Hugging Face's logo

leafspark
/
Llama-3-70b-Toxic-DPO-v0.1

PEFT
Safetensors
Model card Files Community
leafspark commited on
Commit
96bc304
•
1 Parent(s): 0238ff6

model: upload LoRA

Browse files
Files changed (8) hide show
  1. README.md +202 -3
  2. adapter_config.json +31 -0
  3. adapter_model.safetensors +3 -0
  4. optimizer.pt +3 -0
  5. rng_state.pth +3 -0
  6. scheduler.pt +3 -0
  7. trainer_state.json +3861 -0
  8. training_args.bin +3 -0
README.md CHANGED
@@ -1,3 +1,202 @@
1
- ---
2
- license: llama3
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: monsterapi/Meta-Llama-3-70B-Instruct_4bit_bnb
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.10.0
adapter_config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "monsterapi/Meta-Llama-3-70B-Instruct_4bit_bnb",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 16,
14
+ "lora_dropout": 0.0,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 32,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "k_proj",
24
+ "q_proj",
25
+ "o_proj",
26
+ "v_proj"
27
+ ],
28
+ "task_type": "CAUSAL_LM",
29
+ "use_dora": false,
30
+ "use_rslora": false
31
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3ab2e11664f3b402a691506ef38ab09306f2df399300d158907c8f754287271
3
+ size 524375368
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b9fc21ee758a0ea5a00e6f4c7f2eb51ac5c5b3789e36a0a3cd7607749b43a9e
3
+ size 1048957202
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7682299c566684ea51cf26f0c86b6ffaa3c0bc63cbdf84674b29a2c62ac72143
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f40b3ef73d5152cc759fa6f951ad47b3d34d855b864829e062b5dc45d8939e1
3
+ size 1192
trainer_state.json ADDED
@@ -0,0 +1,3861 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6085956692695618,
3
+ "best_model_checkpoint": "outputs/checkpoint-544",
4
+ "epoch": 1.0055452865064696,
5
+ "eval_steps": 136,
6
+ "global_step": 544,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0018484288354898336,
13
+ "grad_norm": 0.466289758682251,
14
+ "learning_rate": 0.001,
15
+ "loss": 1.4884,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0036968576709796672,
20
+ "grad_norm": 0.15260683000087738,
21
+ "learning_rate": 0.001,
22
+ "loss": 1.2641,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.005545286506469501,
27
+ "grad_norm": 2.4310109615325928,
28
+ "learning_rate": 0.001,
29
+ "loss": 0.8826,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.0073937153419593345,
34
+ "grad_norm": 0.3376069664955139,
35
+ "learning_rate": 0.001,
36
+ "loss": 0.9718,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.009242144177449169,
41
+ "grad_norm": 0.5089592337608337,
42
+ "learning_rate": 0.001,
43
+ "loss": 1.0333,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.011090573012939002,
48
+ "grad_norm": 0.7866702079772949,
49
+ "learning_rate": 0.001,
50
+ "loss": 0.7162,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.012939001848428836,
55
+ "grad_norm": 0.2517491579055786,
56
+ "learning_rate": 0.001,
57
+ "loss": 0.7631,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.014787430683918669,
62
+ "grad_norm": 0.306551992893219,
63
+ "learning_rate": 0.001,
64
+ "loss": 0.6545,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.0166358595194085,
69
+ "grad_norm": 0.32659947872161865,
70
+ "learning_rate": 0.001,
71
+ "loss": 0.5165,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.018484288354898338,
76
+ "grad_norm": 0.30095258355140686,
77
+ "learning_rate": 0.001,
78
+ "loss": 0.6372,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.02033271719038817,
83
+ "grad_norm": 0.33985891938209534,
84
+ "learning_rate": 0.001,
85
+ "loss": 0.8448,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.022181146025878003,
90
+ "grad_norm": 0.42997440695762634,
91
+ "learning_rate": 0.001,
92
+ "loss": 0.8519,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.024029574861367836,
97
+ "grad_norm": 0.4363529682159424,
98
+ "learning_rate": 0.001,
99
+ "loss": 0.7915,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.025878003696857672,
104
+ "grad_norm": 0.2837740182876587,
105
+ "learning_rate": 0.001,
106
+ "loss": 0.5573,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.027726432532347505,
111
+ "grad_norm": 0.33749231696128845,
112
+ "learning_rate": 0.001,
113
+ "loss": 0.6843,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.029574861367837338,
118
+ "grad_norm": 0.6295365691184998,
119
+ "learning_rate": 0.001,
120
+ "loss": 0.6791,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.031423290203327174,
125
+ "grad_norm": 0.2831462323665619,
126
+ "learning_rate": 0.001,
127
+ "loss": 0.6832,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.033271719038817,
132
+ "grad_norm": 0.5176781415939331,
133
+ "learning_rate": 0.001,
134
+ "loss": 0.4944,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.03512014787430684,
139
+ "grad_norm": 0.5044294595718384,
140
+ "learning_rate": 0.001,
141
+ "loss": 0.8668,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.036968576709796676,
146
+ "grad_norm": 0.27560847997665405,
147
+ "learning_rate": 0.001,
148
+ "loss": 0.697,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.038817005545286505,
153
+ "grad_norm": 2.1971335411071777,
154
+ "learning_rate": 0.001,
155
+ "loss": 0.6876,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.04066543438077634,
160
+ "grad_norm": 0.4692804217338562,
161
+ "learning_rate": 0.001,
162
+ "loss": 0.6055,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.04251386321626617,
167
+ "grad_norm": 1.2793192863464355,
168
+ "learning_rate": 0.001,
169
+ "loss": 0.8302,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.04436229205175601,
174
+ "grad_norm": 0.7404688000679016,
175
+ "learning_rate": 0.001,
176
+ "loss": 0.4546,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.04621072088724584,
181
+ "grad_norm": 0.2830832600593567,
182
+ "learning_rate": 0.001,
183
+ "loss": 0.6068,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.04805914972273567,
188
+ "grad_norm": 0.3706754744052887,
189
+ "learning_rate": 0.001,
190
+ "loss": 0.5629,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.04990757855822551,
195
+ "grad_norm": 0.5734543800354004,
196
+ "learning_rate": 0.001,
197
+ "loss": 0.7772,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.051756007393715345,
202
+ "grad_norm": 0.4173612594604492,
203
+ "learning_rate": 0.001,
204
+ "loss": 0.7023,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.053604436229205174,
209
+ "grad_norm": 0.35146352648735046,
210
+ "learning_rate": 0.001,
211
+ "loss": 0.7729,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.05545286506469501,
216
+ "grad_norm": 0.3884046673774719,
217
+ "learning_rate": 0.001,
218
+ "loss": 0.7233,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.05730129390018484,
223
+ "grad_norm": 0.34868955612182617,
224
+ "learning_rate": 0.001,
225
+ "loss": 0.5834,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.059149722735674676,
230
+ "grad_norm": 1.9348665475845337,
231
+ "learning_rate": 0.001,
232
+ "loss": 0.6,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.06099815157116451,
237
+ "grad_norm": 0.49200278520584106,
238
+ "learning_rate": 0.001,
239
+ "loss": 0.4783,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.06284658040665435,
244
+ "grad_norm": 0.4063400626182556,
245
+ "learning_rate": 0.001,
246
+ "loss": 0.5797,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.06469500924214418,
251
+ "grad_norm": 0.4601685702800751,
252
+ "learning_rate": 0.001,
253
+ "loss": 0.5789,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.066543438077634,
258
+ "grad_norm": 0.5618835687637329,
259
+ "learning_rate": 0.001,
260
+ "loss": 0.7446,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.06839186691312385,
265
+ "grad_norm": 0.5381736755371094,
266
+ "learning_rate": 0.001,
267
+ "loss": 1.1037,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.07024029574861368,
272
+ "grad_norm": 0.579923689365387,
273
+ "learning_rate": 0.001,
274
+ "loss": 0.5367,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.07208872458410351,
279
+ "grad_norm": 0.4076865613460541,
280
+ "learning_rate": 0.001,
281
+ "loss": 0.6835,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.07393715341959335,
286
+ "grad_norm": 0.3602221608161926,
287
+ "learning_rate": 0.001,
288
+ "loss": 0.5843,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.07578558225508318,
293
+ "grad_norm": 0.4595842957496643,
294
+ "learning_rate": 0.001,
295
+ "loss": 0.8391,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.07763401109057301,
300
+ "grad_norm": 1.456554651260376,
301
+ "learning_rate": 0.001,
302
+ "loss": 0.8826,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.07948243992606285,
307
+ "grad_norm": 0.40481042861938477,
308
+ "learning_rate": 0.001,
309
+ "loss": 0.6189,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.08133086876155268,
314
+ "grad_norm": 2.4103527069091797,
315
+ "learning_rate": 0.001,
316
+ "loss": 0.7887,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.08317929759704251,
321
+ "grad_norm": 0.42522993683815,
322
+ "learning_rate": 0.001,
323
+ "loss": 0.6804,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.08502772643253234,
328
+ "grad_norm": 0.5652860999107361,
329
+ "learning_rate": 0.001,
330
+ "loss": 0.5217,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.08687615526802218,
335
+ "grad_norm": 0.6977353096008301,
336
+ "learning_rate": 0.001,
337
+ "loss": 0.7222,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.08872458410351201,
342
+ "grad_norm": 0.49619829654693604,
343
+ "learning_rate": 0.001,
344
+ "loss": 0.8871,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.09057301293900184,
349
+ "grad_norm": 0.3837599456310272,
350
+ "learning_rate": 0.001,
351
+ "loss": 0.745,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.09242144177449169,
356
+ "grad_norm": 0.35798409581184387,
357
+ "learning_rate": 0.001,
358
+ "loss": 0.7044,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.09426987060998152,
363
+ "grad_norm": 0.3777969777584076,
364
+ "learning_rate": 0.001,
365
+ "loss": 0.5723,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.09611829944547134,
370
+ "grad_norm": 0.35432809591293335,
371
+ "learning_rate": 0.001,
372
+ "loss": 0.6851,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.09796672828096119,
377
+ "grad_norm": 0.4954202175140381,
378
+ "learning_rate": 0.001,
379
+ "loss": 0.8028,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.09981515711645102,
384
+ "grad_norm": 0.3448912501335144,
385
+ "learning_rate": 0.001,
386
+ "loss": 0.6723,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.10166358595194085,
391
+ "grad_norm": 0.4123232960700989,
392
+ "learning_rate": 0.001,
393
+ "loss": 0.6049,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.10351201478743069,
398
+ "grad_norm": 0.37372708320617676,
399
+ "learning_rate": 0.001,
400
+ "loss": 0.5747,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.10536044362292052,
405
+ "grad_norm": 0.4807383120059967,
406
+ "learning_rate": 0.001,
407
+ "loss": 0.6363,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.10720887245841035,
412
+ "grad_norm": 0.8781284093856812,
413
+ "learning_rate": 0.001,
414
+ "loss": 2.1231,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.10905730129390019,
419
+ "grad_norm": 0.5836363434791565,
420
+ "learning_rate": 0.001,
421
+ "loss": 0.7727,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.11090573012939002,
426
+ "grad_norm": 0.41094064712524414,
427
+ "learning_rate": 0.001,
428
+ "loss": 0.6618,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.11275415896487985,
