step": 240
94
+ },
95
+ {
96
+ "epoch": 0.497131931166348,
97
+ "grad_norm": 1.831566333770752,
98
+ "learning_rate": 0.0004971319311663481,
99
+ "loss": 5.6944,
100
+ "step": 260
101
+ },
102
+ {
103
+ "epoch": 0.5353728489483748,
104
+ "grad_norm": 1.7546241283416748,
105
+ "learning_rate": 0.0005353728489483748,
106
+ "loss": 5.6101,
107
+ "step": 280
108
+ },
109
+ {
110
+ "epoch": 0.5736137667304015,
111
+ "grad_norm": 1.8890600204467773,
112
+ "learning_rate": 0.0005736137667304016,
113
+ "loss": 5.5192,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 0.6118546845124283,
118
+ "grad_norm": 1.7542874813079834,
119
+ "learning_rate": 0.0006118546845124283,
120
+ "loss": 5.3822,
121
+ "step": 320
122
+ },
123
+ {
124
+ "epoch": 0.6500956022944551,
125
+ "grad_norm": 1.8762731552124023,
126
+ "learning_rate": 0.000650095602294455,
127
+ "loss": 5.3236,
128
+ "step": 340
129
+ },
130
+ {
131
+ "epoch": 0.6883365200764818,
132
+ "grad_norm": 1.886903166770935,
133
+ "learning_rate": 0.0006883365200764819,
134
+ "loss": 5.2483,
135
+ "step": 360
136
+ },
137
+ {
138
+ "epoch": 0.7265774378585086,
139
+ "grad_norm": 1.9873583316802979,
140
+ "learning_rate": 0.0007265774378585086,
141
+ "loss": 5.1245,
142
+ "step": 380
143
+ },
144
+ {
145
+ "epoch": 0.7648183556405354,
146
+ "grad_norm": 1.953506350517273,
147
+ "learning_rate": 0.0007648183556405354,
148
+ "loss": 5.0771,
149
+ "step": 400
150
+ },
151
+ {
152
+ "epoch": 0.8030592734225621,
153
+ "grad_norm": 1.851192831993103,
154
+ "learning_rate": 0.0008030592734225621,
155
+ "loss": 5.0354,
156
+ "step": 420
157
+ },
158
+ {
159
+ "epoch": 0.8413001912045889,
160
+ "grad_norm": 1.861971139907837,
161
+ "learning_rate": 0.0008413001912045888,
162
+ "loss": 4.9532,
163
+ "step": 440
164
+ },
165
+ {
166
+ "epoch": 0.8795411089866156,
167
+ "grad_norm": 1.9388970136642456,
168
+ "learning_rate": 0.0008795411089866157,
169
+ "loss": 4.8698,
170
+ "step": 460
171
+ },
172
+ {
173
+ "epoch": 0.9177820267686424,
174
+ "grad_norm": 1.919184684753418,
175
+ "learning_rate": 0.0009177820267686424,
176
+ "loss": 4.847,
177
+ "step": 480
178
+ },
179
+ {
180
+ "epoch": 0.9560229445506692,
181
+ "grad_norm": 1.8886795043945312,
182
+ "learning_rate": 0.0009560229445506692,
183
+ "loss": 4.7754,
184
+ "step": 500
185
+ },
186
+ {
187
+ "epoch": 0.994263862332696,
188
+ "grad_norm": 1.9832813739776611,
189
+ "learning_rate": 0.0009942638623326961,
190
+ "loss": 4.6728,
191
+ "step": 520
192
+ },
193
+ {
194
+ "epoch": 1.0,
195
+ "eval_accuracy": 0.15042045072317525,
196
+ "eval_loss": 4.345639228820801,
197
+ "eval_runtime": 211.2136,
198
+ "eval_samples_per_second": 70.379,
199
+ "eval_steps_per_second": 70.379,
200
+ "step": 523
201
+ },
202
+ {
203
+ "epoch": 1.0325047801147227,
204
+ "grad_norm": 1.831260323524475,
205
+ "learning_rate": 0.0009963883577650309,
206
+ "loss": 4.5657,
207
+ "step": 540
208
+ },
209
+ {
210
+ "epoch": 1.0707456978967496,
211
+ "grad_norm": 1.934414029121399,
212
+ "learning_rate": 0.0009921393669003612,
213
+ "loss": 4.5047,
214
+ "step": 560
215
+ },
216
+ {
217
+ "epoch": 1.1089866156787762,
218
+ "grad_norm": 1.7718721628189087,
219
+ "learning_rate": 0.0009878903760356915,
220
+ "loss": 4.4165,
221
+ "step": 580
222
+ },
223
+ {
224
+ "epoch": 1.147227533460803,
225
+ "grad_norm": 1.741455078125,
226
+ "learning_rate": 0.0009836413851710218,
227
+ "loss": 4.3933,
228
+ "step": 600
229
+ },
230
+ {
231
+ "epoch": 1.1854684512428297,
232
+ "grad_norm": 1.8857481479644775,
233
+ "learning_rate": 0.0009793923943063523,
234
+ "loss": 4.3249,
235
+ "step": 620
236
+ },
237
+ {
238
+ "epoch": 1.2237093690248566,
239
+ "grad_norm": 1.8325748443603516,
240
+ "learning_rate": 0.0009751434034416827,
241
+ "loss": 4.2147,
242
+ "step": 640
243
+ },
244
+ {
245
+ "epoch": 1.2619502868068833,
246
+ "grad_norm": 1.8758591413497925,
247
+ "learning_rate": 0.000970894412577013,
248
+ "loss": 4.1569,
249
+ "step": 660
250
+ },
251
+ {
252
+ "epoch": 1.3001912045889101,
253
+ "grad_norm": 1.899542212486267,
254
+ "learning_rate": 0.0009666454217123433,
255
+ "loss": 4.131,
256
+ "step": 680
257
+ },
258
+ {
259
+ "epoch": 1.338432122370937,
260
+ "grad_norm": 1.8188538551330566,
261
+ "learning_rate": 0.0009623964308476737,
262
+ "loss": 4.0467,
263
+ "step": 700
264
+ },
265
+ {
266
+ "epoch": 1.3766730401529637,
267
+ "grad_norm": 1.7679705619812012,
268
+ "learning_rate": 0.000958147439983004,
269
+ "loss": 3.9904,
270
+ "step": 720
271
+ },
272
+ {
273
+ "epoch": 1.4149139579349903,
274
+ "grad_norm": 1.849482774734497,
275
+ "learning_rate": 0.0009538984491183344,
276
+ "loss": 3.9464,
277
+ "step": 740
278
+ },
279
+ {
280
+ "epoch": 1.4531548757170172,
281
+ "grad_norm": 1.8237632513046265,
282
+ "learning_rate": 0.0009496494582536647,
283
+ "loss": 3.9104,
284
+ "step": 760
285
+ },
286
+ {
287
+ "epoch": 1.491395793499044,
288
+ "grad_norm": 1.8175936937332153,
289
+ "learning_rate": 0.0009454004673889951,
290
+ "loss": 3.8441,
291
+ "step": 780
292
+ },
293
+ {
294
+ "epoch": 1.5296367112810707,
295
+ "grad_norm": 1.7967997789382935,
296
+ "learning_rate": 0.0009411514765243255,
297
+ "loss": 3.7898,
298
+ "step": 800
299
+ },
300
+ {
301
+ "epoch": 1.5678776290630974,
302
+ "grad_norm": 1.7681634426116943,
303
+ "learning_rate": 0.0009369024856596558,
304
+ "loss": 3.6894,
305
+ "step": 820
306
+ },
307
+ {
308
+ "epoch": 1.6061185468451242,
309
+ "grad_norm": 1.8655925989151,
310
+ "learning_rate": 0.0009326534947949862,
311
+ "loss": 3.6798,
312
+ "step": 840
313
+ },
314
+ {
315
+ "epoch": 1.644359464627151,
316
+ "grad_norm": 1.853769302368164,
317
+ "learning_rate": 0.0009284045039303166,
318
+ "loss": 3.6297,
319
+ "step": 860
320
+ },
321
+ {
322
+ "epoch": 1.682600382409178,
323
+ "grad_norm": 1.8198288679122925,
324
+ "learning_rate": 0.0009241555130656469,
325
+ "loss": 3.5592,
326
+ "step": 880
327
+ },
328
+ {
329
+ "epoch": 1.7208413001912046,
330
+ "grad_norm": 1.7744460105895996,
331
+ "learning_rate": 0.0009199065222009773,
332
+ "loss": 3.5056,
333
+ "step": 900
334
+ },
335
+ {
336
+ "epoch": 1.7590822179732313,
337
+ "grad_norm": 1.797914981842041,
338
+ "learning_rate": 0.0009156575313363077,
339
+ "loss": 3.4635,
340
+ "step": 920
341
+ },
342
+ {
343
+ "epoch": 1.7973231357552581,
344
+ "grad_norm": 1.8479169607162476,
345
+ "learning_rate": 0.000911408540471638,
346
+ "loss": 3.4434,
347
+ "step": 940
348
+ },
349
+ {
350
+ "epoch": 1.835564053537285,
351
+ "grad_norm": 1.818405032157898,
352
+ "learning_rate": 0.0009071595496069684,
353
+ "loss": 3.441,
354
+ "step": 960
355
+ },
356
+ {
357
+ "epoch": 1.8738049713193117,
358
+ "grad_norm": 1.7609572410583496,
359
+ "learning_rate": 0.0009029105587422988,
360
+ "loss": 3.3934,
361
+ "step": 980
362
+ },
363
+ {
364
+ "epoch": 1.9120458891013383,
365
+ "grad_norm": 1.7228211164474487,
366
+ "learning_rate": 0.0008986615678776291,
367
+ "loss": 3.2961,
368
+ "step": 1000
369
+ },
370
+ {
371
+ "epoch": 1.9502868068833652,
372
+ "grad_norm": 1.8148291110992432,
373
+ "learning_rate": 0.0008944125770129595,
374
+ "loss": 3.2611,
375
+ "step": 1020
376
+ },
377
+ {
378
+ "epoch": 1.988527724665392,
379
+ "grad_norm": 1.933300495147705,
380
+ "learning_rate": 0.0008901635861482899,
381
+ "loss": 3.224,
382
+ "step": 1040
383
+ },
384
+ {
385
+ "epoch": 2.0,
386
+ "eval_accuracy": 0.5140935082408342,
387
+ "eval_loss": 2.258894205093384,
388
+ "eval_runtime": 203.9361,
389
+ "eval_samples_per_second": 72.89,
390
+ "eval_steps_per_second": 72.89,
391
+ "step": 1046
392
+ },
393
+ {
394
+ "epoch": 2.026768642447419,
395
+ "grad_norm": 1.819346308708191,
396
+ "learning_rate": 0.0008859145952836202,
397
+ "loss": 3.1667,
398
+ "step": 1060
399
+ },
400
+ {
401
+ "epoch": 2.0650095602294454,
402
+ "grad_norm": 1.7024896144866943,
403
+ "learning_rate": 0.0008816656044189504,
404
+ "loss": 3.0232,
405
+ "step": 1080
406
+ },
407
+ {
408
+ "epoch": 2.1032504780114722,
409
+ "grad_norm": 1.7023948431015015,
410
+ "learning_rate": 0.000877416613554281,
411
+ "loss": 3.0489,
412
+ "step": 1100
413
+ },
414
+ {
415
+ "epoch": 2.141491395793499,
416
+ "grad_norm": 1.804140329360962,
417
+ "learning_rate": 0.0008731676226896112,
418
+ "loss": 2.9732,
419
+ "step": 1120
420
+ },
421
+ {
422
+ "epoch": 2.179732313575526,
423
+ "grad_norm": 1.7260992527008057,
424
+ "learning_rate": 0.0008689186318249415,
425
+ "loss": 2.9562,
426
+ "step": 1140
427
+ },
428
+ {
429
+ "epoch": 2.2179732313575524,
430
+ "grad_norm": 1.7970356941223145,
431
+ "learning_rate": 0.000864669640960272,
432
+ "loss": 2.8875,
433
+ "step": 1160
434
+ },
435
+ {
436
+ "epoch": 2.2562141491395793,
437
+ "grad_norm": 1.8579261302947998,
438
+ "learning_rate": 0.0008604206500956023,
439
+ "loss": 2.916,
440
+ "step": 1180
441
+ },
442
+ {
443
+ "epoch": 2.294455066921606,
444
+ "grad_norm": 1.852342128753662,
445
+ "learning_rate": 0.0008561716592309326,
446
+ "loss": 2.8963,
447
+ "step": 1200
448
+ },
449
+ {
450
+ "epoch": 2.332695984703633,
451
+ "grad_norm": 1.8845752477645874,
452
+ "learning_rate": 0.000851922668366263,
453
+ "loss": 2.8029,
454
+ "step": 1220
455
+ },
456
+ {
457
+ "epoch": 2.3709369024856595,
458
+ "grad_norm": 1.883952260017395,
459
+ "learning_rate": 0.0008476736775015934,
460
+ "loss": 2.8237,
461
+ "step": 1240
462
+ },
463
+ {
464
+ "epoch": 2.4091778202676863,
465
+ "grad_norm": 1.8383756875991821,
466
+ "learning_rate": 0.0008434246866369237,
467
+ "loss": 2.8473,
468
+ "step": 1260
469
+ },
470
+ {
471
+ "epoch": 2.447418738049713,
472
+ "grad_norm": 1.8900470733642578,
473
+ "learning_rate": 0.0008391756957722541,
474
+ "loss": 2.7722,
475
+ "step": 1280
476
+ },
477
+ {
478
+ "epoch": 2.48565965583174,
479
+ "grad_norm": 1.8097845315933228,
480
+ "learning_rate": 0.0008349267049075845,
481
+ "loss": 2.7584,
482
+ "step": 1300
483
+ },
484
+ {
485
+ "epoch": 2.5239005736137665,
486
+ "grad_norm": 1.7215895652770996,
487
+ "learning_rate": 0.0008306777140429148,
488
+ "loss": 2.7134,
489
+ "step": 1320
490
+ },
491
+ {
492
+ "epoch": 2.5621414913957934,
493
+ "grad_norm": 1.8249051570892334,
494
+ "learning_rate": 0.0008264287231782451,
495
+ "loss": 2.6531,
496
+ "step": 1340
497
+ },
498
+ {
499
+ "epoch": 2.6003824091778203,
500
+ "grad_norm": 1.8082237243652344,
501
+ "learning_rate": 0.0008221797323135756,
502
+ "loss": 2.6675,
503
+ "step": 1360
504
+ },
505
+ {
506
+ "epoch": 2.638623326959847,
507
+ "grad_norm": 1.7981261014938354,
508
+ "learning_rate": 0.0008179307414489059,
509
+ "loss": 2.5702,
510
+ "step": 1380
511
+ },
512
+ {
513
+ "epoch": 2.676864244741874,
514
+ "grad_norm": 1.6964036226272583,
515
+ "learning_rate": 0.0008136817505842362,
516
+ "loss": 2.6339,
517
+ "step": 1400
518
+ },
519
+ {
520
+ "epoch": 2.7151051625239004,
521
+ "grad_norm": 1.755050778388977,
522
+ "learning_rate": 0.0008094327597195667,
523
+ "loss": 2.5489,
524
+ "step": 1420
525
+ },
526
+ {
527
+ "epoch": 2.7533460803059273,
528
+ "grad_norm": 1.7242581844329834,
529
+ "learning_rate": 0.000805183768854897,
530
+ "loss": 2.5908,
531
+ "step": 1440
532
+ },
533
+ {
534
+ "epoch": 2.791586998087954,
535
+ "grad_norm": 1.819612741470337,
536
+ "learning_rate": 0.0008009347779902273,
537
+ "loss": 2.5143,
538
+ "step": 1460
539
+ },
540
+ {
541
+ "epoch": 2.8298279158699806,
542
+ "grad_norm": 1.7033363580703735,
543
+ "learning_rate": 0.0007966857871255578,
544
+ "loss": 2.4662,
545
+ "step": 1480
546
+ },
547
+ {
548
+ "epoch": 2.8680688336520075,
549
+ "grad_norm": 1.7662159204483032,
550
+ "learning_rate": 0.000792436796260888,
551
+ "loss": 2.4044,
552
+ "step": 1500
553
+ },
554
+ {
555
+ "epoch": 2.9063097514340344,
556
+ "grad_norm": 1.7460269927978516,
557
+ "learning_rate": 0.0007881878053962183,
558
+ "loss": 2.4636,
559
+ "step": 1520
560
+ },
561
+ {
562
+ "epoch": 2.9445506692160612,
563
+ "grad_norm": 1.8268380165100098,
564
+ "learning_rate": 0.0007839388145315488,
565
+ "loss": 2.3955,
566
+ "step": 1540
567
+ },
568
+ {
569
+ "epoch": 2.982791586998088,
570
+ "grad_norm": 1.796981930732727,
571
+ "learning_rate": 0.0007796898236668791,