433
+ "grad_norm": 0.4478781521320343,
434
+ "learning_rate": 0.001,
435
+ "loss": 0.73,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.11460258780036968,
440
+ "grad_norm": 0.36589401960372925,
441
+ "learning_rate": 0.001,
442
+ "loss": 0.5947,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.11645101663585952,
447
+ "grad_norm": 0.5363914966583252,
448
+ "learning_rate": 0.001,
449
+ "loss": 0.5481,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.11829944547134935,
454
+ "grad_norm": 0.42670416831970215,
455
+ "learning_rate": 0.001,
456
+ "loss": 0.508,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.12014787430683918,
461
+ "grad_norm": 0.7648505568504333,
462
+ "learning_rate": 0.001,
463
+ "loss": 0.6343,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.12199630314232902,
468
+ "grad_norm": 0.4689379334449768,
469
+ "learning_rate": 0.001,
470
+ "loss": 0.7469,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.12384473197781885,
475
+ "grad_norm": 0.44720444083213806,
476
+ "learning_rate": 0.001,
477
+ "loss": 0.6952,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.1256931608133087,
482
+ "grad_norm": 0.35768693685531616,
483
+ "learning_rate": 0.001,
484
+ "loss": 0.432,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.12754158964879853,
489
+ "grad_norm": 0.4105698764324188,
490
+ "learning_rate": 0.001,
491
+ "loss": 0.6861,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.12939001848428835,
496
+ "grad_norm": 0.38207560777664185,
497
+ "learning_rate": 0.001,
498
+ "loss": 0.5908,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.13123844731977818,
503
+ "grad_norm": 0.46701478958129883,
504
+ "learning_rate": 0.001,
505
+ "loss": 0.6832,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.133086876155268,
510
+ "grad_norm": 0.501502513885498,
511
+ "learning_rate": 0.001,
512
+ "loss": 0.7041,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.13493530499075784,
517
+ "grad_norm": 0.3278079926967621,
518
+ "learning_rate": 0.001,
519
+ "loss": 0.4946,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.1367837338262477,
524
+ "grad_norm": 0.4573265612125397,
525
+ "learning_rate": 0.001,
526
+ "loss": 0.6106,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.13863216266173753,
531
+ "grad_norm": 0.37268367409706116,
532
+ "learning_rate": 0.001,
533
+ "loss": 0.4687,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.14048059149722736,
538
+ "grad_norm": 0.41542476415634155,
539
+ "learning_rate": 0.001,
540
+ "loss": 0.7345,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.1423290203327172,
545
+ "grad_norm": 0.36338284611701965,
546
+ "learning_rate": 0.001,
547
+ "loss": 0.5258,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.14417744916820702,
552
+ "grad_norm": 0.35523292422294617,
553
+ "learning_rate": 0.001,
554
+ "loss": 0.5006,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.14602587800369685,
559
+ "grad_norm": 0.7458345293998718,
560
+ "learning_rate": 0.001,
561
+ "loss": 1.6438,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.1478743068391867,
566
+ "grad_norm": 0.4105502963066101,
567
+ "learning_rate": 0.001,
568
+ "loss": 0.5162,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.14972273567467653,
573
+ "grad_norm": 0.4603978395462036,
574
+ "learning_rate": 0.001,
575
+ "loss": 0.7375,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.15157116451016636,
580
+ "grad_norm": 0.3389419913291931,
581
+ "learning_rate": 0.001,
582
+ "loss": 0.4315,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.1534195933456562,
587
+ "grad_norm": 0.4641773998737335,
588
+ "learning_rate": 0.001,
589
+ "loss": 0.5741,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.15526802218114602,
594
+ "grad_norm": 0.7335869073867798,
595
+ "learning_rate": 0.001,
596
+ "loss": 0.6726,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.15711645101663585,
601
+ "grad_norm": 0.45697247982025146,
602
+ "learning_rate": 0.001,
603
+ "loss": 0.6194,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.1589648798521257,
608
+ "grad_norm": 0.5353591442108154,
609
+ "learning_rate": 0.001,
610
+ "loss": 0.6411,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.16081330868761554,
615
+ "grad_norm": 0.5396345257759094,
616
+ "learning_rate": 0.001,
617
+ "loss": 0.6453,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.16266173752310537,
622
+ "grad_norm": 0.5480281114578247,
623
+ "learning_rate": 0.001,
624
+ "loss": 0.6838,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.1645101663585952,
629
+ "grad_norm": 0.455941379070282,
630
+ "learning_rate": 0.001,
631
+ "loss": 0.58,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.16635859519408502,
636
+ "grad_norm": 0.8435553312301636,
637
+ "learning_rate": 0.001,
638
+ "loss": 0.8912,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.16820702402957485,
643
+ "grad_norm": 0.44801416993141174,
644
+ "learning_rate": 0.001,
645
+ "loss": 0.6285,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.17005545286506468,
650
+ "grad_norm": 0.4071008563041687,
651
+ "learning_rate": 0.001,
652
+ "loss": 0.6715,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.17190388170055454,
657
+ "grad_norm": 0.41972634196281433,
658
+ "learning_rate": 0.001,
659
+ "loss": 0.453,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.17375231053604437,
664
+ "grad_norm": 0.60628741979599,
665
+ "learning_rate": 0.001,
666
+ "loss": 0.7723,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.1756007393715342,
671
+ "grad_norm": 0.4178489148616791,
672
+ "learning_rate": 0.001,
673
+ "loss": 0.4684,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.17744916820702403,
678
+ "grad_norm": 0.5033894777297974,
679
+ "learning_rate": 0.001,
680
+ "loss": 0.6145,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.17929759704251386,
685
+ "grad_norm": 0.6618285179138184,
686
+ "learning_rate": 0.001,
687
+ "loss": 0.9387,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.18114602587800369,
692
+ "grad_norm": 0.47585925459861755,
693
+ "learning_rate": 0.001,
694
+ "loss": 0.7797,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.18299445471349354,
699
+ "grad_norm": 0.41978687047958374,
700
+ "learning_rate": 0.001,
701
+ "loss": 0.5475,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.18484288354898337,
706
+ "grad_norm": 0.5080726742744446,
707
+ "learning_rate": 0.001,
708
+ "loss": 0.5243,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.1866913123844732,
713
+ "grad_norm": 0.38001152873039246,
714
+ "learning_rate": 0.001,
715
+ "loss": 0.532,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.18853974121996303,
720
+ "grad_norm": 0.5791141986846924,
721
+ "learning_rate": 0.001,
722
+ "loss": 0.7756,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.19038817005545286,
727
+ "grad_norm": 0.42902910709381104,
728
+ "learning_rate": 0.001,
729
+ "loss": 0.6425,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.1922365988909427,
734
+ "grad_norm": 0.5157350301742554,
735
+ "learning_rate": 0.001,
736
+ "loss": 0.819,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.19408502772643252,
741
+ "grad_norm": 12.421481132507324,
742
+ "learning_rate": 0.001,
743
+ "loss": 0.8929,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.19593345656192238,
748
+ "grad_norm": 0.5132973790168762,
749
+ "learning_rate": 0.001,
750
+ "loss": 0.6348,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.1977818853974122,
755
+ "grad_norm": 0.4973335862159729,
756
+ "learning_rate": 0.001,
757
+ "loss": 0.6071,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.19963031423290203,
762
+ "grad_norm": 0.326150119304657,
763
+ "learning_rate": 0.001,
764
+ "loss": 0.6531,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.20147874306839186,
769
+ "grad_norm": 0.4327104985713959,
770
+ "learning_rate": 0.001,
771
+ "loss": 0.6975,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.2033271719038817,
776
+ "grad_norm": 0.5976394414901733,
777
+ "learning_rate": 0.001,
778
+ "loss": 0.7276,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.20517560073937152,
783
+ "grad_norm": 0.35819554328918457,
784
+ "learning_rate": 0.001,
785
+ "loss": 0.5249,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.20702402957486138,
790
+ "grad_norm": 0.35644710063934326,
791
+ "learning_rate": 0.001,
792
+ "loss": 0.4523,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.2088724584103512,
797
+ "grad_norm": 0.45812854170799255,
798
+ "learning_rate": 0.001,
799
+ "loss": 0.7442,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.21072088724584104,
804
+ "grad_norm": 0.5240729451179504,
805
+ "learning_rate": 0.001,
806
+ "loss": 0.6039,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.21256931608133087,
811
+ "grad_norm": 0.6027355790138245,
812
+ "learning_rate": 0.001,
813
+ "loss": 0.7314,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.2144177449168207,
818
+ "grad_norm": 0.4197920560836792,
819
+ "learning_rate": 0.001,
820
+ "loss": 0.644,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 0.21626617375231053,
825
+ "grad_norm": 0.5917842388153076,
826
+ "learning_rate": 0.001,
827
+ "loss": 0.6872,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 0.21811460258780038,
832
+ "grad_norm": 0.49236243963241577,
833
+ "learning_rate": 0.001,
834
+ "loss": 0.9002,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 0.2199630314232902,
839
+ "grad_norm": 0.6250895857810974,
840
+ "learning_rate": 0.001,
841
+ "loss": 0.6905,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 0.22181146025878004,
846
+ "grad_norm": 0.5853796005249023,
847
+ "learning_rate": 0.001,
848
+ "loss": 0.7602,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 0.22365988909426987,
853
+ "grad_norm": 0.5543091297149658,
854
+ "learning_rate": 0.001,
855
+ "loss": 0.7356,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 0.2255083179297597,
860
+ "grad_norm": 0.4340505003929138,
861
+ "learning_rate": 0.001,
862
+ "loss": 0.6715,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 0.22735674676524953,
867
+ "grad_norm": 0.4405982792377472,
868
+ "learning_rate": 0.001,
869
+ "loss": 0.6898,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 0.22920517560073936,
874
+ "grad_norm": 0.5105734467506409,
875
+ "learning_rate": 0.001,
876
+ "loss": 0.7762,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 0.23105360443622922,
881
+ "grad_norm": 0.4798790216445923,
882
+ "learning_rate": 0.001,
883
+ "loss": 0.5982,
884
+ "step": 125
885
+ },
886
+ {
887
+ "epoch": 0.23290203327171904,
888
+ "grad_norm": 0.4916364252567291,
889
+ "learning_rate": 0.001,
890
+ "loss": 0.4894,
891
+ "step": 126
892
+ },
893
+ {
894
+ "epoch": 0.23475046210720887,
895
+ "grad_norm": 0.5022260546684265,
896
+ "learning_rate": 0.001,
897
+ "loss": 0.614,
898
+ "step": 127
899
+ },
900
+ {
901
+ "epoch": 0.2365988909426987,
902
+ "grad_norm": 0.4792012870311737,
903
+ "learning_rate": 0.001,
904
+ "loss": 0.6265,
905
+ "step": 128
906
+ },
907
+ {
908
+ "epoch": 0.23844731977818853,
909
+ "grad_norm": 0.5663838982582092,
910
+ "learning_rate": 0.001,
911
+ "loss": 0.5971,
912
+ "step": 129
913
+ },
914
+ {
915
+ "epoch": 0.24029574861367836,
916
+ "grad_norm": 0.49800997972488403,
917
+ "learning_rate": 0.001,
918
+ "loss": 0.6318,
919
+ "step": 130
920
+ },
921
+ {
922
+ "epoch": 0.24214417744916822,
923
+ "grad_norm": 0.35220399498939514,
924
+ "learning_rate": 0.001,
925
+ "loss": 0.3083,
926
+ "step": 131
927
+ },
928
+ {
929
+ "epoch": 0.24399260628465805,
930
+ "grad_norm": 0.4706744849681854,
931
+ "learning_rate": 0.001,
932
+ "loss": 0.6436,
933
+ "step": 132
934
+ },
935
+ {
936