572
+ "loss": 2.3964,
573
+ "step": 1560
574
+ },
575
+ {
576
+ "epoch": 3.0,
577
+ "eval_accuracy": 0.6835519677093844,
578
+ "eval_loss": 1.4662528038024902,
579
+ "eval_runtime": 419.2044,
580
+ "eval_samples_per_second": 35.46,
581
+ "eval_steps_per_second": 35.46,
582
+ "step": 1569
583
+ },
584
+ {
585
+ "epoch": 3.0210325047801145,
586
+ "grad_norm": 1.7852272987365723,
587
+ "learning_rate": 0.0007754408328022094,
588
+ "loss": 2.3174,
589
+ "step": 1580
590
+ },
591
+ {
592
+ "epoch": 3.0592734225621414,
593
+ "grad_norm": 1.8464534282684326,
594
+ "learning_rate": 0.0007711918419375399,
595
+ "loss": 2.2913,
596
+ "step": 1600
597
+ },
598
+ {
599
+ "epoch": 3.0975143403441683,
600
+ "grad_norm": 1.7783145904541016,
601
+ "learning_rate": 0.0007669428510728702,
602
+ "loss": 2.2856,
603
+ "step": 1620
604
+ },
605
+ {
606
+ "epoch": 3.135755258126195,
607
+ "grad_norm": 1.744454264640808,
608
+ "learning_rate": 0.0007626938602082005,
609
+ "loss": 2.2099,
610
+ "step": 1640
611
+ },
612
+ {
613
+ "epoch": 3.173996175908222,
614
+ "grad_norm": 1.8276797533035278,
615
+ "learning_rate": 0.0007584448693435309,
616
+ "loss": 2.23,
617
+ "step": 1660
618
+ },
619
+ {
620
+ "epoch": 3.2122370936902485,
621
+ "grad_norm": 1.8144315481185913,
622
+ "learning_rate": 0.0007541958784788613,
623
+ "loss": 2.1912,
624
+ "step": 1680
625
+ },
626
+ {
627
+ "epoch": 3.2504780114722753,
628
+ "grad_norm": 1.8499830961227417,
629
+ "learning_rate": 0.0007499468876141916,
630
+ "loss": 2.1818,
631
+ "step": 1700
632
+ },
633
+ {
634
+ "epoch": 3.288718929254302,
635
+ "grad_norm": 1.7623099088668823,
636
+ "learning_rate": 0.000745697896749522,
637
+ "loss": 2.1349,
638
+ "step": 1720
639
+ },
640
+ {
641
+ "epoch": 3.3269598470363286,
642
+ "grad_norm": 1.8180640935897827,
643
+ "learning_rate": 0.0007414489058848524,
644
+ "loss": 2.1055,
645
+ "step": 1740
646
+ },
647
+ {
648
+ "epoch": 3.3652007648183555,
649
+ "grad_norm": 1.8159964084625244,
650
+ "learning_rate": 0.0007371999150201827,
651
+ "loss": 2.1077,
652
+ "step": 1760
653
+ },
654
+ {
655
+ "epoch": 3.4034416826003824,
656
+ "grad_norm": 1.7902129888534546,
657
+ "learning_rate": 0.0007329509241555131,
658
+ "loss": 2.0999,
659
+ "step": 1780
660
+ },
661
+ {
662
+ "epoch": 3.4416826003824093,
663
+ "grad_norm": 1.7685898542404175,
664
+ "learning_rate": 0.0007287019332908435,
665
+ "loss": 2.1188,
666
+ "step": 1800
667
+ },
668
+ {
669
+ "epoch": 3.479923518164436,
670
+ "grad_norm": 1.758325219154358,
671
+ "learning_rate": 0.0007244529424261738,
672
+ "loss": 2.0956,
673
+ "step": 1820
674
+ },
675
+ {
676
+ "epoch": 3.5181644359464626,
677
+ "grad_norm": 1.7802537679672241,
678
+ "learning_rate": 0.0007202039515615042,
679
+ "loss": 2.0488,
680
+ "step": 1840
681
+ },
682
+ {
683
+ "epoch": 3.5564053537284894,
684
+ "grad_norm": 1.8220280408859253,
685
+ "learning_rate": 0.0007159549606968346,
686
+ "loss": 2.0776,
687
+ "step": 1860
688
+ },
689
+ {
690
+ "epoch": 3.5946462715105163,
691
+ "grad_norm": 1.8494378328323364,
692
+ "learning_rate": 0.0007117059698321649,
693
+ "loss": 2.0257,
694
+ "step": 1880
695
+ },
696
+ {
697
+ "epoch": 3.632887189292543,
698
+ "grad_norm": 1.719109296798706,
699
+ "learning_rate": 0.0007074569789674953,
700
+ "loss": 2.0332,
701
+ "step": 1900
702
+ },
703
+ {
704
+ "epoch": 3.67112810707457,
705
+ "grad_norm": 1.9517509937286377,
706
+ "learning_rate": 0.0007032079881028257,
707
+ "loss": 2.0052,
708
+ "step": 1920
709
+ },
710
+ {
711
+ "epoch": 3.7093690248565965,
712
+ "grad_norm": 1.7318940162658691,
713
+ "learning_rate": 0.0006989589972381559,
714
+ "loss": 1.9901,
715
+ "step": 1940
716
+ },
717
+ {
718
+ "epoch": 3.7476099426386233,
719
+ "grad_norm": 1.767015814781189,
720
+ "learning_rate": 0.0006947100063734863,
721
+ "loss": 1.9411,
722
+ "step": 1960
723
+ },
724
+ {
725
+ "epoch": 3.78585086042065,
726
+ "grad_norm": 1.761806607246399,
727
+ "learning_rate": 0.0006904610155088166,
728
+ "loss": 2.0048,
729
+ "step": 1980
730
+ },
731
+ {
732
+ "epoch": 3.8240917782026767,
733
+ "grad_norm": 1.7126002311706543,
734
+ "learning_rate": 0.000686212024644147,
735
+ "loss": 1.9312,
736
+ "step": 2000
737
+ },
738
+ {
739
+ "epoch": 3.8623326959847035,
740
+ "grad_norm": 1.7167437076568604,
741
+ "learning_rate": 0.0006819630337794774,
742
+ "loss": 1.9443,
743
+ "step": 2020
744
+ },
745
+ {
746
+ "epoch": 3.9005736137667304,
747
+ "grad_norm": 1.749881386756897,
748
+ "learning_rate": 0.0006777140429148077,
749
+ "loss": 1.893,
750
+ "step": 2040
751
+ },
752
+ {
753
+ "epoch": 3.9388145315487573,
754
+ "grad_norm": 1.6846592426300049,
755
+ "learning_rate": 0.0006734650520501381,
756
+ "loss": 1.8876,
757
+ "step": 2060
758
+ },
759
+ {
760
+ "epoch": 3.977055449330784,
761
+ "grad_norm": 1.8149057626724243,
762
+ "learning_rate": 0.0006692160611854685,
763
+ "loss": 1.8474,
764
+ "step": 2080
765
+ },
766
+ {
767
+ "epoch": 4.0,
768
+ "eval_accuracy": 0.7926673393878237,
769
+ "eval_loss": 0.9547563195228577,
770
+ "eval_runtime": 183.3001,
771
+ "eval_samples_per_second": 81.097,
772
+ "eval_steps_per_second": 81.097,
773
+ "step": 2092
774
+ },
775
+ {
776
+ "epoch": 4.015296367112811,
777
+ "grad_norm": 1.6455098390579224,
778
+ "learning_rate": 0.0006649670703207988,
779
+ "loss": 1.813,
780
+ "step": 2100
781
+ },
782
+ {
783
+ "epoch": 4.053537284894838,
784
+ "grad_norm": 1.6958200931549072,
785
+ "learning_rate": 0.0006607180794561292,
786
+ "loss": 1.7354,
787
+ "step": 2120
788
+ },
789
+ {
790
+ "epoch": 4.091778202676864,
791
+ "grad_norm": 1.7456037998199463,
792
+ "learning_rate": 0.0006564690885914596,
793
+ "loss": 1.7479,
794
+ "step": 2140
795
+ },
796
+ {
797
+ "epoch": 4.130019120458891,
798
+ "grad_norm": 1.7887734174728394,
799
+ "learning_rate": 0.0006522200977267899,
800
+ "loss": 1.7138,
801
+ "step": 2160
802
+ },
803
+ {
804
+ "epoch": 4.168260038240918,
805
+ "grad_norm": 1.7080284357070923,
806
+ "learning_rate": 0.0006479711068621203,
807
+ "loss": 1.7023,
808
+ "step": 2180
809
+ },
810
+ {
811
+ "epoch": 4.2065009560229445,
812
+ "grad_norm": 1.8061983585357666,
813
+ "learning_rate": 0.0006437221159974506,
814
+ "loss": 1.7526,
815
+ "step": 2200
816
+ },
817
+ {
818
+ "epoch": 4.244741873804971,
819
+ "grad_norm": 1.7831811904907227,
820
+ "learning_rate": 0.000639473125132781,
821
+ "loss": 1.7474,
822
+ "step": 2220
823
+ },
824
+ {
825
+ "epoch": 4.282982791586998,
826
+ "grad_norm": 1.752357840538025,
827
+ "learning_rate": 0.0006352241342681113,
828
+ "loss": 1.6688,
829
+ "step": 2240
830
+ },
831
+ {
832
+ "epoch": 4.321223709369025,
833
+ "grad_norm": 1.7843034267425537,
834
+ "learning_rate": 0.0006309751434034417,
835
+ "loss": 1.7009,
836
+ "step": 2260
837
+ },
838
+ {
839
+ "epoch": 4.359464627151052,
840
+ "grad_norm": 1.7608367204666138,
841
+ "learning_rate": 0.0006267261525387721,
842
+ "loss": 1.6727,
843
+ "step": 2280
844
+ },
845
+ {
846
+ "epoch": 4.397705544933078,
847
+ "grad_norm": 1.6877254247665405,
848
+ "learning_rate": 0.0006224771616741024,
849
+ "loss": 1.6801,
850
+ "step": 2300
851
+ },
852
+ {
853
+ "epoch": 4.435946462715105,
854
+ "grad_norm": 1.7891350984573364,
855
+ "learning_rate": 0.0006182281708094328,
856
+ "loss": 1.7108,
857
+ "step": 2320
858
+ },
859
+ {
860
+ "epoch": 4.474187380497132,
861
+ "grad_norm": 1.7104123830795288,
862
+ "learning_rate": 0.0006139791799447631,
863
+ "loss": 1.6442,
864
+ "step": 2340
865
+ },
866
+ {
867
+ "epoch": 4.512428298279159,
868
+ "grad_norm": 1.7026969194412231,
869
+ "learning_rate": 0.0006097301890800934,
870
+ "loss": 1.6531,
871
+ "step": 2360
872
+ },
873
+ {
874
+ "epoch": 4.550669216061186,
875
+ "grad_norm": 1.7890552282333374,
876
+ "learning_rate": 0.0006054811982154238,
877
+ "loss": 1.6539,
878
+ "step": 2380
879
+ },
880
+ {
881
+ "epoch": 4.588910133843212,
882
+ "grad_norm": 1.8423861265182495,
883
+ "learning_rate": 0.0006012322073507542,
884
+ "loss": 1.6681,
885
+ "step": 2400
886
+ },
887
+ {
888
+ "epoch": 4.627151051625239,
889
+ "grad_norm": 1.6434499025344849,
890
+ "learning_rate": 0.0005969832164860845,
891
+ "loss": 1.5935,
892
+ "step": 2420
893
+ },
894
+ {
895
+ "epoch": 4.665391969407266,
896
+ "grad_norm": 1.7261130809783936,
897
+ "learning_rate": 0.0005927342256214149,
898
+ "loss": 1.6273,
899
+ "step": 2440
900
+ },
901
+ {
902
+ "epoch": 4.7036328871892925,
903
+ "grad_norm": 1.7288273572921753,
904
+ "learning_rate": 0.0005884852347567453,
905
+ "loss": 1.6181,
906
+ "step": 2460
907
+ },
908
+ {
909
+ "epoch": 4.741873804971319,
910
+ "grad_norm": 1.773258924484253,
911
+ "learning_rate": 0.0005842362438920756,
912
+ "loss": 1.5719,
913
+ "step": 2480
914
+ },
915
+ {
916
+ "epoch": 4.780114722753346,
917
+ "grad_norm": 1.7676658630371094,
918
+ "learning_rate": 0.000579987253027406,
919
+ "loss": 1.578,
920
+ "step": 2500
921
+ },
922
+ {
923
+ "epoch": 4.818355640535373,
924
+ "grad_norm": 1.8115794658660889,
925
+ "learning_rate": 0.0005757382621627364,
926
+ "loss": 1.535,
927
+ "step": 2520
928
+ },
929
+ {
930
+ "epoch": 4.8565965583174,
931
+ "grad_norm": 1.7989414930343628,
932
+ "learning_rate": 0.0005714892712980667,
933
+ "loss": 1.5493,
934
+ "step": 2540
935
+ },
936
+ {
937
+ "epoch": 4.894837476099426,
938
+ "grad_norm": 1.6607849597930908,
939
+ "learning_rate": 0.000567240280433397,
940
+ "loss": 1.5489,
941
+ "step": 2560
942
+ },
943
+ {
944
+ "epoch": 4.933078393881453,
945
+ "grad_norm": 1.630257487297058,
946
+ "learning_rate": 0.0005629912895687275,
947
+ "loss": 1.5091,
948
+ "step": 2580
949
+ },
950
+ {
951
+ "epoch": 4.97131931166348,
952
+ "grad_norm": 1.7995944023132324,
953
+ "learning_rate": 0.0005587422987040578,
954
+ "loss": 1.5275,
955
+ "step": 2600
956
+ },
957
+ {
958
+ "epoch": 5.0,
959
+ "eval_accuracy": 0.8571140262361251,
960
+ "eval_loss": 0.6697778105735779,
961
+ "eval_runtime": 181.3784,
962
+ "eval_samples_per_second": 81.956,
963
+ "eval_steps_per_second": 81.956,
964
+ "step": 2615
965
+ },
966
+ {
967
+ "epoch": 5.009560229445507,
968
+ "grad_norm": 1.7868553400039673,
969
+ "learning_rate": 0.0005544933078393881,
970
+ "loss": 1.4774,
971
+ "step": 2620
972
+ },
973
+ {
974
+ "epoch": 5.047801147227533,
975
+ "grad_norm": 1.6380654573440552,
976
+ "learning_rate": 0.0005502443169747186,
977
+ "loss": 1.3955,
978
+ "step": 2640
979
+ },
980
+ {
981
+ "epoch": 5.08604206500956,
982
+ "grad_norm": 1.7844533920288086,
983
+ "learning_rate": 0.0005459953261100489,
984
+ "loss": 1.4414,
985
+ "step": 2660
986
+ },
987
+ {
988
+ "epoch": 5.124282982791587,
989
+ "grad_norm": 1.779080867767334,
990
+ "learning_rate": 0.0005417463352453792,
991
+ "loss": 1.3782,
992
+ "step": 2680
993
+ },
994
+ {
995
+ "epoch": 5.162523900573614,
996
+ "grad_norm": 1.741326928138733,
997
+ "learning_rate": 0.0005374973443807097,
998
+ "loss": 1.4152,
999
+ "step": 2700
1000
+ },
1001
+ {
1002
+ "epoch": 5.2007648183556405,
1003
+ "grad_norm": 1.7447401285171509,
1004
+ "learning_rate": 0.00053324835351604,
1005
+ "loss": 1.3996,
1006
+ "step": 2720
1007
+ },
1008
+ {
1009
+ "epoch": 5.239005736137667,
1010
+ "grad_norm": 1.8067736625671387,
1011
+ "learning_rate": 0.0005289993626513702,
1012
+ "loss": 1.4137,
1013
+ "step": 2740
1014
+ },
1015
+ {
1016
+ "epoch": 5.277246653919694,
1017
+ "grad_norm": 1.7393046617507935,
1018
+ "learning_rate": 0.0005247503717867008,
1019
+ "loss": 1.3937,
1020
+ "step": 2760
1021
+ },
1022
+ {
1023
+ "epoch": 5.315487571701721,
1024
+ "grad_norm": 1.756184458732605,
1025
+ "learning_rate": 0.000520501380922031,
1026
+ "loss": 1.3912,
1027
+ "step": 2780
1028
+ },
1029
+ {
1030
+ "epoch": 5.353728489483748,
1031
+ "grad_norm": 1.7133733034133911,
1032
+ "learning_rate": 0.0005162523900573613,
1033
+ "loss": 1.387,
1034
+ "step": 2800
1035
+ },
1036
+ {
1037
+ "epoch": 5.3919694072657744,
1038
+ "grad_norm": 1.6597713232040405,
1039
+ "learning_rate": 0.0005120033991926918,
1040
+ "loss": 1.3551,
1041
+ "step": 2820
1042
+ },
1043
+ {
1044
+ "epoch": 5.430210325047801,
1045
+ "grad_norm": 1.8462845087051392,
1046
+ "learning_rate": 0.0005077544083280221,