+ "epoch": 0.24584103512014788,
937
+ "grad_norm": 0.5163246393203735,
938
+ "learning_rate": 0.001,
939
+ "loss": 0.7155,
940
+ "step": 133
941
+ },
942
+ {
943
+ "epoch": 0.2476894639556377,
944
+ "grad_norm": 0.4994561970233917,
945
+ "learning_rate": 0.001,
946
+ "loss": 0.8529,
947
+ "step": 134
948
+ },
949
+ {
950
+ "epoch": 0.24953789279112754,
951
+ "grad_norm": 0.47701308131217957,
952
+ "learning_rate": 0.001,
953
+ "loss": 0.6465,
954
+ "step": 135
955
+ },
956
+ {
957
+ "epoch": 0.2513863216266174,
958
+ "grad_norm": 0.5003355145454407,
959
+ "learning_rate": 0.001,
960
+ "loss": 0.6787,
961
+ "step": 136
962
+ },
963
+ {
964
+ "epoch": 0.2513863216266174,
965
+ "eval_loss": 0.6116308569908142,
966
+ "eval_runtime": 37.1894,
967
+ "eval_samples_per_second": 1.479,
968
+ "eval_steps_per_second": 0.188,
969
+ "step": 136
970
+ },
971
+ {
972
+ "epoch": 0.2532347504621072,
973
+ "grad_norm": 0.46851846575737,
974
+ "learning_rate": 0.001,
975
+ "loss": 0.7461,
976
+ "step": 137
977
+ },
978
+ {
979
+ "epoch": 0.25508317929759705,
980
+ "grad_norm": 0.703341007232666,
981
+ "learning_rate": 0.001,
982
+ "loss": 0.709,
983
+ "step": 138
984
+ },
985
+ {
986
+ "epoch": 0.25693160813308685,
987
+ "grad_norm": 0.48569703102111816,
988
+ "learning_rate": 0.001,
989
+ "loss": 0.7207,
990
+ "step": 139
991
+ },
992
+ {
993
+ "epoch": 0.2587800369685767,
994
+ "grad_norm": 0.3450978398323059,
995
+ "learning_rate": 0.001,
996
+ "loss": 0.5599,
997
+ "step": 140
998
+ },
999
+ {
1000
+ "epoch": 0.26062846580406657,
1001
+ "grad_norm": 0.5488552451133728,
1002
+ "learning_rate": 0.001,
1003
+ "loss": 0.8029,
1004
+ "step": 141
1005
+ },
1006
+ {
1007
+ "epoch": 0.26247689463955637,
1008
+ "grad_norm": 0.43643897771835327,
1009
+ "learning_rate": 0.001,
1010
+ "loss": 0.6259,
1011
+ "step": 142
1012
+ },
1013
+ {
1014
+ "epoch": 0.2643253234750462,
1015
+ "grad_norm": 0.4441094398498535,
1016
+ "learning_rate": 0.001,
1017
+ "loss": 0.688,
1018
+ "step": 143
1019
+ },
1020
+ {
1021
+ "epoch": 0.266173752310536,
1022
+ "grad_norm": 0.40471598505973816,
1023
+ "learning_rate": 0.001,
1024
+ "loss": 0.666,
1025
+ "step": 144
1026
+ },
1027
+ {
1028
+ "epoch": 0.2680221811460259,
1029
+ "grad_norm": 0.5318337678909302,
1030
+ "learning_rate": 0.001,
1031
+ "loss": 0.6774,
1032
+ "step": 145
1033
+ },
1034
+ {
1035
+ "epoch": 0.2698706099815157,
1036
+ "grad_norm": 0.3838026225566864,
1037
+ "learning_rate": 0.001,
1038
+ "loss": 0.5882,
1039
+ "step": 146
1040
+ },
1041
+ {
1042
+ "epoch": 0.27171903881700554,
1043
+ "grad_norm": 757.2123413085938,
1044
+ "learning_rate": 0.001,
1045
+ "loss": 3.2426,
1046
+ "step": 147
1047
+ },
1048
+ {
1049
+ "epoch": 0.2735674676524954,
1050
+ "grad_norm": 0.4157928228378296,
1051
+ "learning_rate": 0.001,
1052
+ "loss": 0.5649,
1053
+ "step": 148
1054
+ },
1055
+ {
1056
+ "epoch": 0.2754158964879852,
1057
+ "grad_norm": 0.3751083314418793,
1058
+ "learning_rate": 0.001,
1059
+ "loss": 0.5098,
1060
+ "step": 149
1061
+ },
1062
+ {
1063
+ "epoch": 0.27726432532347506,
1064
+ "grad_norm": 0.38269737362861633,
1065
+ "learning_rate": 0.001,
1066
+ "loss": 0.6048,
1067
+ "step": 150
1068
+ },
1069
+ {
1070
+ "epoch": 0.27911275415896486,
1071
+ "grad_norm": 0.9310194849967957,
1072
+ "learning_rate": 0.001,
1073
+ "loss": 0.5548,
1074
+ "step": 151
1075
+ },
1076
+ {
1077
+ "epoch": 0.2809611829944547,
1078
+ "grad_norm": 0.6662898063659668,
1079
+ "learning_rate": 0.001,
1080
+ "loss": 1.0353,
1081
+ "step": 152
1082
+ },
1083
+ {
1084
+ "epoch": 0.2828096118299446,
1085
+ "grad_norm": 0.5112884640693665,
1086
+ "learning_rate": 0.001,
1087
+ "loss": 0.5875,
1088
+ "step": 153
1089
+ },
1090
+ {
1091
+ "epoch": 0.2846580406654344,
1092
+ "grad_norm": 0.682369589805603,
1093
+ "learning_rate": 0.001,
1094
+ "loss": 0.7255,
1095
+ "step": 154
1096
+ },
1097
+ {
1098
+ "epoch": 0.28650646950092423,
1099
+ "grad_norm": 0.37935149669647217,
1100
+ "learning_rate": 0.001,
1101
+ "loss": 0.453,
1102
+ "step": 155
1103
+ },
1104
+ {
1105
+ "epoch": 0.28835489833641403,
1106
+ "grad_norm": 0.5815888047218323,
1107
+ "learning_rate": 0.001,
1108
+ "loss": 0.6359,
1109
+ "step": 156
1110
+ },
1111
+ {
1112
+ "epoch": 0.2902033271719039,
1113
+ "grad_norm": 0.5233355164527893,
1114
+ "learning_rate": 0.001,
1115
+ "loss": 0.588,
1116
+ "step": 157
1117
+ },
1118
+ {
1119
+ "epoch": 0.2920517560073937,
1120
+ "grad_norm": 0.5995686650276184,
1121
+ "learning_rate": 0.001,
1122
+ "loss": 0.6695,
1123
+ "step": 158
1124
+ },
1125
+ {
1126
+ "epoch": 0.29390018484288355,
1127
+ "grad_norm": 0.5288013815879822,
1128
+ "learning_rate": 0.001,
1129
+ "loss": 0.6172,
1130
+ "step": 159
1131
+ },
1132
+ {
1133
+ "epoch": 0.2957486136783734,
1134
+ "grad_norm": 0.4866679310798645,
1135
+ "learning_rate": 0.001,
1136
+ "loss": 0.6383,
1137
+ "step": 160
1138
+ },
1139
+ {
1140
+ "epoch": 0.2975970425138632,
1141
+ "grad_norm": 0.4867928922176361,
1142
+ "learning_rate": 0.001,
1143
+ "loss": 0.6561,
1144
+ "step": 161
1145
+ },
1146
+ {
1147
+ "epoch": 0.29944547134935307,
1148
+ "grad_norm": 85.95575714111328,
1149
+ "learning_rate": 0.001,
1150
+ "loss": 7.2015,
1151
+ "step": 162
1152
+ },
1153
+ {
1154
+ "epoch": 0.30129390018484287,
1155
+ "grad_norm": 0.9067592620849609,
1156
+ "learning_rate": 0.001,
1157
+ "loss": 0.7286,
1158
+ "step": 163
1159
+ },
1160
+ {
1161
+ "epoch": 0.3031423290203327,
1162
+ "grad_norm": 0.8694595098495483,
1163
+ "learning_rate": 0.001,
1164
+ "loss": 0.5795,
1165
+ "step": 164
1166
+ },
1167
+ {
1168
+ "epoch": 0.3049907578558225,
1169
+ "grad_norm": 0.5374184846878052,
1170
+ "learning_rate": 0.001,
1171
+ "loss": 0.5078,
1172
+ "step": 165
1173
+ },
1174
+ {
1175
+ "epoch": 0.3068391866913124,
1176
+ "grad_norm": 0.42466646432876587,
1177
+ "learning_rate": 0.001,
1178
+ "loss": 0.5351,
1179
+ "step": 166
1180
+ },
1181
+ {
1182
+ "epoch": 0.30868761552680224,
1183
+ "grad_norm": 0.707733690738678,
1184
+ "learning_rate": 0.001,
1185
+ "loss": 0.8727,
1186
+ "step": 167
1187
+ },
1188
+ {
1189
+ "epoch": 0.31053604436229204,
1190
+ "grad_norm": 0.7297586798667908,
1191
+ "learning_rate": 0.001,
1192
+ "loss": 0.6873,
1193
+ "step": 168
1194
+ },
1195
+ {
1196
+ "epoch": 0.3123844731977819,
1197
+ "grad_norm": 0.587944507598877,
1198
+ "learning_rate": 0.001,
1199
+ "loss": 2.0879,
1200
+ "step": 169
1201
+ },
1202
+ {
1203
+ "epoch": 0.3142329020332717,
1204
+ "grad_norm": 0.9605491161346436,
1205
+ "learning_rate": 0.001,
1206
+ "loss": 1.2765,
1207
+ "step": 170
1208
+ },
1209
+ {
1210
+ "epoch": 0.31608133086876156,
1211
+ "grad_norm": 0.5181152820587158,
1212
+ "learning_rate": 0.001,
1213
+ "loss": 0.6069,
1214
+ "step": 171
1215
+ },
1216
+ {
1217
+ "epoch": 0.3179297597042514,
1218
+ "grad_norm": 2.2382378578186035,
1219
+ "learning_rate": 0.001,
1220
+ "loss": 0.7284,
1221
+ "step": 172
1222
+ },
1223
+ {
1224
+ "epoch": 0.3197781885397412,
1225
+ "grad_norm": 0.42649027705192566,
1226
+ "learning_rate": 0.001,
1227
+ "loss": 0.5089,
1228
+ "step": 173
1229
+ },
1230
+ {
1231
+ "epoch": 0.32162661737523107,
1232
+ "grad_norm": 0.5831953287124634,
1233
+ "learning_rate": 0.001,
1234
+ "loss": 0.6019,
1235
+ "step": 174
1236
+ },
1237
+ {
1238
+ "epoch": 0.3234750462107209,
1239
+ "grad_norm": 11.331945419311523,
1240
+ "learning_rate": 0.001,
1241
+ "loss": 0.7763,
1242
+ "step": 175
1243
+ },
1244
+ {
1245
+ "epoch": 0.32532347504621073,
1246
+ "grad_norm": 0.4289061725139618,
1247
+ "learning_rate": 0.001,
1248
+ "loss": 0.4493,
1249
+ "step": 176
1250
+ },
1251
+ {
1252
+ "epoch": 0.32717190388170053,
1253
+ "grad_norm": 0.7920455932617188,
1254
+ "learning_rate": 0.001,
1255
+ "loss": 0.7945,
1256
+ "step": 177
1257
+ },
1258
+ {
1259
+ "epoch": 0.3290203327171904,
1260
+ "grad_norm": 1.0137889385223389,
1261
+ "learning_rate": 0.001,
1262
+ "loss": 0.7158,
1263
+ "step": 178
1264
+ },
1265
+ {
1266
+ "epoch": 0.33086876155268025,
1267
+ "grad_norm": 1.1164557933807373,
1268
+ "learning_rate": 0.001,
1269
+ "loss": 0.7007,
1270
+ "step": 179
1271
+ },
1272
+ {
1273
+ "epoch": 0.33271719038817005,
1274
+ "grad_norm": 0.5898825526237488,
1275
+ "learning_rate": 0.001,
1276
+ "loss": 0.7362,
1277
+ "step": 180
1278
+ },
1279
+ {
1280
+ "epoch": 0.3345656192236599,
1281
+ "grad_norm": 2144.9521484375,
1282
+ "learning_rate": 0.001,
1283
+ "loss": 5.333,
1284
+ "step": 181
1285
+ },
1286
+ {
1287
+ "epoch": 0.3364140480591497,
1288
+ "grad_norm": 1.0912034511566162,
1289
+ "learning_rate": 0.001,
1290
+ "loss": 0.8123,
1291
+ "step": 182
1292
+ },
1293
+ {
1294
+ "epoch": 0.33826247689463956,
1295
+ "grad_norm": 0.5837512612342834,
1296
+ "learning_rate": 0.001,
1297
+ "loss": 0.738,
1298
+ "step": 183
1299
+ },
1300
+ {
1301
+ "epoch": 0.34011090573012936,
1302
+ "grad_norm": 0.5561927556991577,
1303
+ "learning_rate": 0.001,
1304
+ "loss": 0.5875,
1305
+ "step": 184
1306
+ },
1307
+ {
1308
+ "epoch": 0.3419593345656192,
1309
+ "grad_norm": 3.1537868976593018,
1310
+ "learning_rate": 0.001,
1311
+ "loss": 0.8469,
1312
+ "step": 185
1313
+ },
1314
+ {
1315
+ "epoch": 0.3438077634011091,
1316
+ "grad_norm": 0.751020610332489,
1317
+ "learning_rate": 0.001,
1318
+ "loss": 0.6567,
1319
+ "step": 186
1320
+ },
1321
+ {
1322
+ "epoch": 0.3456561922365989,
1323
+ "grad_norm": 2.071126699447632,
1324
+ "learning_rate": 0.001,
1325
+ "loss": 0.7666,
1326
+ "step": 187
1327
+ },
1328
+ {
1329
+ "epoch": 0.34750462107208874,
1330
+ "grad_norm": 0.8807157874107361,
1331
+ "learning_rate": 0.001,
1332
+ "loss": 1.0371,
1333
+ "step": 188
1334
+ },
1335
+ {
1336
+ "epoch": 0.34935304990757854,
1337
+ "grad_norm": 2.8368020057678223,
1338
+ "learning_rate": 0.001,
1339
+ "loss": 0.9064,
1340
+ "step": 189
1341
+ },
1342
+ {
1343
+ "epoch": 0.3512014787430684,
1344
+ "grad_norm": 16.864635467529297,
1345
+ "learning_rate": 0.001,
1346
+ "loss": 0.9614,
1347
+ "step": 190
1348
+ },
1349
+ {
1350
+ "epoch": 0.35304990757855825,
1351
+ "grad_norm": 0.7199917435646057,
1352
+ "learning_rate": 0.001,
1353
+ "loss": 0.7248,
1354
+ "step": 191
1355
+ },
1356
+ {
1357
+ "epoch": 0.35489833641404805,
1358
+ "grad_norm": 0.6968075633049011,
1359
+ "learning_rate": 0.001,
1360
+ "loss": 0.6518,
1361
+ "step": 192
1362
+ },
1363
+ {
1364
+ "epoch": 0.3567467652495379,
1365
+ "grad_norm": 0.7219743728637695,
1366
+ "learning_rate": 0.001,
1367
+ "loss": 0.7583,
1368
+ "step": 193
1369
+ },
1370
+ {
1371
+ "epoch": 0.3585951940850277,
1372
+ "grad_norm": 0.5318032503128052,
1373
+ "learning_rate": 0.001,
1374
+ "loss": 0.638,
1375
+ "step": 194
1376
+ },
1377
+ {
1378
+ "epoch": 0.36044362292051757,
1379
+ "grad_norm": 0.740088701248169,
1380
+ "learning_rate": 0.001,
1381
+ "loss": 0.6744,
1382
+ "step": 195
1383
+ },
1384
+ {
1385
+ "epoch": 0.36229205175600737,
1386
+ "grad_norm": 0.743021547794342,
1387
+ "learning_rate": 0.001,
1388
+ "loss": 0.6755,
1389
+ "step": 196
1390
+ },
1391
+ {
1392
+ "epoch": 0.36414048059149723,
1393
+ "grad_norm": 4.2798004150390625,
1394
+ "learning_rate": 0.001,
1395
+ "loss": 0.6902,
1396
+ "step": 197
1397
+ },
1398
+ {
1399
+ "epoch": 0.3659889094269871,
1400
+ "grad_norm": 0.7763252854347229,
1401
+ "learning_rate": 0.001,
1402
+ "loss": 0.7363,
1403
+ "step": 198
1404
+ },
1405
+ {
1406
+ "epoch": 0.3678373382624769,
1407
+ "grad_norm": 0.8664296865463257,
1408
+ "learning_rate": 0.001,
1409
+ "loss": 0.9194,
1410
+ "step": 199
1411
+ },
1412
+ {
1413
+ "epoch": 0.36968576709796674,
1414
+ "grad_norm": 1.4949946403503418,
1415
+ "learning_rate": 0.001,
1416
+ "loss": 0.6825,
1417
+ "step": 200
1418
+ },
1419
+ {
1420
+ "epoch": 0.37153419593345655,
1421
+ "grad_norm": 0.47144782543182373,
1422
+ "learning_rate": 0.001,
1423
+ "loss": 0.4892,
1424
+ "step": 201
1425
+ },
1426
+ {
1427
+ "epoch": 0.3733826247689464,
1428
+ "grad_norm": 2.078660249710083,
1429
+ "learning_rate": 0.001,
1430
+ "loss": 0.9382,
1431
+ "step": 202
1432
+ },
1433
+ {
1434
+ "epoch": 0.3752310536044362,
1435
+ "grad_norm": 0.8569644689559937,
1436
+ "learning_rate": 0.001,
1437
+ "loss": 0.6567,
1438
+ "step": 203
1439
+ },
1440
+ {
1441
+ "epoch": 0.37707948243992606,
1442
+ "grad_norm": 0.7167682647705078,
1443
+ "learning_rate": 0.001,
1444
+ "loss": 0.8144,
1445
+ "step": 204
1446
+ },
1447
+ {
1448
+ "epoch": 0.3789279112754159,
1449
+ "grad_norm": 2.2069590091705322,
1450
+ "learning_rate": 0.001,
1451
+ "loss": 0.8378,
1452
+ "step": 205
1453
+ },
1454
+ {
1455
+ "epoch": 0.3807763401109057,
1456
+ "grad_norm": 4.755973815917969,
1457
+ "learning_rate": 0.001,
1458
+ "loss": 0.6908,
1459
+ "step": 206
1460
+ },
1461
+ {
1462
+ "epoch": 0.3826247689463956,
1463
+ "grad_norm": 0.6981136202812195,
1464
+ "learning_rate": 0.001,
1465
+ "loss": 0.5822,
1466
+ "step": 207
1467
+ },
1468
+ {
1469
+ "epoch": 0.3844731977818854,
1470
+ "grad_norm": 0.7765632271766663,
1471
+ "learning_rate": 0.001,
1472
+ "loss": 0.8156,
1473
+ "step": 208
1474
+ },
1475
+ {
1476
+ "epoch": 0.38632162661737524,
1477
+ "grad_norm": 2.0891387462615967,
1478
+ "learning_rate": 0.001,
1479
+ "loss": 0.558,
1480
+ "step": 209
1481
+ },
1482
+ {
1483
+ "epoch": 0.38817005545286504,
1484
+ "grad_norm": 0.7609770894050598,
1485
+ "learning_rate": 0.001,
1486
+ "loss": 0.6879,
1487
+ "step": 210
1488
+ },
1489
+ {
1490
+ "epoch": 0.3900184842883549,
1491
+ "grad_norm": 0.6121675372123718,
1492
+ "learning_rate": 0.001,
1493
+ "loss": 0.8023,
1494
+ "step": 211
1495
+ },
1496
+ {
1497
+ "epoch": 0.39186691312384475,
1498
+ "grad_norm": 1.2764517068862915,
1499
+ "learning_rate": 0.001,
1500
+ "loss": 0.671,
1501
+ "step": 212
1502
+ },
1503
+ {
1504
+ "epoch": 0.39371534195933455,
1505
+ "grad_norm": 0.9839686155319214,
1506
+ "learning_rate": 0.001,
1507
+ "loss": 0.6025,
1508
+ "step": 213
1509
+ },
1510
+ {
1511
+ "epoch": 0.3955637707948244,
1512
+ "grad_norm": 0.8942749500274658,
1513
+ "learning_rate": 0.001,
1514
+ "loss": 0.7837,
1515
+ "step": 214
1516
+ },
1517
+ {
1518
+ "epoch": 0.3974121996303142,
1519
+ "grad_norm": 8.554291725158691,
1520
+ "learning_rate": 0.001,
1521
+ "loss": 0.9271,
1522
+ "step": 215
1523
+ },
1524
+ {
1525
+ "epoch": 0.39926062846580407,
1526
+ "grad_norm": 36.47178649902344,
1527
+ "learning_rate": 0.001,
1528
+ "loss": 1.0102,
1529
+ "step": 216
1530
+ },
1531
+ {
1532
+ "epoch": 0.4011090573012939,
1533
+ "grad_norm": 4.423219203948975,
1534
+ "learning_rate": 0.001,
1535
+ "loss": 0.614,
1536
+ "step": 217
1537
+ },
1538
+ {
1539
+ "epoch": 0.4029574861367837,
1540
+ "grad_norm": 0.673527181148529,
1541
+ "learning_rate": 0.001,
1542
+ "loss": 0.6804,
1543
+ "step": 218
1544
+ },
1545
+ {
1546
+ "epoch": 0.4048059149722736,
1547
+ "grad_norm": 0.8505294322967529,
1548
+ "learning_rate": 0.001,
1549
+ "loss": 0.6223,
1550
+ "step": 219
1551
+ },
1552
+ {
1553
+ "epoch": 0.4066543438077634,
1554
+ "grad_norm": 0.6489795446395874,
1555
+ "learning_rate": 0.001,
1556
+ "loss": 0.6429,
1557
+ "step": 220
1558
+ },
1559
+ {
1560
+ "epoch": 0.40850277264325324,
1561
+ "grad_norm": 0.8948644399642944,
1562
+ "learning_rate": 0.001,
1563
+ "loss": 1.4723,
1564
+ "step": 221
1565
+ },
1566
+ {
1567
+ "epoch": 0.41035120147874304,
1568
+ "grad_norm": 210.18814086914062,
1569
+ "learning_rate": 0.001,
1570
+ "loss": 1.6274,
1571
+ "step": 222
1572
+ },
1573
+ {
1574
+ "epoch": 0.4121996303142329,
1575
+ "grad_norm": 23.37410545349121,
1576
+ "learning_rate": 0.001,
1577
+ "loss": 0.6821,
1578
+ "step": 223
1579
+ },
1580
+ {
1581
+ "epoch": 0.41404805914972276,
1582
+ "grad_norm": 25.335840225219727,
1583
+ "learning_rate": 0.001,
1584
+ "loss": 1.5518,
1585
+ "step": 224
1586
+ },
1587
+ {
1588
+ "epoch": 0.41589648798521256,
1589
+ "grad_norm": 1.2805469036102295,
1590
+ "learning_rate": 0.001,
1591
+ "loss": 0.755,
1592
+ "step": 225
1593
+ },
1594
+ {
1595
+ "epoch": 0.4177449168207024,
1596
+ "grad_norm": 3.5078518390655518,
1597
+ "learning_rate": 0.001,
1598
+ "loss": 0.765,
1599
+ "step": 226
1600
+ },
1601
+ {
1602
+ "epoch": 0.4195933456561922,
1603
+ "grad_norm": 2.665682315826416,
1604
+ "learning_rate": 0.001,
1605
+ "loss": 1.0047,
1606
+ "step": 227
1607
+ },
1608
+ {
1609
+ "epoch": 0.4214417744916821,
1610
+ "grad_norm": 0.8399984240531921,
1611
+ "learning_rate": 0.001,
1612
+ "loss": 0.6717,
1613
+ "step": 228
1614
+ },
1615
+ {
1616
+ "epoch": 0.4232902033271719,
1617
+ "grad_norm": 0.5785517692565918,
1618
+ "learning_rate": 0.001,
1619
+ "loss": 0.6277,
1620
+ "step": 229
1621
+ },
1622
+ {
1623
+ "epoch": 0.42513863216266173,
1624
+ "grad_norm": 0.7574679851531982,
1625
+ "learning_rate": 0.001,
1626
+ "loss": 0.6001,
1627
+ "step": 230
1628
+ },
1629
+ {
1630
+ "epoch": 0.4269870609981516,
1631
+ "grad_norm": 1.1060686111450195,
1632
+ "learning_rate": 0.001,
1633
+ "loss": 0.4657,
1634
+ "step": 231
1635
+ },
1636
+ {
1637
+ "epoch": 0.4288354898336414,
1638
+ "grad_norm": 0.8456437587738037,
1639
+ "learning_rate": 0.001,
1640
+ "loss": 0.6504,
1641
+ "step": 232
1642
+ },
1643
+ {
1644
+ "epoch": 0.43068391866913125,
1645
+ "grad_norm": 1.1631237268447876,
1646
+ "learning_rate": 0.001,
1647
+ "loss": 0.6474,
1648
+ "step": 233
1649
+ },
1650
+ {
1651
+ "epoch": 0.43253234750462105,
1652
+ "grad_norm": 0.8575699925422668,
1653
+ "learning_rate": 0.001,
1654
+ "loss": 0.8597,
1655
+ "step": 234
1656
+ },
1657
+ {
1658
+ "epoch": 0.4343807763401109,
1659
+ "grad_norm": 1.1094099283218384,
1660
+ "learning_rate": 0.001,
1661
+ "loss": 0.6128,
1662
+ "step": 235
1663
+ },
1664
+ {
1665
+ "epoch": 0.43622920517560076,
1666
+ "grad_norm": 0.765487015247345,
1667
+ "learning_rate": 0.001,
1668
+ "loss": 0.685,
1669
+ "step": 236
1670
+ },
1671
+ {
1672
+ "epoch": 0.43807763401109057,
1673
+ "grad_norm": 0.713035523891449,
1674
+ "learning_rate": 0.001,
1675
+ "loss": 0.6375,
1676
+ "step": 237
1677
+ },
1678
+ {
1679
+ "epoch": 0.4399260628465804,
1680
+ "grad_norm": 1.1346995830535889,
1681
+ "learning_rate": 0.001,
1682
+ "loss": 0.8121,
1683
+ "step": 238
1684
+ },
1685
+ {
1686
+ "epoch": 0.4417744916820702,
1687
+ "grad_norm": 10.435094833374023,
1688
+ "learning_rate": 0.001,
1689
+ "loss": 0.7499,
1690
+ "step": 239
1691
+ },
1692
+ {
1693
+ "epoch": 0.4436229205175601,
1694
+ "grad_norm": 1.442493200302124,
1695
+ "learning_rate": 0.001,
1696
+ "loss": 0.6899,
1697
+ "step": 240
1698
+ },
1699
+ {
1700
+ "epoch": 0.4454713493530499,
1701
+ "grad_norm": 1.5144755840301514,
1702
+ "learning_rate": 0.001,
1703
+ "loss": 0.566,
1704
+ "step": 241
1705
+ },
1706
+ {
1707
+ "epoch": 0.44731977818853974,
1708
+ "grad_norm": 1.2616835832595825,
1709
+ "learning_rate": 0.001,
1710
+ "loss": 0.6848,
1711
+ "step": 242
1712
+ },
1713
+ {
1714
+ "epoch": 0.4491682070240296,
1715
+ "grad_norm": 1.6803869009017944,
1716
+ "learning_rate": 0.001,
1717
+ "loss": 1.9734,
1718
+ "step": 243
1719
+ },
1720
+ {
1721
+ "epoch": 0.4510166358595194,
1722
+ "grad_norm": 1.6784708499908447,
1723
+ "learning_rate": 0.001,
1724
+ "loss": 0.8113,
1725
+ "step": 244
1726
+ },
1727
+ {
1728
+ "epoch": 0.45286506469500926,
1729
+ "grad_norm": 0.6301059126853943,
1730
+ "learning_rate": 0.001,
1731
+ "loss": 0.5922,
1732
+ "step": 245
1733
+ },
1734
+ {
1735
+ "epoch": 0.45471349353049906,
1736
+ "grad_norm": 2.435321807861328,
1737
+ "learning_rate": 0.001,
1738
+ "loss": 0.7648,
1739
+ "step": 246
1740
+ },
1741
+ {
1742
+ "epoch": 0.4565619223659889,
1743
+ "grad_norm": 0.9158065915107727,
1744
+ "learning_rate": 0.001,
1745
+ "loss": 0.6473,
1746
+ "step": 247
1747
+ },
1748
+ {
1749
+ "epoch": 0.4584103512014787,
1750
+ "grad_norm": 1.1794534921646118,
1751
+ "learning_rate": 0.001,
1752
+ "loss": 0.6715,
1753
+ "step": 248
1754
+ },
1755
+ {
1756
+ "epoch": 0.4602587800369686,
1757
+ "grad_norm": 5.524970054626465,
1758
+ "learning_rate": 0.001,
1759
+ "loss": 0.7828,
1760
+ "step": 249
1761
+ },
1762
+ {
1763
+ "epoch": 0.46210720887245843,
1764
+ "grad_norm": 1.7100472450256348,
1765
+ "learning_rate": 0.001,
1766
+ "loss": 0.8253,
1767
+ "step": 250
1768
+ },
1769
+ {
1770
+ "epoch": 0.46395563770794823,
1771
+ "grad_norm": 43.962615966796875,
1772
+ "learning_rate": 0.001,
1773
+ "loss": 7.6281,
1774
+ "step": 251
1775
+ },
1776
+ {
1777
+ "epoch": 0.4658040665434381,
1778
+ "grad_norm": 8.905851364135742,
1779
+ "learning_rate": 0.001,
1780
+ "loss": 1.2425,
1781
+ "step": 252
1782
+ },
1783
+ {
1784
+ "epoch": 0.4676524953789279,
1785
+ "grad_norm": 1.4537878036499023,
1786
+ "learning_rate": 0.001,
1787
+ "loss": 0.6607,
1788
+ "step": 253
1789
+ },
1790
+ {
1791
+ "epoch": 0.46950092421441775,
1792
+ "grad_norm": 1.0906081199645996,
1793
+ "learning_rate": 0.001,
1794
+ "loss": 0.7711,
1795
+ "step": 254
1796
+ },
1797
+ {
1798
+ "epoch": 0.4713493530499076,
1799
+ "grad_norm": 4.024237155914307,
1800
+ "learning_rate": 0.001,
1801
+ "loss": 0.8851,
1802
+ "step": 255
1803
+ },
1804
+ {
1805
+ "epoch": 0.4731977818853974,
1806
+ "grad_norm": 0.7822116017341614,
1807
+ "learning_rate": 0.001,
1808
+ "loss": 0.5662,
1809
+ "step": 256
1810
+ },
1811
+ {
1812
+ "epoch": 0.47504621072088726,
1813
+ "grad_norm": 1.1747384071350098,
1814
+ "learning_rate": 0.001,
1815
+ "loss": 0.6728,
1816
+ "step": 257
1817
+ },
1818
+ {
1819
+ "epoch": 0.47689463955637706,
1820
+ "grad_norm": 1.1935127973556519,
1821
+ "learning_rate": 0.001,
1822
+ "loss": 0.5782,
1823
+ "step": 258
1824
+ },
1825
+ {
1826
+ "epoch": 0.4787430683918669,
1827
+ "grad_norm": 0.5861594080924988,
1828
+ "learning_rate": 0.001,
1829
+ "loss": 0.6337,
1830
+ "step": 259
1831
+ },
1832
+ {
1833
+ "epoch": 0.4805914972273567,
1834
+ "grad_norm": 1.6729638576507568,
1835
+ "learning_rate": 0.001,
1836
+ "loss": 0.7732,
1837
+ "step": 260
1838
+ },
1839
+ {
1840
+ "epoch": 0.4824399260628466,
1841
+ "grad_norm": 6.095468044281006,
1842
+ "learning_rate": 0.001,
1843
+ "loss": 0.8832,
1844
+ "step": 261
1845
+ },
1846
+ {
1847
+ "epoch": 0.48428835489833644,
1848
+ "grad_norm": 366.1634521484375,
1849
+ "learning_rate": 0.001,
1850
+ "loss": 1.8354,
1851
+ "step": 262
1852
+ },
1853
+ {
1854
+ "epoch": 0.48613678373382624,
1855
+ "grad_norm": 119.24848175048828,
1856
+ "learning_rate": 0.001,
1857
+ "loss": 2.4536,
1858
+ "step": 263
1859
+ },
1860
+ {
1861
+ "epoch": 0.4879852125693161,
1862
+ "grad_norm": 1.6337554454803467,
1863
+ "learning_rate": 0.001,
1864
+ "loss": 0.7948,
1865
+ "step": 264
1866
+ },
1867
+ {
1868
+ "epoch": 0.4898336414048059,
1869
+ "grad_norm": 0.7200238704681396,
1870
+ "learning_rate": 0.001,
1871
+ "loss": 0.5746,
1872
+ "step": 265
1873
+ },
1874
+ {
1875
+ "epoch": 0.49168207024029575,
1876
+ "grad_norm": 0.9898024201393127,
1877
+ "learning_rate": 0.001,
1878
+ "loss": 0.9548,
1879
+ "step": 266
1880
+ },
1881
+ {
1882
+ "epoch": 0.49353049907578556,
1883
+ "grad_norm": 48.905426025390625,
1884
+ "learning_rate": 0.001,
1885
+ "loss": 2.7047,
1886
+ "step": 267
1887
+ },
1888
+ {
1889
+ "epoch": 0.4953789279112754,
1890
+ "grad_norm": 2.0077579021453857,
1891
+ "learning_rate": 0.001,
1892
+ "loss": 0.9014,
1893
+ "step": 268
1894
+ },
1895
+ {
1896
+ "epoch": 0.49722735674676527,
1897
+ "grad_norm": 0.7598840594291687,
1898
+ "learning_rate": 0.001,
1899
+ "loss": 0.6205,
1900
+ "step": 269
1901
+ },
1902
+ {
1903
+ "epoch": 0.49907578558225507,
1904
+ "grad_norm": 0.6317667365074158,
1905
+ "learning_rate": 0.001,
1906
+ "loss": 0.6323,
1907
+ "step": 270
1908
+ },
1909
+ {
1910
+ "epoch": 0.5009242144177449,
1911
+ "grad_norm": 1.1243677139282227,
1912