1047
+ "loss": 1.3557,
1048
+ "step": 2840
1049
+ },
1050
+ {
1051
+ "epoch": 5.468451242829828,
1052
+ "grad_norm": 1.6737143993377686,
1053
+ "learning_rate": 0.0005035054174633524,
1054
+ "loss": 1.3495,
1055
+ "step": 2860
1056
+ },
1057
+ {
1058
+ "epoch": 5.506692160611855,
1059
+ "grad_norm": 1.7071157693862915,
1060
+ "learning_rate": 0.0004992564265986828,
1061
+ "loss": 1.394,
1062
+ "step": 2880
1063
+ },
1064
+ {
1065
+ "epoch": 5.544933078393882,
1066
+ "grad_norm": 1.663072943687439,
1067
+ "learning_rate": 0.0004950074357340132,
1068
+ "loss": 1.3263,
1069
+ "step": 2900
1070
+ },
1071
+ {
1072
+ "epoch": 5.583173996175908,
1073
+ "grad_norm": 1.640093207359314,
1074
+ "learning_rate": 0.0004907584448693436,
1075
+ "loss": 1.3474,
1076
+ "step": 2920
1077
+ },
1078
+ {
1079
+ "epoch": 5.621414913957935,
1080
+ "grad_norm": 1.762568712234497,
1081
+ "learning_rate": 0.0004865094540046739,
1082
+ "loss": 1.3375,
1083
+ "step": 2940
1084
+ },
1085
+ {
1086
+ "epoch": 5.659655831739962,
1087
+ "grad_norm": 1.6714434623718262,
1088
+ "learning_rate": 0.00048226046314000425,
1089
+ "loss": 1.3218,
1090
+ "step": 2960
1091
+ },
1092
+ {
1093
+ "epoch": 5.6978967495219885,
1094
+ "grad_norm": 1.7594107389450073,
1095
+ "learning_rate": 0.0004780114722753346,
1096
+ "loss": 1.3008,
1097
+ "step": 2980
1098
+ },
1099
+ {
1100
+ "epoch": 5.736137667304015,
1101
+ "grad_norm": 1.6483973264694214,
1102
+ "learning_rate": 0.000473762481410665,
1103
+ "loss": 1.3331,
1104
+ "step": 3000
1105
+ },
1106
+ {
1107
+ "epoch": 5.774378585086042,
1108
+ "grad_norm": 1.7252651453018188,
1109
+ "learning_rate": 0.00046951349054599533,
1110
+ "loss": 1.2775,
1111
+ "step": 3020
1112
+ },
1113
+ {
1114
+ "epoch": 5.812619502868069,
1115
+ "grad_norm": 1.7860745191574097,
1116
+ "learning_rate": 0.0004652644996813257,
1117
+ "loss": 1.2747,
1118
+ "step": 3040
1119
+ },
1120
+ {
1121
+ "epoch": 5.850860420650095,
1122
+ "grad_norm": 1.749874234199524,
1123
+ "learning_rate": 0.0004610155088166561,
1124
+ "loss": 1.2946,
1125
+ "step": 3060
1126
+ },
1127
+ {
1128
+ "epoch": 5.8891013384321225,
1129
+ "grad_norm": 1.7197644710540771,
1130
+ "learning_rate": 0.0004567665179519864,
1131
+ "loss": 1.2849,
1132
+ "step": 3080
1133
+ },
1134
+ {
1135
+ "epoch": 5.927342256214149,
1136
+ "grad_norm": 1.6396132707595825,
1137
+ "learning_rate": 0.00045251752708731676,
1138
+ "loss": 1.2544,
1139
+ "step": 3100
1140
+ },
1141
+ {
1142
+ "epoch": 5.965583173996176,
1143
+ "grad_norm": 1.720376968383789,
1144
+ "learning_rate": 0.0004482685362226471,
1145
+ "loss": 1.248,
1146
+ "step": 3120
1147
+ },
1148
+ {
1149
+ "epoch": 6.0,
1150
+ "eval_accuracy": 0.8899428187016482,
1151
+ "eval_loss": 0.5270123481750488,
1152
+ "eval_runtime": 183.7621,
1153
+ "eval_samples_per_second": 80.893,
1154
+ "eval_steps_per_second": 80.893,
1155
+ "step": 3138
1156
+ },
1157
+ {
1158
+ "epoch": 6.003824091778203,
1159
+ "grad_norm": 1.5206599235534668,
1160
+ "learning_rate": 0.0004440195453579775,
1161
+ "loss": 1.2398,
1162
+ "step": 3140
1163
+ },
1164
+ {
1165
+ "epoch": 6.042065009560229,
1166
+ "grad_norm": 1.7172082662582397,
1167
+ "learning_rate": 0.00043977055449330785,
1168
+ "loss": 1.2097,
1169
+ "step": 3160
1170
+ },
1171
+ {
1172
+ "epoch": 6.080305927342256,
1173
+ "grad_norm": 1.5570909976959229,
1174
+ "learning_rate": 0.0004355215636286382,
1175
+ "loss": 1.1669,
1176
+ "step": 3180
1177
+ },
1178
+ {
1179
+ "epoch": 6.118546845124283,
1180
+ "grad_norm": 1.7044614553451538,
1181
+ "learning_rate": 0.0004312725727639686,
1182
+ "loss": 1.1647,
1183
+ "step": 3200
1184
+ },
1185
+ {
1186
+ "epoch": 6.15678776290631,
1187
+ "grad_norm": 1.5819571018218994,
1188
+ "learning_rate": 0.0004270235818992989,
1189
+ "loss": 1.1627,
1190
+ "step": 3220
1191
+ },
1192
+ {
1193
+ "epoch": 6.195028680688337,
1194
+ "grad_norm": 1.7076871395111084,
1195
+ "learning_rate": 0.0004227745910346293,
1196
+ "loss": 1.1728,
1197
+ "step": 3240
1198
+ },
1199
+ {
1200
+ "epoch": 6.233269598470363,
1201
+ "grad_norm": 1.7301490306854248,
1202
+ "learning_rate": 0.0004185256001699597,
1203
+ "loss": 1.1459,
1204
+ "step": 3260
1205
+ },
1206
+ {
1207
+ "epoch": 6.27151051625239,
1208
+ "grad_norm": 1.7135626077651978,
1209
+ "learning_rate": 0.00041427660930528997,
1210
+ "loss": 1.1676,
1211
+ "step": 3280
1212
+ },
1213
+ {
1214
+ "epoch": 6.309751434034417,
1215
+ "grad_norm": 1.602142572402954,
1216
+ "learning_rate": 0.00041002761844062037,
1217
+ "loss": 1.1488,
1218
+ "step": 3300
1219
+ },
1220
+ {
1221
+ "epoch": 6.347992351816444,
1222
+ "grad_norm": 1.755293846130371,
1223
+ "learning_rate": 0.00040577862757595076,
1224
+ "loss": 1.1395,
1225
+ "step": 3320
1226
+ },
1227
+ {
1228
+ "epoch": 6.3862332695984705,
1229
+ "grad_norm": 1.663662314414978,
1230
+ "learning_rate": 0.00040152963671128105,
1231
+ "loss": 1.1321,
1232
+ "step": 3340
1233
+ },
1234
+ {
1235
+ "epoch": 6.424474187380497,
1236
+ "grad_norm": 1.7366993427276611,
1237
+ "learning_rate": 0.00039728064584661145,
1238
+ "loss": 1.1317,
1239
+ "step": 3360
1240
+ },
1241
+ {
1242
+ "epoch": 6.462715105162524,
1243
+ "grad_norm": 1.7560149431228638,
1244
+ "learning_rate": 0.0003930316549819418,
1245
+ "loss": 1.1449,
1246
+ "step": 3380
1247
+ },
1248
+ {
1249
+ "epoch": 6.500956022944551,
1250
+ "grad_norm": 1.7576582431793213,
1251
+ "learning_rate": 0.00038878266411727214,
1252
+ "loss": 1.13,
1253
+ "step": 3400
1254
+ },
1255
+ {
1256
+ "epoch": 6.539196940726577,
1257
+ "grad_norm": 1.7916873693466187,
1258
+ "learning_rate": 0.00038453367325260254,
1259
+ "loss": 1.1419,
1260
+ "step": 3420
1261
+ },
1262
+ {
1263
+ "epoch": 6.577437858508604,
1264
+ "grad_norm": 1.5987508296966553,
1265
+ "learning_rate": 0.0003802846823879329,
1266
+ "loss": 1.1107,
1267
+ "step": 3440
1268
+ },
1269
+ {
1270
+ "epoch": 6.615678776290631,
1271
+ "grad_norm": 1.8192518949508667,
1272
+ "learning_rate": 0.0003760356915232632,
1273
+ "loss": 1.1162,
1274
+ "step": 3460
1275
+ },
1276
+ {
1277
+ "epoch": 6.653919694072657,