+ "learning_rate": 0.001,
1913
+ "loss": 0.699,
1914
+ "step": 271
1915
+ },
1916
+ {
1917
+ "epoch": 0.5027726432532348,
1918
+ "grad_norm": 0.7714710235595703,
1919
+ "learning_rate": 0.001,
1920
+ "loss": 0.6488,
1921
+ "step": 272
1922
+ },
1923
+ {
1924
+ "epoch": 0.5027726432532348,
1925
+ "eval_loss": 2.399911403656006,
1926
+ "eval_runtime": 36.8979,
1927
+ "eval_samples_per_second": 1.491,
1928
+ "eval_steps_per_second": 0.19,
1929
+ "step": 272
1930
+ },
1931
+ {
1932
+ "epoch": 0.5046210720887245,
1933
+ "grad_norm": 5.736697196960449,
1934
+ "learning_rate": 0.001,
1935
+ "loss": 0.7481,
1936
+ "step": 273
1937
+ },
1938
+ {
1939
+ "epoch": 0.5064695009242144,
1940
+ "grad_norm": 0.6620627641677856,
1941
+ "learning_rate": 0.001,
1942
+ "loss": 0.6781,
1943
+ "step": 274
1944
+ },
1945
+ {
1946
+ "epoch": 0.5083179297597042,
1947
+ "grad_norm": 1.1013222932815552,
1948
+ "learning_rate": 0.001,
1949
+ "loss": 0.8447,
1950
+ "step": 275
1951
+ },
1952
+ {
1953
+ "epoch": 0.5101663585951941,
1954
+ "grad_norm": 0.6351478099822998,
1955
+ "learning_rate": 0.001,
1956
+ "loss": 0.486,
1957
+ "step": 276
1958
+ },
1959
+ {
1960
+ "epoch": 0.512014787430684,
1961
+ "grad_norm": 0.7216677069664001,
1962
+ "learning_rate": 0.001,
1963
+ "loss": 0.6427,
1964
+ "step": 277
1965
+ },
1966
+ {
1967
+ "epoch": 0.5138632162661737,
1968
+ "grad_norm": 0.9114554524421692,
1969
+ "learning_rate": 0.001,
1970
+ "loss": 2.2048,
1971
+ "step": 278
1972
+ },
1973
+ {
1974
+ "epoch": 0.5157116451016636,
1975
+ "grad_norm": 0.9319707155227661,
1976
+ "learning_rate": 0.001,
1977
+ "loss": 0.6169,
1978
+ "step": 279
1979
+ },
1980
+ {
1981
+ "epoch": 0.5175600739371534,
1982
+ "grad_norm": 1.0675102472305298,
1983
+ "learning_rate": 0.001,
1984
+ "loss": 0.8317,
1985
+ "step": 280
1986
+ },
1987
+ {
1988
+ "epoch": 0.5194085027726433,
1989
+ "grad_norm": 1.91627037525177,
1990
+ "learning_rate": 0.001,
1991
+ "loss": 0.7027,
1992
+ "step": 281
1993
+ },
1994
+ {
1995
+ "epoch": 0.5212569316081331,
1996
+ "grad_norm": 1.4550344944000244,
1997
+ "learning_rate": 0.001,
1998
+ "loss": 0.7876,
1999
+ "step": 282
2000
+ },
2001
+ {
2002
+ "epoch": 0.5231053604436229,
2003
+ "grad_norm": 1.139053463935852,
2004
+ "learning_rate": 0.001,
2005
+ "loss": 2.3306,
2006
+ "step": 283
2007
+ },
2008
+ {
2009
+ "epoch": 0.5249537892791127,
2010
+ "grad_norm": 1.3426302671432495,
2011
+ "learning_rate": 0.001,
2012
+ "loss": 0.4473,
2013
+ "step": 284
2014
+ },
2015
+ {
2016
+ "epoch": 0.5268022181146026,
2017
+ "grad_norm": 1.4722453355789185,
2018
+ "learning_rate": 0.001,
2019
+ "loss": 0.7528,
2020
+ "step": 285
2021
+ },
2022
+ {
2023
+ "epoch": 0.5286506469500925,
2024
+ "grad_norm": 0.6636189222335815,
2025
+ "learning_rate": 0.001,
2026
+ "loss": 0.7003,
2027
+ "step": 286
2028
+ },
2029
+ {
2030
+ "epoch": 0.5304990757855823,
2031
+ "grad_norm": 0.6289682388305664,
2032
+ "learning_rate": 0.001,
2033
+ "loss": 0.6891,
2034
+ "step": 287
2035
+ },
2036
+ {
2037
+ "epoch": 0.532347504621072,
2038
+ "grad_norm": 0.518354058265686,
2039
+ "learning_rate": 0.001,
2040
+ "loss": 0.6492,
2041
+ "step": 288
2042
+ },
2043
+ {
2044
+ "epoch": 0.5341959334565619,
2045
+ "grad_norm": 0.6105506420135498,
2046
+ "learning_rate": 0.001,
2047
+ "loss": 0.642,
2048
+ "step": 289
2049
+ },
2050
+ {
2051
+ "epoch": 0.5360443622920518,
2052
+ "grad_norm": 0.663215160369873,
2053
+ "learning_rate": 0.001,
2054
+ "loss": 0.7104,
2055
+ "step": 290
2056
+ },
2057
+ {
2058
+ "epoch": 0.5378927911275416,
2059
+ "grad_norm": 0.6891048550605774,
2060
+ "learning_rate": 0.001,
2061
+ "loss": 0.7431,
2062
+ "step": 291
2063
+ },
2064
+ {
2065
+ "epoch": 0.5397412199630314,
2066
+ "grad_norm": 1.533583641052246,
2067
+ "learning_rate": 0.001,
2068
+ "loss": 1.886,
2069
+ "step": 292
2070
+ },
2071
+ {
2072
+ "epoch": 0.5415896487985212,
2073
+ "grad_norm": 0.933929979801178,
2074
+ "learning_rate": 0.001,
2075
+ "loss": 0.6183,
2076
+ "step": 293
2077
+ },
2078
+ {
2079
+ "epoch": 0.5434380776340111,
2080
+ "grad_norm": 0.7768216729164124,
2081
+ "learning_rate": 0.001,
2082
+ "loss": 0.7606,
2083
+ "step": 294
2084
+ },
2085
+ {
2086
+ "epoch": 0.5452865064695009,
2087
+ "grad_norm": 2.2823166847229004,
2088
+ "learning_rate": 0.001,
2089
+ "loss": 0.7254,
2090
+ "step": 295
2091
+ },
2092
+ {
2093
+ "epoch": 0.5471349353049908,
2094
+ "grad_norm": 0.7304850816726685,
2095
+ "learning_rate": 0.001,
2096
+ "loss": 0.7621,
2097
+ "step": 296
2098
+ },
2099
+ {
2100
+ "epoch": 0.5489833641404805,
2101
+ "grad_norm": 0.593639075756073,
2102
+ "learning_rate": 0.001,
2103
+ "loss": 0.6813,
2104
+ "step": 297
2105
+ },
2106
+ {
2107
+ "epoch": 0.5508317929759704,
2108
+ "grad_norm": 2.0610456466674805,
2109
+ "learning_rate": 0.001,
2110
+ "loss": 1.0654,
2111
+ "step": 298
2112
+ },
2113
+ {
2114
+ "epoch": 0.5526802218114603,
2115
+ "grad_norm": 1.1367557048797607,
2116
+ "learning_rate": 0.001,
2117
+ "loss": 0.6536,
2118
+ "step": 299
2119
+ },
2120
+ {
2121
+ "epoch": 0.5545286506469501,
2122
+ "grad_norm": 5.493245601654053,
2123
+ "learning_rate": 0.001,
2124
+ "loss": 0.6969,
2125
+ "step": 300
2126
+ },
2127
+ {
2128
+ "epoch": 0.55637707948244,
2129
+ "grad_norm": 0.56303471326828,
2130
+ "learning_rate": 0.001,
2131
+ "loss": 0.7435,
2132
+ "step": 301
2133
+ },
2134
+ {
2135
+ "epoch": 0.5582255083179297,
2136
+ "grad_norm": 0.5343788266181946,
2137
+ "learning_rate": 0.001,
2138
+ "loss": 0.6129,
2139
+ "step": 302
2140
+ },
2141
+ {
2142
+ "epoch": 0.5600739371534196,
2143
+ "grad_norm": 0.6194228529930115,
2144
+ "learning_rate": 0.001,
2145
+ "loss": 0.6848,
2146
+ "step": 303
2147
+ },
2148
+ {
2149
+ "epoch": 0.5619223659889094,
2150
+ "grad_norm": 0.7855530977249146,
2151
+ "learning_rate": 0.001,
2152
+ "loss": 0.8936,
2153
+ "step": 304
2154
+ },
2155
+ {
2156
+ "epoch": 0.5637707948243993,
2157
+ "grad_norm": 0.5858953595161438,
2158
+ "learning_rate": 0.001,
2159
+ "loss": 0.7653,
2160
+ "step": 305
2161
+ },
2162
+ {
2163
+ "epoch": 0.5656192236598891,
2164
+ "grad_norm": 48.524574279785156,
2165
+ "learning_rate": 0.001,
2166
+ "loss": 9.1166,
2167
+ "step": 306
2168
+ },
2169
+ {
2170
+ "epoch": 0.5674676524953789,
2171
+ "grad_norm": 0.6151651740074158,
2172
+ "learning_rate": 0.001,
2173
+ "loss": 0.7313,
2174
+ "step": 307
2175
+ },
2176
+ {
2177
+ "epoch": 0.5693160813308688,
2178
+ "grad_norm": 0.9248428344726562,
2179
+ "learning_rate": 0.001,
2180
+ "loss": 0.5781,
2181
+ "step": 308
2182
+ },
2183
+ {
2184
+ "epoch": 0.5711645101663586,
2185
+ "grad_norm": 0.49555471539497375,
2186
+ "learning_rate": 0.001,
2187
+ "loss": 0.5315,
2188
+ "step": 309
2189
+ },
2190
+ {
2191
+ "epoch": 0.5730129390018485,
2192
+ "grad_norm": 0.5464370250701904,
2193
+ "learning_rate": 0.001,
2194
+ "loss": 0.6076,
2195
+ "step": 310
2196
+ },
2197
+ {
2198
+ "epoch": 0.5748613678373382,
2199
+ "grad_norm": 0.6241145730018616,
2200
+ "learning_rate": 0.001,
2201
+ "loss": 0.3794,
2202
+ "step": 311
2203
+ },
2204
+ {
2205
+ "epoch": 0.5767097966728281,
2206
+ "grad_norm": 0.6810182929039001,
2207
+ "learning_rate": 0.001,
2208
+ "loss": 0.8058,
2209
+ "step": 312
2210
+ },
2211
+ {
2212
+ "epoch": 0.5785582255083179,
2213
+ "grad_norm": 0.6118370890617371,
2214
+ "learning_rate": 0.001,
2215
+ "loss": 0.6892,
2216
+ "step": 313
2217
+ },
2218
+ {
2219
+ "epoch": 0.5804066543438078,
2220
+ "grad_norm": 0.8108903169631958,
2221
+ "learning_rate": 0.001,
2222
+ "loss": 0.7409,
2223
+ "step": 314
2224
+ },
2225
+ {
2226
+ "epoch": 0.5822550831792976,
2227
+ "grad_norm": 0.6275304555892944,
2228
+ "learning_rate": 0.001,
2229
+ "loss": 0.5313,
2230
+ "step": 315
2231
+ },
2232
+ {
2233
+ "epoch": 0.5841035120147874,
2234
+ "grad_norm": 0.8600186705589294,
2235
+ "learning_rate": 0.001,
2236
+ "loss": 0.7863,
2237
+ "step": 316
2238
+ },
2239
+ {
2240
+ "epoch": 0.5859519408502772,
2241
+ "grad_norm": 1.5396201610565186,
2242
+ "learning_rate": 0.001,
2243
+ "loss": 0.8304,
2244
+ "step": 317
2245
+ },
2246
+ {
2247
+ "epoch": 0.5878003696857671,
2248
+ "grad_norm": 0.6472541689872742,
2249
+ "learning_rate": 0.001,
2250
+ "loss": 0.5984,
2251
+ "step": 318
2252
+ },
2253
+ {
2254
+ "epoch": 0.589648798521257,
2255
+ "grad_norm": 1.398506760597229,
2256
+ "learning_rate": 0.001,
2257
+ "loss": 0.9278,
2258
+ "step": 319
2259
+ },
2260
+ {
2261
+ "epoch": 0.5914972273567468,
2262
+ "grad_norm": 0.8885821104049683,
2263
+ "learning_rate": 0.001,
2264
+ "loss": 0.7355,
2265
+ "step": 320
2266
+ },
2267
+ {
2268
+ "epoch": 0.5933456561922366,
2269
+ "grad_norm": 0.6919491291046143,
2270
+ "learning_rate": 0.001,
2271
+ "loss": 0.7955,
2272
+ "step": 321
2273
+ },
2274
+ {
2275
+ "epoch": 0.5951940850277264,
2276
+ "grad_norm": 0.5543484687805176,
2277
+ "learning_rate": 0.001,
2278
+ "loss": 0.5661,
2279
+ "step": 322
2280
+ },
2281
+ {
2282
+ "epoch": 0.5970425138632163,
2283
+ "grad_norm": 0.9218202829360962,
2284
+ "learning_rate": 0.001,
2285
+ "loss": 0.6754,
2286
+ "step": 323
2287
+ },
2288
+ {
2289
+ "epoch": 0.5988909426987061,
2290
+ "grad_norm": 0.7282502055168152,
2291
+ "learning_rate": 0.001,
2292
+ "loss": 0.7551,
2293
+ "step": 324
2294
+ },
2295
+ {
2296
+ "epoch": 0.600739371534196,
2297
+ "grad_norm": 0.8452831506729126,
2298
+ "learning_rate": 0.001,
2299
+ "loss": 0.7975,
2300
+ "step": 325
2301
+ },
2302
+ {
2303
+ "epoch": 0.6025878003696857,
2304
+ "grad_norm": 0.6374185085296631,
2305
+ "learning_rate": 0.001,
2306
+ "loss": 0.6155,
2307
+ "step": 326
2308
+ },
2309
+ {
2310
+ "epoch": 0.6044362292051756,
2311
+ "grad_norm": 0.8224793076515198,
2312
+ "learning_rate": 0.001,
2313
+ "loss": 0.8842,
2314
+ "step": 327
2315
+ },
2316
+ {
2317
+ "epoch": 0.6062846580406654,
2318
+ "grad_norm": 0.8151664137840271,
2319
+ "learning_rate": 0.001,
2320
+ "loss": 0.6684,
2321
+ "step": 328
2322
+ },
2323
+ {
2324
+ "epoch": 0.6081330868761553,
2325
+ "grad_norm": 0.8286120295524597,
2326
+ "learning_rate": 0.001,
2327
+ "loss": 0.6752,
2328
+ "step": 329
2329
+ },
2330
+ {
2331
+ "epoch": 0.609981515711645,
2332
+ "grad_norm": 0.6680192351341248,
2333
+ "learning_rate": 0.001,
2334
+ "loss": 0.5938,
2335
+ "step": 330
2336
+ },
2337
+ {
2338
+ "epoch": 0.6118299445471349,
2339
+ "grad_norm": 0.6495516896247864,
2340
+ "learning_rate": 0.001,
2341
+ "loss": 0.6121,
2342
+ "step": 331
2343
+ },
2344
+ {
2345
+ "epoch": 0.6136783733826248,
2346
+ "grad_norm": 0.484153151512146,
2347
+ "learning_rate": 0.001,
2348
+ "loss": 0.5139,
2349
+ "step": 332
2350
+ },
2351
+ {
2352
+ "epoch": 0.6155268022181146,
2353
+ "grad_norm": 0.7952211499214172,
2354
+ "learning_rate": 0.001,
2355
+ "loss": 1.9693,
2356
+ "step": 333
2357
+ },
2358
+ {
2359
+ "epoch": 0.6173752310536045,
2360
+ "grad_norm": 0.6049392819404602,
2361
+ "learning_rate": 0.001,
2362
+ "loss": 0.6355,
2363
+ "step": 334
2364
+ },
2365
+ {
2366
+ "epoch": 0.6192236598890942,
2367
+ "grad_norm": 0.5543565154075623,
2368
+ "learning_rate": 0.001,
2369
+ "loss": 0.5789,
2370
+ "step": 335
2371
+ },
2372
+ {
2373
+ "epoch": 0.6210720887245841,
2374
+ "grad_norm": 0.6608275771141052,
2375
+ "learning_rate": 0.001,
2376
+ "loss": 0.7809,
2377
+ "step": 336
2378
+ },
2379
+ {
2380
+ "epoch": 0.6229205175600739,
2381
+ "grad_norm": 0.7678483724594116,
2382
+ "learning_rate": 0.001,
2383
+ "loss": 0.901,
2384
+ "step": 337
2385
+ },
2386
+ {
2387
+ "epoch": 0.6247689463955638,
2388
+ "grad_norm": 0.7045119404792786,
2389
+ "learning_rate": 0.001,
2390
+ "loss": 0.6893,
2391
+ "step": 338
2392
+ },
2393
+ {
2394
+ "epoch": 0.6266173752310537,
2395
+ "grad_norm": 0.6836982369422913,
2396
+ "learning_rate": 0.001,
2397
+ "loss": 0.6864,
2398
+ "step": 339
2399
+ },
2400
+ {
2401
+ "epoch": 0.6284658040665434,
2402
+ "grad_norm": 0.5288094878196716,
2403
+ "learning_rate": 0.001,
2404
+ "loss": 0.5089,
2405
+ "step": 340
2406
+ },
2407
+ {
2408
+ "epoch": 0.6303142329020333,
2409
+ "grad_norm": 0.6091610193252563,
2410
+ "learning_rate": 0.001,
2411
+ "loss": 0.8265,
2412
+ "step": 341
2413
+ },
2414
+ {
2415
+ "epoch": 0.6321626617375231,
2416
+ "grad_norm": 0.7190775871276855,
2417
+ "learning_rate": 0.001,
2418
+ "loss": 0.5927,
2419
+ "step": 342
2420
+ },
2421
+ {
2422
+ "epoch": 0.634011090573013,
2423
+ "grad_norm": 16.79271697998047,
2424
+ "learning_rate": 0.001,
2425
+ "loss": 2.4376,
2426
+ "step": 343
2427
+ },
2428
+ {
2429
+ "epoch": 0.6358595194085028,
2430
+ "grad_norm": 0.6407002806663513,
2431
+ "learning_rate": 0.001,
2432
+ "loss": 0.6132,
2433
+ "step": 344
2434
+ },
2435
+ {
2436
+ "epoch": 0.6377079482439926,
2437
+ "grad_norm": 0.8412754535675049,
2438
+ "learning_rate": 0.001,
2439
+ "loss": 0.6561,
2440
+ "step": 345
2441
+ },
2442
+ {
2443
+ "epoch": 0.6395563770794824,
2444
+ "grad_norm": 0.6675889492034912,
2445
+ "learning_rate": 0.001,
2446
+ "loss": 0.5961,
2447
+ "step": 346
2448
+ },
2449
+ {
2450
+ "epoch": 0.6414048059149723,
2451
+ "grad_norm": 0.6772903203964233,
2452
+ "learning_rate": 0.001,
2453
+ "loss": 0.7657,
2454
+ "step": 347
2455
+ },
2456
+ {
2457
+ "epoch": 0.6432532347504621,
2458
+ "grad_norm": 0.568170964717865,
2459
+ "learning_rate": 0.001,
2460
+ "loss": 0.5673,
2461
+ "step": 348
2462
+ },
2463
+ {
2464
+ "epoch": 0.6451016635859519,
2465
+ "grad_norm": 0.7691311240196228,
2466
+ "learning_rate": 0.001,
2467
+ "loss": 0.6749,
2468
+ "step": 349
2469
+ },
2470
+ {
2471
+ "epoch": 0.6469500924214417,
2472
+ "grad_norm": 1.38419771194458,
2473
+ "learning_rate": 0.001,
2474
+ "loss": 0.7093,
2475
+ "step": 350
2476
+ },
2477
+ {
2478
+ "epoch": 0.6487985212569316,
2479
+ "grad_norm": 17.263277053833008,
2480
+ "learning_rate": 0.001,
2481
+ "loss": 1.152,
2482
+ "step": 351
2483
+ },
2484
+ {
2485
+ "epoch": 0.6506469500924215,
2486
+ "grad_norm": 1.9294712543487549,
2487
+ "learning_rate": 0.001,
2488
+ "loss": 0.857,
2489
+ "step": 352
2490
+ },
2491
+ {
2492
+ "epoch": 0.6524953789279113,
2493
+ "grad_norm": 0.6687116026878357,
2494
+ "learning_rate": 0.001,
2495
+ "loss": 0.7798,
2496
+ "step": 353
2497
+ },
2498
+ {
2499
+ "epoch": 0.6543438077634011,
2500
+ "grad_norm": 0.600409984588623,
2501
+ "learning_rate": 0.001,
2502
+ "loss": 0.5035,
2503
+ "step": 354
2504
+ },
2505
+ {
2506
+ "epoch": 0.6561922365988909,
2507
+ "grad_norm": 0.7064170837402344,
2508
+ "learning_rate": 0.001,
2509
+ "loss": 0.8408,
2510
+ "step": 355
2511
+ },
2512
+ {
2513
+ "epoch": 0.6580406654343808,
2514
+ "grad_norm": 0.5997954607009888,
2515
+ "learning_rate": 0.001,
2516
+ "loss": 0.7668,
2517
+ "step": 356
2518
+ },
2519
+ {
2520
+ "epoch": 0.6598890942698706,
2521
+ "grad_norm": 0.8595522046089172,
2522
+ "learning_rate": 0.001,
2523
+ "loss": 0.6487,
2524
+ "step": 357
2525
+ },
2526
+ {
2527
+ "epoch": 0.6617375231053605,
2528
+ "grad_norm": 0.8264375329017639,
2529
+ "learning_rate": 0.001,
2530
+ "loss": 0.8053,
2531
+ "step": 358
2532
+ },
2533
+ {
2534
+ "epoch": 0.6635859519408502,
2535
+ "grad_norm": 0.45927339792251587,
2536
+ "learning_rate": 0.001,
2537
+ "loss": 0.6095,
2538
+ "step": 359
2539
+ },
2540
+ {
2541
+ "epoch": 0.6654343807763401,
2542
+ "grad_norm": 0.734035849571228,
2543
+ "learning_rate": 0.001,
2544
+ "loss": 0.6616,
2545
+ "step": 360
2546
+ },
2547
+ {
2548
+ "epoch": 0.66728280961183,
2549
+ "grad_norm": 0.639272928237915,
2550
+ "learning_rate": 0.001,
2551
+ "loss": 0.7055,
2552
+ "step": 361
2553
+ },
2554
+ {
2555
+ "epoch": 0.6691312384473198,
2556
+ "grad_norm": 0.5212395787239075,
2557
+ "learning_rate": 0.001,
2558
+ "loss": 0.6112,
2559
+ "step": 362
2560
+ },
2561
+ {
2562
+ "epoch": 0.6709796672828097,
2563
+ "grad_norm": 0.5960317850112915,
2564
+ "learning_rate": 0.001,
2565
+ "loss": 0.6009,
2566
+ "step": 363
2567
+ },
2568
+ {
2569
+ "epoch": 0.6728280961182994,
2570
+ "grad_norm": 0.7983746528625488,
2571
+ "learning_rate": 0.001,
2572
+ "loss": 0.8488,
2573
+ "step": 364
2574
+ },
2575
+ {
2576
+ "epoch": 0.6746765249537893,
2577
+ "grad_norm": 0.5350251793861389,
2578
+ "learning_rate": 0.001,
2579
+ "loss": 0.6422,
2580
+ "step": 365
2581
+ },
2582
+ {
2583
+ "epoch": 0.6765249537892791,
2584
+ "grad_norm": 34.95353698730469,
2585
+ "learning_rate": 0.001,
2586
+ "loss": 2.1827,
2587
+ "step": 366
2588
+ },
2589
+ {
2590
+ "epoch": 0.678373382624769,
2591
+ "grad_norm": 0.6277118921279907,
2592
+ "learning_rate": 0.001,
2593
+ "loss": 0.636,
2594
+ "step": 367
2595
+ },
2596
+ {
2597
+ "epoch": 0.6802218114602587,
2598
+ "grad_norm": 0.5147830843925476,
2599
+ "learning_rate": 0.001,
2600
+ "loss": 0.6665,
2601
+ "step": 368
2602
+ },
2603
+ {
2604
+ "epoch": 0.6820702402957486,
2605
+ "grad_norm": 0.9916329979896545,
2606
+ "learning_rate": 0.001,
2607
+ "loss": 0.8003,
2608
+ "step": 369
2609
+ },
2610
+ {
2611
+ "epoch": 0.6839186691312384,
2612
+ "grad_norm": 0.49549809098243713,
2613
+ "learning_rate": 0.001,
2614
+ "loss": 0.7006,
2615
+ "step": 370
2616
+ },
2617
+ {
2618
+ "epoch": 0.6857670979667283,
2619
+ "grad_norm": 0.6405476331710815,
2620
+ "learning_rate": 0.001,
2621
+ "loss": 0.6889,
2622
+ "step": 371
2623
+ },
2624
+ {
2625
+ "epoch": 0.6876155268022182,
2626
+ "grad_norm": 17.999553680419922,
2627
+ "learning_rate": 0.001,
2628
+ "loss": 1.08,
2629
+ "step": 372
2630
+ },
2631
+ {
2632
+ "epoch": 0.6894639556377079,
2633
+ "grad_norm": 0.6254942417144775,
2634
+ "learning_rate": 0.001,
2635
+ "loss": 0.7336,
2636
+ "step": 373
2637
+ },
2638
+ {
2639
+ "epoch": 0.6913123844731978,
2640
+ "grad_norm": 0.7496931552886963,
2641
+ "learning_rate": 0.001,
2642
+ "loss": 0.7513,
2643
+ "step": 374
2644
+ },
2645
+ {
2646
+ "epoch": 0.6931608133086876,
2647
+ "grad_norm": 0.7643265128135681,
2648
+ "learning_rate": 0.001,
2649
+ "loss": 0.6091,
2650
+ "step": 375
2651
+ },
2652
+ {
2653
+ "epoch": 0.6950092421441775,
2654
+ "grad_norm": 0.6992492079734802,
2655
+ "learning_rate": 0.001,
2656
+ "loss": 0.6633,
2657
+ "step": 376
2658
+ },
2659
+ {
2660
+ "epoch": 0.6968576709796673,
2661
+ "grad_norm": 1.2083274126052856,
2662
+ "learning_rate": 0.001,
2663
+ "loss": 0.9582,
2664
+ "step": 377
2665
+ },
2666
+ {
2667
+ "epoch": 0.6987060998151571,
2668
+ "grad_norm": 1.052863359451294,
2669
+ "learning_rate": 0.001,
2670
+ "loss": 0.7435,
2671
+ "step": 378
2672
+ },
2673
+ {
2674
+ "epoch": 0.7005545286506469,
2675
+ "grad_norm": 5.199072360992432,
2676
+ "learning_rate": 0.001,
2677
+ "loss": 1.9021,
2678
+ "step": 379
2679
+ },
2680
+ {
2681
+ "epoch": 0.7024029574861368,
2682
+ "grad_norm": 0.6658142805099487,
2683
+ "learning_rate": 0.001,
2684
+ "loss": 0.578,
2685
+ "step": 380
2686
+ },
2687
+ {
2688
+ "epoch": 0.7042513863216266,
2689
+ "grad_norm": 0.9965769052505493,
2690
+ "learning_rate": 0.001,
2691
+ "loss": 0.6766,
2692
+ "step": 381
2693
+ },
2694
+ {
2695
+ "epoch": 0.7060998151571165,
2696
+ "grad_norm": 1.3572355508804321,
2697
+ "learning_rate": 0.001,
2698
+ "loss": 0.925,
2699
+ "step": 382
2700
+ },
2701
+ {
2702
+ "epoch": 0.7079482439926063,
2703
+ "grad_norm": 0.796585202217102,
2704
+ "learning_rate": 0.001,
2705
+ "loss": 0.8363,
2706
+ "step": 383
2707
+ },
2708
+ {
2709
+ "epoch": 0.7097966728280961,
2710
+ "grad_norm": 1.294584035873413,
2711
+ "learning_rate": 0.001,
2712
+ "loss": 0.6018,
2713
+ "step": 384
2714
+ },
2715
+ {
2716
+ "epoch": 0.711645101663586,
2717
+ "grad_norm": 0.5120808482170105,
2718
+ "learning_rate": 0.001,
2719
+ "loss": 0.6919,
2720
+ "step": 385
2721
+ },
2722
+ {
2723
+ "epoch": 0.7134935304990758,
2724
+ "grad_norm": 0.6790513396263123,
2725
+ "learning_rate": 0.001,
2726
+ "loss": 0.7894,
2727
+ "step": 386
2728
+ },
2729
+ {
2730
+ "epoch": 0.7153419593345656,
2731
+ "grad_norm": 0.796168863773346,
2732
+ "learning_rate": 0.001,
2733
+ "loss": 0.7301,
2734
+ "step": 387
2735
+ },
2736
+ {
2737
+ "epoch": 0.7171903881700554,
2738
+ "grad_norm": 0.7078443169593811,
2739
+ "learning_rate": 0.001,
2740
+ "loss": 0.614,
2741
+ "step": 388
2742
+ },
2743
+ {
2744
+ "epoch": 0.7190388170055453,
2745
+ "grad_norm": 0.656707227230072,
2746
+ "learning_rate": 0.001,
2747
+ "loss": 0.7422,
2748
+ "step": 389
2749
+ },
2750
+ {
2751
+ "epoch": 0.7208872458410351,
2752
+ "grad_norm": 1.1826304197311401,
2753
+ "learning_rate": 0.001,
2754
+ "loss": 0.9052,
2755
+ "step": 390
2756
+ },
2757
+ {
2758
+ "epoch": 0.722735674676525,
2759
+ "grad_norm": 1.317765474319458,
2760
+ "learning_rate": 0.001,
2761
+ "loss": 0.7672,
2762
+ "step": 391
2763
+ },
2764
+ {
2765
+ "epoch": 0.7245841035120147,
2766
+ "grad_norm": 0.6513253450393677,
2767
+ "learning_rate": 0.001,
2768
+ "loss": 0.6832,
2769
+ "step": 392
2770
+ },
2771
+ {
2772
+ "epoch": 0.7264325323475046,
2773
+ "grad_norm": 0.6068792939186096,
2774
+ "learning_rate": 0.001,
2775
+ "loss": 0.6591,
2776
+ "step": 393
2777
+ },
2778
+ {
2779
+ "epoch": 0.7282809611829945,
2780
+ "grad_norm": 0.7942211031913757,
2781
+ "learning_rate": 0.001,
2782
+ "loss": 2.1942,
2783
+ "step": 394
2784
+ },
2785
+ {
2786
+ "epoch": 0.7301293900184843,
2787
+ "grad_norm": 0.7121322751045227,
2788
+ "learning_rate": 0.001,
2789
+ "loss": 0.7641,
2790
+ "step": 395
2791
+ },
2792
+ {
2793
+ "epoch": 0.7319778188539742,
2794
+ "grad_norm": 0.5792383551597595,
2795
+ "learning_rate": 0.001,
2796
+ "loss": 0.6137,
2797
+ "step": 396
2798
+ },
2799
+ {
2800
+ "epoch": 0.7338262476894639,
2801
+ "grad_norm": 0.40939533710479736,
2802
+ "learning_rate": 0.001,
2803
+ "loss": 0.4706,
2804
+ "step": 397
2805
+ },
2806
+ {
2807
+ "epoch": 0.7356746765249538,
2808
+ "grad_norm": 0.8896576762199402,
2809
+ "learning_rate": 0.001,
2810
+ "loss": 0.7001,
2811
+ "step": 398
2812
+ },
2813
+ {
2814
+ "epoch": 0.7375231053604436,
2815
+ "grad_norm": 0.5687329173088074,
2816
+ "learning_rate": 0.001,
2817
+ "loss": 0.6926,
2818
+ "step": 399
2819
+ },
2820
+ {
2821
+ "epoch": 0.7393715341959335,
2822
+ "grad_norm": 0.7827416658401489,
2823
+ "learning_rate": 0.001,
2824
+ "loss": 0.8797,
2825
+ "step": 400
2826
+ },
2827
+ {
2828
+ "epoch": 0.7412199630314233,
2829
+ "grad_norm": 0.6891829371452332,
2830
+ "learning_rate": 0.001,
2831
+ "loss": 0.6442,
2832
+ "step": 401
2833
+ },
2834
+ {
2835
+ "epoch": 0.7430683918669131,
2836
+ "grad_norm": 0.9651913046836853,
2837
+ "learning_rate": 0.001,
2838
+ "loss": 0.7578,
2839
+ "step": 402
2840
+ },
2841
+ {
2842
+ "epoch": 0.744916820702403,
2843
+ "grad_norm": 1.134435772895813,
2844
+ "learning_rate": 0.001,
2845
+ "loss": 0.7586,
2846
+ "step": 403
2847
+ },
2848
+ {
2849
+ "epoch": 0.7467652495378928,
2850
+ "grad_norm": 0.7820314764976501,
2851
+ "learning_rate": 0.001,
2852
+ "loss": 0.7503,
2853
+ "step": 404
2854
+ },
2855
+ {
2856
+ "epoch": 0.7486136783733827,
2857
+ "grad_norm": 1.1618527173995972,
2858
+ "learning_rate": 0.001,
2859
+ "loss": 0.7242,
2860
+ "step": 405
2861
+ },
2862
+ {
2863
+ "epoch": 0.7504621072088724,
2864
+ "grad_norm": 0.8468978404998779,
2865
+ "learning_rate": 0.001,
2866
+ "loss": 0.804,
2867
+ "step": 406
2868
+ },
2869
+ {
2870
+ "epoch": 0.7523105360443623,
2871
+ "grad_norm": 0.6495232582092285,
2872
+ "learning_rate": 0.001,
2873
+ "loss": 0.4844,
2874
+ "step": 407
2875
+ },
2876
+ {
2877
+ "epoch": 0.7541589648798521,
2878
+ "grad_norm": 62.157989501953125,
2879
+ "learning_rate": 0.001,
2880
+ "loss": 2.3715,
2881
+ "step": 408
2882
+ },
2883
+ {
2884
+ "epoch": 0.7541589648798521,
2885
+ "eval_loss": 0.6921550631523132,
2886
+ "eval_runtime": 36.9225,
2887
+ "eval_samples_per_second": 1.49,
2888
+ "eval_steps_per_second": 0.19,