1278
+ "grad_norm": 1.7236486673355103,
1279
+ "learning_rate": 0.0003717867006585936,
1280
+ "loss": 1.1255,
1281
+ "step": 3480
1282
+ },
1283
+ {
1284
+ "epoch": 6.692160611854685,
1285
+ "grad_norm": 1.8209389448165894,
1286
+ "learning_rate": 0.0003675377097939239,
1287
+ "loss": 1.0629,
1288
+ "step": 3500
1289
+ },
1290
+ {
1291
+ "epoch": 6.730401529636711,
1292
+ "grad_norm": 1.652782678604126,
1293
+ "learning_rate": 0.0003632887189292543,
1294
+ "loss": 1.0809,
1295
+ "step": 3520
1296
+ },
1297
+ {
1298
+ "epoch": 6.768642447418738,
1299
+ "grad_norm": 1.6148645877838135,
1300
+ "learning_rate": 0.00035903972806458466,
1301
+ "loss": 1.1286,
1302
+ "step": 3540
1303
+ },
1304
+ {
1305
+ "epoch": 6.806883365200765,
1306
+ "grad_norm": 1.6869423389434814,
1307
+ "learning_rate": 0.000354790737199915,
1308
+ "loss": 1.1069,
1309
+ "step": 3560
1310
+ },
1311
+ {
1312
+ "epoch": 6.845124282982791,
1313
+ "grad_norm": 1.6373172998428345,
1314
+ "learning_rate": 0.0003505417463352454,
1315
+ "loss": 1.0911,
1316
+ "step": 3580
1317
+ },
1318
+ {
1319
+ "epoch": 6.8833652007648185,
1320
+ "grad_norm": 1.6761549711227417,
1321
+ "learning_rate": 0.00034629275547057574,
1322
+ "loss": 1.0808,
1323
+ "step": 3600
1324
+ },
1325
+ {
1326
+ "epoch": 6.921606118546845,
1327
+ "grad_norm": 1.6510460376739502,
1328
+ "learning_rate": 0.0003420437646059061,
1329
+ "loss": 1.0809,
1330
+ "step": 3620
1331
+ },
1332
+ {
1333
+ "epoch": 6.959847036328872,
1334
+ "grad_norm": 1.7351855039596558,
1335
+ "learning_rate": 0.0003377947737412365,
1336
+ "loss": 1.0912,
1337
+ "step": 3640
1338
+ },
1339
+ {
1340
+ "epoch": 6.998087954110899,
1341
+ "grad_norm": 1.7165274620056152,
1342
+ "learning_rate": 0.00033354578287656683,
1343
+ "loss": 1.0991,
1344
+ "step": 3660
1345
+ },
1346
+ {
1347
+ "epoch": 7.0,
1348
+ "eval_accuracy": 0.9037336024217961,
1349
+ "eval_loss": 0.44995447993278503,
1350
+ "eval_runtime": 189.5634,
1351
+ "eval_samples_per_second": 78.417,
1352
+ "eval_steps_per_second": 78.417,
1353
+ "step": 3661
1354
+ },
1355
+ {
1356
+ "epoch": 7.036328871892925,
1357
+ "grad_norm": 1.6468501091003418,
1358
+ "learning_rate": 0.0003292967920118972,
1359
+ "loss": 1.0154,
1360
+ "step": 3680
1361
+ },
1362
+ {
1363
+ "epoch": 7.074569789674952,
1364
+ "grad_norm": 1.79421067237854,
1365
+ "learning_rate": 0.0003250478011472275,
1366
+ "loss": 1.0378,
1367
+ "step": 3700
1368
+ },
1369
+ {
1370
+ "epoch": 7.112810707456979,
1371
+ "grad_norm": 1.7234885692596436,
1372
+ "learning_rate": 0.0003207988102825579,
1373
+ "loss": 1.0145,
1374
+ "step": 3720
1375
+ },
1376
+ {
1377
+ "epoch": 7.151051625239006,
1378
+ "grad_norm": 1.6947157382965088,
1379
+ "learning_rate": 0.00031654981941788826,
1380
+ "loss": 1.0012,
1381
+ "step": 3740
1382
+ },
1383
+ {
1384
+ "epoch": 7.189292543021033,
1385
+ "grad_norm": 1.6818758249282837,
1386
+ "learning_rate": 0.0003123008285532186,
1387
+ "loss": 1.0191,
1388
+ "step": 3760
1389
+ },
1390
+ {
1391
+ "epoch": 7.227533460803059,
1392
+ "grad_norm": 1.557080864906311,
1393
+ "learning_rate": 0.000308051837688549,
1394
+ "loss": 1.0437,
1395
+ "step": 3780
1396
+ },
1397
+ {
1398
+ "epoch": 7.265774378585086,
1399
+ "grad_norm": 1.6532793045043945,
1400
+ "learning_rate": 0.00030380284682387935,
1401
+ "loss": 1.0079,
1402
+ "step": 3800
1403
+ },
1404
+ {
1405
+ "epoch": 7.304015296367113,
1406
+ "grad_norm": 1.646686315536499,
1407
+ "learning_rate": 0.0002995538559592097,
1408
+ "loss": 0.9904,
1409
+ "step": 3820
1410
+ },
1411
+ {
1412
+ "epoch": 7.342256214149139,
1413
+ "grad_norm": 1.6772829294204712,
1414
+ "learning_rate": 0.0002953048650945401,
1415
+ "loss": 1.0002,
1416
+ "step": 3840
1417
+ },
1418
+ {
1419
+ "epoch": 7.3804971319311665,
1420
+ "grad_norm": 1.6452054977416992,
1421
+ "learning_rate": 0.0002910558742298704,
1422
+ "loss": 0.9674,
1423
+ "step": 3860
1424
+ },
1425
+ {
1426
+ "epoch": 7.418738049713193,
1427
+ "grad_norm": 1.592207908630371,
1428
+ "learning_rate": 0.0002868068833652008,
1429
+ "loss": 0.9642,
1430
+ "step": 3880
1431
+ },
1432
+ {
1433
+ "epoch": 7.45697896749522,
1434
+ "grad_norm": 1.7015941143035889,
1435
+ "learning_rate": 0.0002825578925005312,
1436
+ "loss": 0.973,
1437
+ "step": 3900
1438
+ },
1439
+ {
1440
+ "epoch": 7.495219885277247,
1441
+ "grad_norm": 1.6589232683181763,
1442
+ "learning_rate": 0.00027830890163586146,
1443
+ "loss": 0.9803,
1444
+ "step": 3920
1445
+ },
1446
+ {
1447
+ "epoch": 7.533460803059273,
1448
+ "grad_norm": 1.660190463066101,
1449
+ "learning_rate": 0.00027405991077119186,
1450
+ "loss": 0.9702,
1451
+ "step": 3940
1452
+ },
1453
+ {
1454
+ "epoch": 7.5717017208413,
1455
+ "grad_norm": 1.7052509784698486,
1456
+ "learning_rate": 0.00026981091990652226,
1457
+ "loss": 0.9919,
1458
+ "step": 3960
1459
+ },
1460
+ {
1461
+ "epoch": 7.609942638623327,
1462
+ "grad_norm": 1.6874445676803589,
1463
+ "learning_rate": 0.00026556192904185255,
1464
+ "loss": 0.938,
1465
+ "step": 3980
1466
+ },
1467
+ {
1468
+ "epoch": 7.648183556405353,
1469
+ "grad_norm": 1.811640739440918,
1470
+ "learning_rate": 0.00026131293817718295,
1471
+ "loss": 0.9564,
1472
+ "step": 4000
1473
+ },
1474
+ {
1475
+ "epoch": 7.686424474187381,
1476
+ "grad_norm": 1.741968035697937,
1477
+ "learning_rate": 0.00025706394731251324,
1478
+ "loss": 0.9432,
1479
+ "step": 4020
1480
+ },
1481
+ {
1482
+ "epoch": 7.724665391969407,
1483
+ "grad_norm": 1.6731518507003784,
1484
+ "learning_rate": 0.00025281495644784364,
1485
+ "loss": 0.9082,
1486
+ "step": 4040
1487
+ },
1488
+ {
1489
+ "epoch": 7.762906309751434,
1490
+ "grad_norm": 1.7399870157241821,
1491
+ "learning_rate": 0.00024856596558317403,
1492
+ "loss": 0.92,
1493
+ "step": 4060
1494
+ },
1495
+ {
1496
+ "epoch": 7.801147227533461,
1497
+ "grad_norm": 1.580674171447754,
1498
+ "learning_rate": 0.0002443169747185044,
1499
+ "loss": 0.9433,
1500
+ "step": 4080
1501
+ },
1502
+ {
1503
+ "epoch": 7.839388145315487,
1504
+ "grad_norm": 1.5683550834655762,
1505
+ "learning_rate": 0.00024006798385383472,