2889
+ "step": 408
2890
+ },
2891
+ {
2892
+ "epoch": 0.756007393715342,
2893
+ "grad_norm": 0.9253891110420227,
2894
+ "learning_rate": 0.001,
2895
+ "loss": 1.0238,
2896
+ "step": 409
2897
+ },
2898
+ {
2899
+ "epoch": 0.7578558225508318,
2900
+ "grad_norm": 1.3141051530838013,
2901
+ "learning_rate": 0.001,
2902
+ "loss": 1.2068,
2903
+ "step": 410
2904
+ },
2905
+ {
2906
+ "epoch": 0.7597042513863216,
2907
+ "grad_norm": 0.5310596227645874,
2908
+ "learning_rate": 0.001,
2909
+ "loss": 0.5432,
2910
+ "step": 411
2911
+ },
2912
+ {
2913
+ "epoch": 0.7615526802218114,
2914
+ "grad_norm": 0.9271712899208069,
2915
+ "learning_rate": 0.001,
2916
+ "loss": 0.7643,
2917
+ "step": 412
2918
+ },
2919
+ {
2920
+ "epoch": 0.7634011090573013,
2921
+ "grad_norm": 0.9340744018554688,
2922
+ "learning_rate": 0.001,
2923
+ "loss": 0.8286,
2924
+ "step": 413
2925
+ },
2926
+ {
2927
+ "epoch": 0.7652495378927912,
2928
+ "grad_norm": 1.0287261009216309,
2929
+ "learning_rate": 0.001,
2930
+ "loss": 0.9113,
2931
+ "step": 414
2932
+ },
2933
+ {
2934
+ "epoch": 0.767097966728281,
2935
+ "grad_norm": 0.6159982085227966,
2936
+ "learning_rate": 0.001,
2937
+ "loss": 0.6518,
2938
+ "step": 415
2939
+ },
2940
+ {
2941
+ "epoch": 0.7689463955637708,
2942
+ "grad_norm": 0.6261630654335022,
2943
+ "learning_rate": 0.001,
2944
+ "loss": 0.8691,
2945
+ "step": 416
2946
+ },
2947
+ {
2948
+ "epoch": 0.7707948243992606,
2949
+ "grad_norm": 0.5993965268135071,
2950
+ "learning_rate": 0.001,
2951
+ "loss": 0.7519,
2952
+ "step": 417
2953
+ },
2954
+ {
2955
+ "epoch": 0.7726432532347505,
2956
+ "grad_norm": 0.738401472568512,
2957
+ "learning_rate": 0.001,
2958
+ "loss": 0.6912,
2959
+ "step": 418
2960
+ },
2961
+ {
2962
+ "epoch": 0.7744916820702403,
2963
+ "grad_norm": 0.7634181976318359,
2964
+ "learning_rate": 0.001,
2965
+ "loss": 0.5789,
2966
+ "step": 419
2967
+ },
2968
+ {
2969
+ "epoch": 0.7763401109057301,
2970
+ "grad_norm": 1.065432071685791,
2971
+ "learning_rate": 0.001,
2972
+ "loss": 0.7503,
2973
+ "step": 420
2974
+ },
2975
+ {
2976
+ "epoch": 0.7781885397412199,
2977
+ "grad_norm": 0.9008420705795288,
2978
+ "learning_rate": 0.001,
2979
+ "loss": 0.5917,
2980
+ "step": 421
2981
+ },
2982
+ {
2983
+ "epoch": 0.7800369685767098,
2984
+ "grad_norm": 0.6660546660423279,
2985
+ "learning_rate": 0.001,
2986
+ "loss": 0.685,
2987
+ "step": 422
2988
+ },
2989
+ {
2990
+ "epoch": 0.7818853974121996,
2991
+ "grad_norm": 0.6387668251991272,
2992
+ "learning_rate": 0.001,
2993
+ "loss": 0.7161,
2994
+ "step": 423
2995
+ },
2996
+ {
2997
+ "epoch": 0.7837338262476895,
2998
+ "grad_norm": 1.0390757322311401,
2999
+ "learning_rate": 0.001,
3000
+ "loss": 0.8252,
3001
+ "step": 424
3002
+ },
3003
+ {
3004
+ "epoch": 0.7855822550831792,
3005
+ "grad_norm": 0.7057231664657593,
3006
+ "learning_rate": 0.001,
3007
+ "loss": 0.6607,
3008
+ "step": 425
3009
+ },
3010
+ {
3011
+ "epoch": 0.7874306839186691,
3012
+ "grad_norm": 1.4526139497756958,
3013
+ "learning_rate": 0.001,
3014
+ "loss": 0.7296,
3015
+ "step": 426
3016
+ },
3017
+ {
3018
+ "epoch": 0.789279112754159,
3019
+ "grad_norm": 0.6918472051620483,
3020
+ "learning_rate": 0.001,
3021
+ "loss": 0.6665,
3022
+ "step": 427
3023
+ },
3024
+ {
3025
+ "epoch": 0.7911275415896488,
3026
+ "grad_norm": 0.7483262419700623,
3027
+ "learning_rate": 0.001,
3028
+ "loss": 0.7229,
3029
+ "step": 428
3030
+ },
3031
+ {
3032
+ "epoch": 0.7929759704251387,
3033
+ "grad_norm": 0.5912255048751831,
3034
+ "learning_rate": 0.001,
3035
+ "loss": 0.765,
3036
+ "step": 429
3037
+ },
3038
+ {
3039
+ "epoch": 0.7948243992606284,
3040
+ "grad_norm": 0.8008376955986023,
3041
+ "learning_rate": 0.001,
3042
+ "loss": 0.7337,
3043
+ "step": 430
3044
+ },
3045
+ {
3046
+ "epoch": 0.7966728280961183,
3047
+ "grad_norm": 0.687903106212616,
3048
+ "learning_rate": 0.001,
3049
+ "loss": 0.7869,
3050
+ "step": 431
3051
+ },
3052
+ {
3053
+ "epoch": 0.7985212569316081,
3054
+ "grad_norm": 1.3599655628204346,
3055
+ "learning_rate": 0.001,
3056
+ "loss": 1.124,
3057
+ "step": 432
3058
+ },
3059
+ {
3060
+ "epoch": 0.800369685767098,
3061
+ "grad_norm": 0.8680609464645386,
3062
+ "learning_rate": 0.001,
3063
+ "loss": 0.7317,
3064
+ "step": 433
3065
+ },
3066
+ {
3067
+ "epoch": 0.8022181146025879,
3068
+ "grad_norm": 0.712820291519165,
3069
+ "learning_rate": 0.001,
3070
+ "loss": 0.6848,
3071
+ "step": 434
3072
+ },
3073
+ {
3074
+ "epoch": 0.8040665434380776,
3075
+ "grad_norm": 0.5497101545333862,
3076
+ "learning_rate": 0.001,
3077
+ "loss": 0.7699,
3078
+ "step": 435
3079
+ },
3080
+ {
3081
+ "epoch": 0.8059149722735675,
3082
+ "grad_norm": 0.657538115978241,
3083
+ "learning_rate": 0.001,
3084
+ "loss": 0.6637,
3085
+ "step": 436
3086
+ },
3087
+ {
3088
+ "epoch": 0.8077634011090573,
3089
+ "grad_norm": 0.5894818305969238,
3090
+ "learning_rate": 0.001,
3091
+ "loss": 0.7878,
3092
+ "step": 437
3093
+ },
3094
+ {
3095
+ "epoch": 0.8096118299445472,
3096
+ "grad_norm": 0.596224844455719,
3097
+ "learning_rate": 0.001,
3098
+ "loss": 0.7473,
3099
+ "step": 438
3100
+ },
3101
+ {
3102
+ "epoch": 0.8114602587800369,
3103
+ "grad_norm": 1.2735570669174194,
3104
+ "learning_rate": 0.001,
3105
+ "loss": 0.6826,
3106
+ "step": 439
3107
+ },
3108
+ {
3109
+ "epoch": 0.8133086876155268,
3110
+ "grad_norm": 0.9592067003250122,
3111
+ "learning_rate": 0.001,
3112
+ "loss": 0.9536,
3113
+ "step": 440
3114
+ },
3115
+ {
3116
+ "epoch": 0.8151571164510166,
3117
+ "grad_norm": 0.7559153437614441,
3118
+ "learning_rate": 0.001,
3119
+ "loss": 0.8591,
3120
+ "step": 441
3121
+ },
3122
+ {
3123
+ "epoch": 0.8170055452865065,
3124
+ "grad_norm": 0.49736127257347107,
3125
+ "learning_rate": 0.001,
3126
+ "loss": 0.6603,
3127
+ "step": 442
3128
+ },
3129
+ {
3130
+ "epoch": 0.8188539741219963,
3131
+ "grad_norm": 36.26533126831055,
3132
+ "learning_rate": 0.001,
3133
+ "loss": 2.834,
3134
+ "step": 443
3135
+ },
3136
+ {
3137
+ "epoch": 0.8207024029574861,
3138
+ "grad_norm": 0.7133930325508118,
3139
+ "learning_rate": 0.001,
3140
+ "loss": 0.6979,
3141
+ "step": 444
3142
+ },
3143
+ {
3144
+ "epoch": 0.822550831792976,
3145
+ "grad_norm": 0.5370646119117737,
3146
+ "learning_rate": 0.001,
3147
+ "loss": 0.549,
3148
+ "step": 445
3149
+ },
3150
+ {
3151
+ "epoch": 0.8243992606284658,
3152
+ "grad_norm": 0.4864414930343628,
3153
+ "learning_rate": 0.001,
3154
+ "loss": 0.516,
3155
+ "step": 446
3156
+ },
3157
+ {
3158
+ "epoch": 0.8262476894639557,
3159
+ "grad_norm": 1.0422941446304321,
3160
+ "learning_rate": 0.001,
3161
+ "loss": 0.7549,
3162
+ "step": 447
3163
+ },
3164
+ {
3165
+ "epoch": 0.8280961182994455,
3166
+ "grad_norm": 0.796935498714447,
3167
+ "learning_rate": 0.001,
3168
+ "loss": 1.1122,
3169
+ "step": 448
3170
+ },
3171
+ {
3172
+ "epoch": 0.8299445471349353,
3173
+ "grad_norm": 0.708077073097229,
3174
+ "learning_rate": 0.001,
3175
+ "loss": 0.5895,
3176
+ "step": 449
3177
+ },
3178
+ {
3179
+ "epoch": 0.8317929759704251,
3180
+ "grad_norm": 0.6083373427391052,
3181
+ "learning_rate": 0.001,
3182
+ "loss": 0.7167,
3183
+ "step": 450
3184
+ },
3185
+ {
3186
+ "epoch": 0.833641404805915,
3187
+ "grad_norm": 0.568143904209137,
3188
+ "learning_rate": 0.001,
3189
+ "loss": 0.5395,
3190
+ "step": 451
3191
+ },
3192
+ {
3193
+ "epoch": 0.8354898336414048,
3194
+ "grad_norm": 1.630936861038208,
3195
+ "learning_rate": 0.001,
3196
+ "loss": 0.9712,
3197
+ "step": 452
3198
+ },
3199
+ {
3200
+ "epoch": 0.8373382624768947,
3201
+ "grad_norm": 0.8898182511329651,
3202
+ "learning_rate": 0.001,
3203
+ "loss": 0.7966,
3204
+ "step": 453
3205
+ },
3206
+ {
3207
+ "epoch": 0.8391866913123844,
3208
+ "grad_norm": 0.6210492253303528,
3209
+ "learning_rate": 0.001,
3210
+ "loss": 0.7789,
3211
+ "step": 454
3212
+ },
3213
+ {
3214
+ "epoch": 0.8410351201478743,
3215
+ "grad_norm": 0.5374466180801392,
3216
+ "learning_rate": 0.001,
3217
+ "loss": 0.5124,
3218
+ "step": 455
3219
+ },
3220
+ {
3221
+ "epoch": 0.8428835489833642,
3222
+ "grad_norm": 0.731884241104126,
3223
+ "learning_rate": 0.001,
3224
+ "loss": 0.657,
3225
+ "step": 456
3226
+ },
3227
+ {
3228
+ "epoch": 0.844731977818854,
3229
+ "grad_norm": 0.6285644173622131,
3230
+ "learning_rate": 0.001,
3231
+ "loss": 0.878,
3232
+ "step": 457
3233
+ },
3234
+ {
3235
+ "epoch": 0.8465804066543438,
3236
+ "grad_norm": 0.6428564786911011,
3237
+ "learning_rate": 0.001,
3238
+ "loss": 0.6888,
3239
+ "step": 458
3240
+ },
3241
+ {
3242
+ "epoch": 0.8484288354898336,
3243
+ "grad_norm": 0.9038918614387512,
3244
+ "learning_rate": 0.001,
3245
+ "loss": 0.7343,
3246
+ "step": 459
3247
+ },
3248
+ {
3249
+ "epoch": 0.8502772643253235,
3250
+ "grad_norm": 0.6318102478981018,
3251
+ "learning_rate": 0.001,
3252
+ "loss": 0.803,
3253
+ "step": 460
3254
+ },
3255
+ {
3256
+ "epoch": 0.8521256931608133,
3257
+ "grad_norm": 0.7558817863464355,
3258
+ "learning_rate": 0.001,
3259
+ "loss": 0.847,
3260
+ "step": 461
3261
+ },
3262
+ {
3263
+ "epoch": 0.8539741219963032,
3264
+ "grad_norm": 0.6051446199417114,
3265
+ "learning_rate": 0.001,
3266
+ "loss": 0.7294,
3267
+ "step": 462
3268
+ },
3269
+ {
3270
+ "epoch": 0.8558225508317929,
3271
+ "grad_norm": 0.8236597776412964,
3272
+ "learning_rate": 0.001,
3273
+ "loss": 0.8389,
3274
+ "step": 463
3275
+ },
3276
+ {
3277
+ "epoch": 0.8576709796672828,
3278
+ "grad_norm": 0.5667428970336914,
3279
+ "learning_rate": 0.001,
3280
+ "loss": 0.5978,
3281
+ "step": 464
3282
+ },
3283
+ {
3284
+ "epoch": 0.8595194085027726,
3285
+ "grad_norm": 0.6218537092208862,
3286
+ "learning_rate": 0.001,
3287
+ "loss": 0.6539,
3288
+ "step": 465
3289
+ },
3290
+ {
3291
+ "epoch": 0.8613678373382625,
3292
+ "grad_norm": 0.6580603718757629,
3293
+ "learning_rate": 0.001,
3294
+ "loss": 0.5846,
3295
+ "step": 466
3296
+ },
3297
+ {
3298
+ "epoch": 0.8632162661737524,
3299
+ "grad_norm": 0.8344849348068237,
3300
+ "learning_rate": 0.001,
3301
+ "loss": 0.819,
3302
+ "step": 467
3303
+ },
3304
+ {
3305
+ "epoch": 0.8650646950092421,
3306
+ "grad_norm": 0.6353901028633118,
3307
+ "learning_rate": 0.001,
3308
+ "loss": 0.6255,
3309
+ "step": 468
3310
+ },
3311
+ {
3312
+ "epoch": 0.866913123844732,
3313
+ "grad_norm": 0.5879977345466614,
3314
+ "learning_rate": 0.001,
3315
+ "loss": 0.8907,
3316
+ "step": 469
3317
+ },
3318
+ {
3319
+ "epoch": 0.8687615526802218,
3320
+ "grad_norm": 0.7070827484130859,
3321
+ "learning_rate": 0.001,
3322
+ "loss": 0.7093,
3323
+ "step": 470
3324
+ },
3325
+ {
3326
+ "epoch": 0.8706099815157117,
3327
+ "grad_norm": 0.6677271127700806,
3328
+ "learning_rate": 0.001,
3329
+ "loss": 0.6718,
3330
+ "step": 471
3331
+ },
3332
+ {
3333
+ "epoch": 0.8724584103512015,
3334
+ "grad_norm": 0.5725215673446655,
3335
+ "learning_rate": 0.001,
3336
+ "loss": 0.7003,
3337
+ "step": 472
3338
+ },
3339
+ {
3340
+ "epoch": 0.8743068391866913,
3341
+ "grad_norm": 0.6160258054733276,
3342
+ "learning_rate": 0.001,
3343
+ "loss": 0.6447,
3344
+ "step": 473
3345
+ },
3346
+ {
3347
+ "epoch": 0.8761552680221811,
3348
+ "grad_norm": 0.8822168111801147,
3349
+ "learning_rate": 0.001,
3350
+ "loss": 0.6368,
3351
+ "step": 474
3352
+ },
3353
+ {
3354
+ "epoch": 0.878003696857671,
3355
+ "grad_norm": 1.4470586776733398,
3356
+ "learning_rate": 0.001,
3357
+ "loss": 0.9277,
3358
+ "step": 475
3359
+ },
3360
+ {
3361
+ "epoch": 0.8798521256931608,
3362
+ "grad_norm": 0.6264105439186096,
3363
+ "learning_rate": 0.001,
3364
+ "loss": 0.6375,
3365
+ "step": 476
3366
+ },
3367
+ {
3368
+ "epoch": 0.8817005545286506,
3369
+ "grad_norm": 0.7535873651504517,
3370
+ "learning_rate": 0.001,
3371
+ "loss": 0.6433,
3372
+ "step": 477
3373
+ },
3374
+ {
3375
+ "epoch": 0.8835489833641405,
3376
+ "grad_norm": 0.6162166595458984,
3377
+ "learning_rate": 0.001,
3378
+ "loss": 0.6918,