1506
+ "loss": 0.9584,
1507
+ "step": 4100
1508
+ },
1509
+ {
1510
+ "epoch": 7.8776290630975145,
1511
+ "grad_norm": 1.7664682865142822,
1512
+ "learning_rate": 0.00023581899298916507,
1513
+ "loss": 0.9318,
1514
+ "step": 4120
1515
+ },
1516
+ {
1517
+ "epoch": 7.915869980879541,
1518
+ "grad_norm": 1.5522887706756592,
1519
+ "learning_rate": 0.00023157000212449544,
1520
+ "loss": 0.8852,
1521
+ "step": 4140
1522
+ },
1523
+ {
1524
+ "epoch": 7.954110898661568,
1525
+ "grad_norm": 1.626836895942688,
1526
+ "learning_rate": 0.00022732101125982578,
1527
+ "loss": 0.9121,
1528
+ "step": 4160
1529
+ },
1530
+ {
1531
+ "epoch": 7.992351816443595,
1532
+ "grad_norm": 1.8042898178100586,
1533
+ "learning_rate": 0.00022307202039515615,
1534
+ "loss": 0.9221,
1535
+ "step": 4180
1536
+ },
1537
+ {
1538
+ "epoch": 8.0,
1539
+ "eval_accuracy": 0.9266733938782374,
1540
+ "eval_loss": 0.3572401702404022,
1541
+ "eval_runtime": 168.0336,
1542
+ "eval_samples_per_second": 88.464,
1543
+ "eval_steps_per_second": 88.464,
1544
+ "step": 4184
1545
+ },
1546
+ {
1547
+ "epoch": 8.030592734225621,
1548
+ "grad_norm": 1.545024037361145,
1549
+ "learning_rate": 0.00021882302953048652,
1550
+ "loss": 0.8979,
1551
+ "step": 4200
1552
+ },
1553
+ {
1554
+ "epoch": 8.068833652007648,
1555
+ "grad_norm": 1.592607855796814,
1556
+ "learning_rate": 0.00021457403866581687,
1557
+ "loss": 0.8464,
1558
+ "step": 4220
1559
+ },
1560
+ {
1561
+ "epoch": 8.107074569789676,
1562
+ "grad_norm": 1.5347646474838257,
1563
+ "learning_rate": 0.0002103250478011472,
1564
+ "loss": 0.8634,
1565
+ "step": 4240
1566
+ },
1567
+ {
1568
+ "epoch": 8.145315487571702,
1569
+ "grad_norm": 1.621201515197754,
1570
+ "learning_rate": 0.0002060760569364776,
1571
+ "loss": 0.8425,
1572
+ "step": 4260
1573
+ },
1574
+ {
1575
+ "epoch": 8.183556405353729,
1576
+ "grad_norm": 1.7381062507629395,
1577
+ "learning_rate": 0.00020182706607180795,
1578
+ "loss": 0.8776,
1579
+ "step": 4280
1580
+ },
1581
+ {
1582
+ "epoch": 8.221797323135755,
1583
+ "grad_norm": 1.5798373222351074,
1584
+ "learning_rate": 0.0001975780752071383,
1585
+ "loss": 0.854,
1586
+ "step": 4300
1587
+ },
1588
+ {
1589
+ "epoch": 8.260038240917781,
1590
+ "grad_norm": 1.5751338005065918,
1591
+ "learning_rate": 0.00019332908434246867,
1592
+ "loss": 0.8646,
1593
+ "step": 4320
1594
+ },
1595
+ {
1596
+ "epoch": 8.29827915869981,
1597
+ "grad_norm": 1.570742130279541,
1598
+ "learning_rate": 0.00018908009347779904,
1599
+ "loss": 0.8521,
1600
+ "step": 4340
1601
+ },
1602
+ {
1603
+ "epoch": 8.336520076481836,
1604
+ "grad_norm": 1.7959846258163452,
1605
+ "learning_rate": 0.00018483110261312938,
1606
+ "loss": 0.8858,
1607
+ "step": 4360
1608
+ },
1609
+ {
1610
+ "epoch": 8.374760994263863,
1611
+ "grad_norm": 1.7537908554077148,
1612
+ "learning_rate": 0.00018058211174845973,
1613
+ "loss": 0.8707,
1614
+ "step": 4380
1615
+ },
1616
+ {
1617
+ "epoch": 8.413001912045889,
1618
+ "grad_norm": 1.635578989982605,
1619
+ "learning_rate": 0.0001763331208837901,
1620
+ "loss": 0.8375,
1621
+ "step": 4400
1622
+ },
1623
+ {
1624
+ "epoch": 8.451242829827915,
1625
+ "grad_norm": 1.5729222297668457,
1626
+ "learning_rate": 0.00017208413001912047,
1627
+ "loss": 0.8631,
1628
+ "step": 4420
1629
+ },
1630
+ {
1631
+ "epoch": 8.489483747609942,
1632
+ "grad_norm": 1.6586476564407349,
1633
+ "learning_rate": 0.00016783513915445082,
1634
+ "loss": 0.8551,
1635
+ "step": 4440
1636
+ },
1637
+ {
1638
+ "epoch": 8.52772466539197,
1639
+ "grad_norm": 1.6118619441986084,
1640
+ "learning_rate": 0.00016358614828978119,
1641
+ "loss": 0.8637,
1642
+ "step": 4460
1643
+ },
1644
+ {
1645
+ "epoch": 8.565965583173996,
1646
+ "grad_norm": 1.5538595914840698,
1647
+ "learning_rate": 0.00015933715742511153,
1648
+ "loss": 0.8484,
1649
+ "step": 4480
1650
+ },
1651
+ {
1652
+ "epoch": 8.604206500956023,
1653
+ "grad_norm": 1.5646642446517944,
1654
+ "learning_rate": 0.0001550881665604419,
1655
+ "loss": 0.8433,
1656
+ "step": 4500
1657
+ },
1658
+ {
1659
+ "epoch": 8.64244741873805,
1660
+ "grad_norm": 1.7190415859222412,
1661
+ "learning_rate": 0.00015083917569577227,
1662
+ "loss": 0.8592,
1663
+ "step": 4520
1664
+ },
1665
+ {
1666
+ "epoch": 8.680688336520076,
1667
+ "grad_norm": 1.4950307607650757,
1668
+ "learning_rate": 0.00014659018483110262,
1669
+ "loss": 0.8236,
1670
+ "step": 4540
1671
+ },
1672
+ {
1673
+ "epoch": 8.718929254302104,
1674
+ "grad_norm": 1.5117732286453247,
1675
+ "learning_rate": 0.00014234119396643296,
1676
+ "loss": 0.8421,
1677
+ "step": 4560
1678
+ },
1679
+ {
1680
+ "epoch": 8.75717017208413,
1681
+ "grad_norm": 1.5558750629425049,
1682
+ "learning_rate": 0.00013809220310176336,
1683
+ "loss": 0.8287,
1684
+ "step": 4580
1685
+ },
1686
+ {
1687
+ "epoch": 8.795411089866157,
1688
+ "grad_norm": 1.7955564260482788,
1689
+ "learning_rate": 0.0001338432122370937,
1690
+ "loss": 0.8492,
1691
+ "step": 4600
1692
+ },
1693
+ {
1694
+ "epoch": 8.833652007648183,
1695
+ "grad_norm": 1.6532599925994873,
1696
+ "learning_rate": 0.00012959422137242405,
1697
+ "loss": 0.8419,
1698
+ "step": 4620
1699
+ },
1700
+ {
1701
+ "epoch": 8.87189292543021,
1702
+ "grad_norm": 1.7040739059448242,
1703
+ "learning_rate": 0.0001253452305077544,
1704
+ "loss": 0.8125,
1705
+ "step": 4640
1706
+ },
1707
+ {
1708
+ "epoch": 8.910133843212238,
1709
+ "grad_norm": 1.7040703296661377,
1710
+ "learning_rate": 0.00012109623964308478,
1711
+ "loss": 0.8187,
1712
+ "step": 4660
1713
+ },
1714
+ {
1715
+ "epoch": 8.948374760994264,
1716
+ "grad_norm": 1.7090845108032227,
1717
+ "learning_rate": 0.00011684724877841513,
1718
+ "loss": 0.8155,
1719
+ "step": 4680
1720
+ },
1721
+ {
1722
+ "epoch": 8.98661567877629,
1723
+ "grad_norm": 1.6070616245269775,
1724
+ "learning_rate": 0.00011259825791374549,
1725
+ "loss": 0.7997,
1726
+ "step": 4700
1727
+ },
1728
+ {
1729
+ "epoch": 9.0,
1730
+ "eval_accuracy": 0.9352842246888665,
1731
+ "eval_loss": 0.3138497769832611,
1732
+ "eval_runtime": 177.561,
1733
+ "eval_samples_per_second": 83.718,
1734