3379
+ "step": 478
3380
+ },
3381
+ {
3382
+ "epoch": 0.8853974121996303,
3383
+ "grad_norm": 0.6916104555130005,
3384
+ "learning_rate": 0.001,
3385
+ "loss": 0.6636,
3386
+ "step": 479
3387
+ },
3388
+ {
3389
+ "epoch": 0.8872458410351202,
3390
+ "grad_norm": 0.7060137987136841,
3391
+ "learning_rate": 0.001,
3392
+ "loss": 0.4684,
3393
+ "step": 480
3394
+ },
3395
+ {
3396
+ "epoch": 0.88909426987061,
3397
+ "grad_norm": 0.6374285221099854,
3398
+ "learning_rate": 0.001,
3399
+ "loss": 0.9017,
3400
+ "step": 481
3401
+ },
3402
+ {
3403
+ "epoch": 0.8909426987060998,
3404
+ "grad_norm": 0.5305145382881165,
3405
+ "learning_rate": 0.001,
3406
+ "loss": 0.5699,
3407
+ "step": 482
3408
+ },
3409
+ {
3410
+ "epoch": 0.8927911275415896,
3411
+ "grad_norm": 0.6026877164840698,
3412
+ "learning_rate": 0.001,
3413
+ "loss": 0.9496,
3414
+ "step": 483
3415
+ },
3416
+ {
3417
+ "epoch": 0.8946395563770795,
3418
+ "grad_norm": 0.6786549091339111,
3419
+ "learning_rate": 0.001,
3420
+ "loss": 0.804,
3421
+ "step": 484
3422
+ },
3423
+ {
3424
+ "epoch": 0.8964879852125693,
3425
+ "grad_norm": 0.7018754482269287,
3426
+ "learning_rate": 0.001,
3427
+ "loss": 0.737,
3428
+ "step": 485
3429
+ },
3430
+ {
3431
+ "epoch": 0.8983364140480592,
3432
+ "grad_norm": 0.4089026153087616,
3433
+ "learning_rate": 0.001,
3434
+ "loss": 0.5953,
3435
+ "step": 486
3436
+ },
3437
+ {
3438
+ "epoch": 0.9001848428835489,
3439
+ "grad_norm": 0.48161014914512634,
3440
+ "learning_rate": 0.001,
3441
+ "loss": 0.5167,
3442
+ "step": 487
3443
+ },
3444
+ {
3445
+ "epoch": 0.9020332717190388,
3446
+ "grad_norm": 0.7609049677848816,
3447
+ "learning_rate": 0.001,
3448
+ "loss": 0.7631,
3449
+ "step": 488
3450
+ },
3451
+ {
3452
+ "epoch": 0.9038817005545287,
3453
+ "grad_norm": 0.8507420420646667,
3454
+ "learning_rate": 0.001,
3455
+ "loss": 0.8331,
3456
+ "step": 489
3457
+ },
3458
+ {
3459
+ "epoch": 0.9057301293900185,
3460
+ "grad_norm": 0.549686074256897,
3461
+ "learning_rate": 0.001,
3462
+ "loss": 0.5609,
3463
+ "step": 490
3464
+ },
3465
+ {
3466
+ "epoch": 0.9075785582255084,
3467
+ "grad_norm": 0.8549586534500122,
3468
+ "learning_rate": 0.001,
3469
+ "loss": 0.7975,
3470
+ "step": 491
3471
+ },
3472
+ {
3473
+ "epoch": 0.9094269870609981,
3474
+ "grad_norm": 0.5464617609977722,
3475
+ "learning_rate": 0.001,
3476
+ "loss": 0.7603,
3477
+ "step": 492
3478
+ },
3479
+ {
3480
+ "epoch": 0.911275415896488,
3481
+ "grad_norm": 0.8106308579444885,
3482
+ "learning_rate": 0.001,
3483
+ "loss": 0.5753,
3484
+ "step": 493
3485
+ },
3486
+ {
3487
+ "epoch": 0.9131238447319778,
3488
+ "grad_norm": 1.0989980697631836,
3489
+ "learning_rate": 0.001,
3490
+ "loss": 0.8119,
3491
+ "step": 494
3492
+ },
3493
+ {
3494
+ "epoch": 0.9149722735674677,
3495
+ "grad_norm": 53.603729248046875,
3496
+ "learning_rate": 0.001,
3497
+ "loss": 3.5664,
3498
+ "step": 495
3499
+ },
3500
+ {
3501
+ "epoch": 0.9168207024029574,
3502
+ "grad_norm": 0.6865960359573364,
3503
+ "learning_rate": 0.001,
3504
+ "loss": 0.6459,
3505
+ "step": 496
3506
+ },
3507
+ {
3508
+ "epoch": 0.9186691312384473,
3509
+ "grad_norm": 0.7828227877616882,
3510
+ "learning_rate": 0.001,
3511
+ "loss": 0.9751,
3512
+ "step": 497
3513
+ },
3514
+ {
3515
+ "epoch": 0.9205175600739371,
3516
+ "grad_norm": 0.8201976418495178,
3517
+ "learning_rate": 0.001,
3518
+ "loss": 0.9485,
3519
+ "step": 498
3520
+ },
3521
+ {
3522
+ "epoch": 0.922365988909427,
3523
+ "grad_norm": 0.7697083353996277,
3524
+ "learning_rate": 0.001,
3525
+ "loss": 0.8864,
3526
+ "step": 499
3527
+ },
3528
+ {
3529
+ "epoch": 0.9242144177449169,
3530
+ "grad_norm": 0.4081413447856903,
3531
+ "learning_rate": 0.001,
3532
+ "loss": 0.5863,
3533
+ "step": 500
3534
+ },
3535
+ {
3536
+ "epoch": 0.9260628465804066,
3537
+ "grad_norm": 0.592320442199707,
3538
+ "learning_rate": 0.001,
3539
+ "loss": 0.7524,
3540
+ "step": 501
3541
+ },
3542
+ {
3543
+ "epoch": 0.9279112754158965,
3544
+ "grad_norm": 0.4706171452999115,
3545
+ "learning_rate": 0.001,
3546
+ "loss": 0.5649,
3547
+ "step": 502
3548
+ },
3549
+ {
3550
+ "epoch": 0.9297597042513863,
3551
+ "grad_norm": 0.4379749000072479,
3552
+ "learning_rate": 0.001,
3553
+ "loss": 0.5178,
3554
+ "step": 503
3555
+ },
3556
+ {
3557
+ "epoch": 0.9316081330868762,
3558
+ "grad_norm": 0.6452725529670715,
3559
+ "learning_rate": 0.001,
3560
+ "loss": 0.6012,
3561
+ "step": 504
3562
+ },
3563
+ {
3564
+ "epoch": 0.933456561922366,
3565
+ "grad_norm": 0.7428618669509888,
3566
+ "learning_rate": 0.001,
3567
+ "loss": 0.6258,
3568
+ "step": 505
3569
+ },
3570
+ {
3571
+ "epoch": 0.9353049907578558,
3572
+ "grad_norm": 0.7185813188552856,
3573
+ "learning_rate": 0.001,
3574
+ "loss": 0.899,
3575
+ "step": 506
3576
+ },
3577
+ {
3578
+ "epoch": 0.9371534195933456,
3579
+ "grad_norm": 0.8039231896400452,
3580
+ "learning_rate": 0.001,
3581
+ "loss": 0.8567,
3582
+ "step": 507
3583
+ },
3584
+ {
3585
+ "epoch": 0.9390018484288355,
3586
+ "grad_norm": 0.6219761371612549,
3587
+ "learning_rate": 0.001,
3588
+ "loss": 0.6672,
3589
+ "step": 508
3590
+ },
3591
+ {
3592
+ "epoch": 0.9408502772643254,
3593
+ "grad_norm": 0.8613112568855286,
3594
+ "learning_rate": 0.001,
3595
+ "loss": 1.0144,
3596
+ "step": 509
3597
+ },
3598
+ {
3599
+ "epoch": 0.9426987060998152,
3600
+ "grad_norm": 0.6800719499588013,
3601
+ "learning_rate": 0.001,
3602
+ "loss": 0.6831,
3603
+ "step": 510
3604
+ },
3605
+ {
3606
+ "epoch": 0.944547134935305,
3607
+ "grad_norm": 0.8042735457420349,
3608
+ "learning_rate": 0.001,
3609
+ "loss": 0.8125,
3610
+ "step": 511
3611
+ },
3612
+ {
3613
+ "epoch": 0.9463955637707948,
3614
+ "grad_norm": 0.8199442625045776,
3615
+ "learning_rate": 0.001,
3616
+ "loss": 1.2534,
3617
+ "step": 512
3618
+ },
3619
+ {
3620
+ "epoch": 0.9482439926062847,
3621
+ "grad_norm": 6.547449111938477,
3622
+ "learning_rate": 0.001,
3623
+ "loss": 0.8561,
3624
+ "step": 513
3625
+ },
3626
+ {
3627
+ "epoch": 0.9500924214417745,
3628
+ "grad_norm": 1.0684571266174316,
3629
+ "learning_rate": 0.001,
3630
+ "loss": 0.7813,
3631
+ "step": 514
3632
+ },
3633
+ {
3634
+ "epoch": 0.9519408502772643,
3635
+ "grad_norm": 4.597997665405273,
3636
+ "learning_rate": 0.001,
3637
+ "loss": 0.9715,
3638
+ "step": 515
3639
+ },
3640
+ {
3641
+ "epoch": 0.9537892791127541,
3642
+ "grad_norm": 1.0539919137954712,
3643
+ "learning_rate": 0.001,
3644
+ "loss": 0.7927,
3645
+ "step": 516
3646
+ },
3647
+ {
3648
+ "epoch": 0.955637707948244,
3649
+ "grad_norm": 0.7296995520591736,
3650
+ "learning_rate": 0.001,
3651
+ "loss": 0.7152,
3652
+ "step": 517
3653
+ },
3654
+ {
3655
+ "epoch": 0.9574861367837338,
3656
+ "grad_norm": 0.6043654084205627,
3657
+ "learning_rate": 0.001,
3658
+ "loss": 0.6799,
3659
+ "step": 518
3660
+ },
3661
+ {
3662
+ "epoch": 0.9593345656192237,
3663
+ "grad_norm": 0.5910910964012146,
3664
+ "learning_rate": 0.001,
3665
+ "loss": 0.6224,
3666
+ "step": 519
3667
+ },
3668
+ {
3669
+ "epoch": 0.9611829944547134,
3670
+ "grad_norm": 2.4806816577911377,
3671
+ "learning_rate": 0.001,
3672
+ "loss": 0.4854,
3673
+ "step": 520
3674
+ },
3675
+ {
3676
+ "epoch": 0.9630314232902033,
3677
+ "grad_norm": 1.4178262948989868,
3678
+ "learning_rate": 0.001,
3679
+ "loss": 0.9269,
3680
+ "step": 521
3681
+ },
3682
+ {
3683
+ "epoch": 0.9648798521256932,
3684
+ "grad_norm": 0.5048137307167053,
3685
+ "learning_rate": 0.001,
3686
+ "loss": 0.5043,
3687
+ "step": 522
3688
+ },
3689
+ {
3690
+ "epoch": 0.966728280961183,
3691
+ "grad_norm": 1.0623221397399902,
3692
+ "learning_rate": 0.001,
3693
+ "loss": 1.0081,
3694
+ "step": 523
3695
+ },
3696
+ {
3697
+ "epoch": 0.9685767097966729,
3698
+ "grad_norm": 0.6304468512535095,
3699
+ "learning_rate": 0.001,
3700
+ "loss": 0.6194,
3701
+ "step": 524
3702
+ },
3703
+ {
3704
+ "epoch": 0.9704251386321626,
3705
+ "grad_norm": 0.5621396899223328,
3706
+ "learning_rate": 0.001,
3707
+ "loss": 0.4772,
3708
+ "step": 525
3709
+ },
3710
+ {
3711
+ "epoch": 0.9722735674676525,
3712
+ "grad_norm": 0.7014063000679016,
3713
+ "learning_rate": 0.001,
3714
+ "loss": 0.7481,
3715
+ "step": 526
3716
+ },
3717
+ {
3718
+ "epoch": 0.9741219963031423,
3719
+ "grad_norm": 0.7803249359130859,
3720
+ "learning_rate": 0.001,
3721
+ "loss": 0.6818,
3722
+ "step": 527
3723
+ },
3724
+ {
3725
+ "epoch": 0.9759704251386322,
3726
+ "grad_norm": 0.56339430809021,
3727
+ "learning_rate": 0.001,
3728
+ "loss": 0.529,
3729
+ "step": 528
3730
+ },
3731
+ {
3732
+ "epoch": 0.977818853974122,
3733
+ "grad_norm": 1.3150532245635986,
3734
+ "learning_rate": 0.001,
3735
+ "loss": 0.85,
3736
+ "step": 529
3737
+ },
3738
+ {
3739
+ "epoch": 0.9796672828096118,
3740
+ "grad_norm": 0.73233562707901,
3741
+ "learning_rate": 0.001,
3742
+ "loss": 0.6697,
3743
+ "step": 530
3744
+ },
3745
+ {
3746
+ "epoch": 0.9815157116451017,
3747
+ "grad_norm": 15.183834075927734,
3748
+ "learning_rate": 0.001,
3749
+ "loss": 0.6668,
3750
+ "step": 531
3751
+ },
3752
+ {
3753
+ "epoch": 0.9833641404805915,
3754
+ "grad_norm": 0.710564136505127,
3755
+ "learning_rate": 0.001,
3756
+ "loss": 0.5376,
3757
+ "step": 532
3758
+ },
3759
+ {
3760
+ "epoch": 0.9852125693160814,
3761
+ "grad_norm": 0.9392178654670715,
3762
+ "learning_rate": 0.001,
3763
+ "loss": 0.8452,
3764
+ "step": 533
3765
+ },
3766
+ {
3767
+ "epoch": 0.9870609981515711,
3768
+ "grad_norm": 0.6461498737335205,
3769
+ "learning_rate": 0.001,
3770
+ "loss": 0.7486,
3771
+ "step": 534
3772
+ },
3773
+ {
3774
+ "epoch": 0.988909426987061,
3775
+ "grad_norm": 0.8957709670066833,
3776
+ "learning_rate": 0.001,
3777
+ "loss": 0.6943,
3778
+ "step": 535
3779
+ },
3780
+ {
3781
+ "epoch": 0.9907578558225508,
3782
+ "grad_norm": 1.0718097686767578,
3783
+ "learning_rate": 0.001,
3784
+ "loss": 0.7135,
3785
+ "step": 536
3786
+ },
3787
+ {
3788
+ "epoch": 0.9926062846580407,
3789
+ "grad_norm": 0.8124735355377197,
3790
+ "learning_rate": 0.001,
3791
+ "loss": 0.9046,
3792
+ "step": 537
3793
+ },
3794
+ {
3795
+ "epoch": 0.9944547134935305,
3796
+ "grad_norm": 0.6564399600028992,
3797
+ "learning_rate": 0.001,
3798
+ "loss": 0.7277,
3799
+ "step": 538
3800
+ },
3801
+ {
3802
+ "epoch": 0.9963031423290203,
3803
+ "grad_norm": 4.252345085144043,
3804
+ "learning_rate": 0.001,
3805
+ "loss": 0.9342,
3806
+ "step": 539
3807
+ },
3808
+ {
3809
+ "epoch": 0.9981515711645101,
3810
+ "grad_norm": 0.8855632543563843,
3811
+ "learning_rate": 0.001,
3812
+ "loss": 0.7394,
3813
+ "step": 540
3814
+ },
3815
+ {
3816
+ "epoch": 1.0,
3817
+ "grad_norm": 3.0222349166870117,
3818
+ "learning_rate": 0.001,
3819
+ "loss": 0.8512,
3820
+ "step": 541
3821
+ },
3822
+ {
3823
+ "epoch": 1.0018484288354899,
3824
+ "grad_norm": 0.7332062721252441,
3825
+ "learning_rate": 0.001,
3826
+ "loss": 0.7089,
3827
+ "step": 542
3828
+ },
3829
+ {
3830
+ "epoch": 1.0036968576709797,
3831
+ "grad_norm": 0.7275331020355225,
3832
+ "learning_rate": 0.001,
3833
+ "loss": 0.6,
3834
+ "step": 543
3835
+ },
3836
+ {
3837
+ "epoch": 1.0055452865064696,
3838
+ "grad_norm": 0.7912113070487976,
3839
+ "learning_rate": 0.001,
3840
+ "loss": 0.7719,
3841
+ "step": 544
3842
+ },
3843
+ {
3844
+ "epoch": 1.0055452865064696,
3845
+ "eval_loss": 0.6085956692695618,
3846
+ "eval_runtime": 36.9382,
3847
+ "eval_samples_per_second": 1.489,
3848
+ "eval_steps_per_second": 0.19,
3849
+ "step": 544
3850
+ }
3851
+ ],
3852
+ "logging_steps": 1,
3853
+ "max_steps": 1623,
3854
+ "num_input_tokens_seen": 0,
3855
+ "num_train_epochs": 3,
3856
+ "save_steps": 136,
3857
+ "total_flos": 7.871105024380109e+16,
3858
+ "train_batch_size": 1,
3859
+ "trial_name": null,
3860
+ "trial_params": null
3861
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9dc8611a986d3b9b655d3d1b19926c19d0e166cd233c0a740387a39dad9e60f
3
+ size 5048