+ "eval_steps_per_second": 83.718,
1735
+ "step": 4707
1736
+ },
1737
+ {
1738
+ "epoch": 9.024856596558317,
1739
+ "grad_norm": 1.5590825080871582,
1740
+ "learning_rate": 0.00010834926704907585,
1741
+ "loss": 0.7906,
1742
+ "step": 4720
1743
+ },
1744
+ {
1745
+ "epoch": 9.063097514340344,
1746
+ "grad_norm": 1.4745252132415771,
1747
+ "learning_rate": 0.0001041002761844062,
1748
+ "loss": 0.7494,
1749
+ "step": 4740
1750
+ },
1751
+ {
1752
+ "epoch": 9.101338432122372,
1753
+ "grad_norm": 1.61099112033844,
1754
+ "learning_rate": 9.985128531973658e-05,
1755
+ "loss": 0.7854,
1756
+ "step": 4760
1757
+ },
1758
+ {
1759
+ "epoch": 9.139579349904398,
1760
+ "grad_norm": 1.5839650630950928,
1761
+ "learning_rate": 9.560229445506692e-05,
1762
+ "loss": 0.7636,
1763
+ "step": 4780
1764
+ },
1765
+ {
1766
+ "epoch": 9.177820267686425,
1767
+ "grad_norm": 1.7259138822555542,
1768
+ "learning_rate": 9.135330359039729e-05,
1769
+ "loss": 0.776,
1770
+ "step": 4800
1771
+ },
1772
+ {
1773
+ "epoch": 9.216061185468451,
1774
+ "grad_norm": 1.5495970249176025,
1775
+ "learning_rate": 8.710431272572764e-05,
1776
+ "loss": 0.7444,
1777
+ "step": 4820
1778
+ },
1779
+ {
1780
+ "epoch": 9.254302103250478,
1781
+ "grad_norm": 1.5250838994979858,
1782
+ "learning_rate": 8.2855321861058e-05,
1783
+ "loss": 0.7603,
1784
+ "step": 4840
1785
+ },
1786
+ {
1787
+ "epoch": 9.292543021032504,
1788
+ "grad_norm": 1.6244220733642578,
1789
+ "learning_rate": 7.860633099638836e-05,
1790
+ "loss": 0.7561,
1791
+ "step": 4860
1792
+ },
1793
+ {
1794
+ "epoch": 9.330783938814532,
1795
+ "grad_norm": 1.6825993061065674,
1796
+ "learning_rate": 7.435734013171871e-05,
1797
+ "loss": 0.7908,
1798
+ "step": 4880
1799
+ },
1800
+ {
1801
+ "epoch": 9.369024856596559,
1802
+ "grad_norm": 1.563707947731018,
1803
+ "learning_rate": 7.010834926704908e-05,
1804
+ "loss": 0.7517,
1805
+ "step": 4900
1806
+ },
1807
+ {
1808
+ "epoch": 9.407265774378585,
1809
+ "grad_norm": 1.7463629245758057,
1810
+ "learning_rate": 6.585935840237942e-05,
1811
+ "loss": 0.7679,
1812
+ "step": 4920
1813
+ },
1814
+ {
1815
+ "epoch": 9.445506692160611,
1816
+ "grad_norm": 1.5689053535461426,
1817
+ "learning_rate": 6.16103675377098e-05,
1818
+ "loss": 0.7426,
1819
+ "step": 4940
1820
+ },
1821
+ {
1822
+ "epoch": 9.483747609942638,
1823
+ "grad_norm": 1.6512914896011353,
1824
+ "learning_rate": 5.736137667304015e-05,
1825
+ "loss": 0.7695,
1826
+ "step": 4960
1827
+ },
1828
+ {
1829
+ "epoch": 9.521988527724666,
1830
+ "grad_norm": 1.6542084217071533,
1831
+ "learning_rate": 5.311238580837052e-05,
1832
+ "loss": 0.7603,
1833
+ "step": 4980
1834
+ },
1835
+ {
1836
+ "epoch": 9.560229445506693,
1837
+ "grad_norm": 1.6929945945739746,
1838
+ "learning_rate": 4.8863394943700874e-05,
1839
+ "loss": 0.754,
1840
+ "step": 5000
1841
+ },
1842
+ {
1843
+ "epoch": 9.598470363288719,
1844
+ "grad_norm": 1.4880517721176147,
1845
+ "learning_rate": 4.461440407903123e-05,
1846
+ "loss": 0.7597,
1847
+ "step": 5020
1848
+ },
1849
+ {
1850
+ "epoch": 9.636711281070745,
1851
+ "grad_norm": 1.578971266746521,
1852
+ "learning_rate": 4.036541321436159e-05,
1853
+ "loss": 0.7624,
1854
+ "step": 5040
1855
+ },
1856
+ {
1857
+ "epoch": 9.674952198852772,
1858
+ "grad_norm": 1.616727352142334,
1859
+ "learning_rate": 3.6116422349691954e-05,
1860
+ "loss": 0.7653,
1861
+ "step": 5060
1862
+ },
1863
+ {
1864
+ "epoch": 9.7131931166348,
1865
+ "grad_norm": 1.6762784719467163,
1866
+ "learning_rate": 3.186743148502231e-05,
1867
+ "loss": 0.7352,
1868
+ "step": 5080
1869
+ },
1870
+ {
1871
+ "epoch": 9.751434034416826,
1872
+ "grad_norm": 1.5666388273239136,
1873
+ "learning_rate": 2.7618440620352666e-05,
1874
+ "loss": 0.732,
1875
+ "step": 5100
1876
+ },
1877
+ {
1878
+ "epoch": 9.789674952198853,
1879
+ "grad_norm": 1.641012191772461,
1880
+ "learning_rate": 2.3369449755683023e-05,
1881
+ "loss": 0.7631,
1882
+ "step": 5120
1883
+ },
1884
+ {
1885
+ "epoch": 9.82791586998088,
1886
+ "grad_norm": 1.7024327516555786,
1887
+ "learning_rate": 1.9120458891013384e-05,
1888
+ "loss": 0.7153,
1889
+ "step": 5140
1890
+ },
1891
+ {
1892
+ "epoch": 9.866156787762906,
1893
+ "grad_norm": 1.4840829372406006,
1894
+ "learning_rate": 1.4871468026343743e-05,
1895
+ "loss": 0.7247,
1896
+ "step": 5160
1897
+ },
1898
+ {
1899
+ "epoch": 9.904397705544934,
1900
+ "grad_norm": 1.627562165260315,
1901
+ "learning_rate": 1.0622477161674103e-05,
1902
+ "loss": 0.7289,
1903
+ "step": 5180
1904
+ },
1905
+ {
1906
+ "epoch": 9.94263862332696,
1907
+ "grad_norm": 1.6473079919815063,
1908
+ "learning_rate": 6.373486297004461e-06,
1909
+ "loss": 0.7563,
1910
+ "step": 5200
1911
+ },
1912
+ {
1913
+ "epoch": 9.980879541108987,
1914
+ "grad_norm": 1.577776312828064,
1915
+ "learning_rate": 2.1244954323348204e-06,
1916
+ "loss": 0.7603,
1917
+ "step": 5220
1918
+ },
1919
+ {
1920
+ "epoch": 10.0,
1921
+ "eval_accuracy": 0.9410023545240498,
1922
+ "eval_loss": 0.29460111260414124,
1923
+ "eval_runtime": 187.2975,
1924
+ "eval_samples_per_second": 79.366,
1925
+ "eval_steps_per_second": 79.366,
1926
+ "step": 5230
1927
+ },
1928
+ {
1929
+ "epoch": 10.0,
1930
+ "step": 5230,
1931
+ "total_flos": 1.96318398191328e+18,
1932
+ "train_loss": 2.1099621225725396,
1933
+ "train_runtime": 19361.582,
1934
+ "train_samples_per_second": 69.094,
1935
+ "train_steps_per_second": 0.27
1936
+ }
1937
+ ],
1938
+ "logging_steps": 20,
1939
+ "max_steps": 5230,
1940
+ "num_input_tokens_seen": 0,
1941
+ "num_train_epochs": 10,
1942
+ "save_steps": 500,
1943
+ "stateful_callbacks": {
1944
+ "TrainerControl": {
1945
+ "args": {
1946
+ "should_epoch_stop": false,
1947
+ "should_evaluate": false,
1948
+ "should_log": false,
1949
+ "should_save": true,
1950
+ "should_training_stop": true
1951
+ },
1952
+ "attributes": {}
1953
+ }
1954
+ },
1955
+ "total_flos": 1.96318398191328e+18,
1956
+ "train_batch_size": 256,
1957
+ "trial_name": null,
1958
+ "trial_params": null
1959
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d1bb824344c6ddde9b20da9c3f5c9f3402868c8cc34b80554b6d30758a81850
3
+ size 5051