Hugging Face's logo

lemingshen
/
GPIoT_Task_Decomposition

Safetensors
Model card Files
xet
Community
lemingshen commited on
Commit
1ea676e
·
1 Parent(s): 39e15c2

update

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +6 -0
  2. adapter_config.json +29 -0
  3. adapter_model.safetensors +3 -0
  4. checkpoint-1000/README.md +202 -0
  5. checkpoint-1000/adapter_config.json +29 -0
  6. checkpoint-1000/adapter_model.safetensors +3 -0
  7. checkpoint-1000/optimizer.pt +3 -0
  8. checkpoint-1000/rng_state.pth +3 -0
  9. checkpoint-1000/scheduler.pt +3 -0
  10. checkpoint-1000/special_tokens_map.json +24 -0
  11. checkpoint-1000/tokenizer.json +0 -0
  12. checkpoint-1000/tokenizer.model +3 -0
  13. checkpoint-1000/tokenizer_config.json +43 -0
  14. checkpoint-1000/trainer_state.json +313 -0
  15. checkpoint-1000/training_args.bin +3 -0
  16. checkpoint-10000/README.md +202 -0
  17. checkpoint-10000/adapter_config.json +29 -0
  18. checkpoint-10000/adapter_model.safetensors +3 -0
  19. checkpoint-10000/optimizer.pt +3 -0
  20. checkpoint-10000/rng_state.pth +3 -0
  21. checkpoint-10000/scheduler.pt +3 -0
  22. checkpoint-10000/special_tokens_map.json +24 -0
  23. checkpoint-10000/tokenizer.json +0 -0
  24. checkpoint-10000/tokenizer.model +3 -0
  25. checkpoint-10000/tokenizer_config.json +43 -0
  26. checkpoint-10000/trainer_state.json +2833 -0
  27. checkpoint-10000/training_args.bin +3 -0
  28. checkpoint-13000/README.md +202 -0
  29. checkpoint-13000/adapter_config.json +29 -0
  30. checkpoint-13000/adapter_model.safetensors +3 -0
  31. checkpoint-13000/optimizer.pt +3 -0
  32. checkpoint-13000/rng_state.pth +3 -0
  33. checkpoint-13000/scheduler.pt +3 -0
  34. checkpoint-13000/special_tokens_map.json +24 -0
  35. checkpoint-13000/tokenizer.json +0 -0
  36. checkpoint-13000/tokenizer.model +3 -0
  37. checkpoint-13000/tokenizer_config.json +43 -0
  38. checkpoint-13000/trainer_state.json +3673 -0
  39. checkpoint-13000/training_args.bin +3 -0
  40. checkpoint-5000/README.md +202 -0
  41. checkpoint-5000/adapter_config.json +29 -0
  42. checkpoint-5000/adapter_model.safetensors +3 -0
  43. checkpoint-5000/optimizer.pt +3 -0
  44. checkpoint-5000/rng_state.pth +3 -0
  45. checkpoint-5000/scheduler.pt +3 -0
  46. checkpoint-5000/special_tokens_map.json +24 -0
  47. checkpoint-5000/tokenizer.json +0 -0
  48. checkpoint-5000/tokenizer.model +3 -0
  49. checkpoint-5000/tokenizer_config.json +43 -0
  50. checkpoint-5000/trainer_state.json +1433 -0
README.md CHANGED
@@ -1,3 +1,9 @@
1
  ---
2
  license: mit
3
  ---
 
 
 
 
 
 
 
1
  ---
2
  license: mit
3
  ---
4
+
5
+ # GPIoT: Tailoring Small Language Models for IoT Program Synthesis and Development
6
+ - Foundation model: Llama-2-13b-chat-hf
7
+ - LoRA fine tuned with INT8 quantization
8
+
9
+ ### Feel free to choose any checkpoint
adapter_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf",
5
+ "bias": "lora_only",
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": 0.001,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 64,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "q_proj",
24
+ "v_proj"
25
+ ],
26
+ "task_type": "CAUSAL_LM",
27
+ "use_dora": false,
28
+ "use_rslora": false
29
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86d0ff6d67840efa0f3bb43f925e5e44fdf67a41a26a053e97062f777917fb67
3
+ size 209736952
checkpoint-1000/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: meta-llama/Llama-2-13b-chat-hf
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.12.0
checkpoint-1000/adapter_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf",
5
+ "bias": "lora_only",
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": 0.001,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 64,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "q_proj",
24
+ "v_proj"
25
+ ],
26
+ "task_type": "CAUSAL_LM",
27
+ "use_dora": false,
28
+ "use_rslora": false
29
+ }
checkpoint-1000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8edefefbf568dc991704857cffabbca95b925eb0f62f3af863a93162d0a11bc6
3
+ size 209736952
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b9a0ac956de7513e2a08d3f405bb39bc8b68bc7d1e4f700bde5671dbbd28187
3
+ size 419529285
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:574d0133fbcb68f97b92c9639c0846141a3ee9ff99d17b5b92c2644afac8737d
3
+ size 14575
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:caa069f9a4a007a38bfc435840b395335c193bb1bf78f30a831ff387adb6205a
3
+ size 627
checkpoint-1000/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-1000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<<SYS>>\\n' + system_message + '\\n<</SYS>>\\n\\n' + message['content'] %}{% else %}{% set content = message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ bos_token + '[INST] ' + content.strip() + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ ' ' + content.strip() + ' ' + eos_token }}{% endif %}{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": false,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,313 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.22168033695411218,
5
+ "eval_steps": 500,
6
+ "global_step": 1000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.005542008423852805,
13
+ "grad_norm": 0.07243233174085617,
14
+ "learning_rate": 1.2315270935960592e-05,
15
+ "loss": 1.4594,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.01108401684770561,
20
+ "grad_norm": 0.40484485030174255,
21
+ "learning_rate": 2.4630541871921184e-05,
22
+ "loss": 2.2032,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.016626025271558414,
27
+ "grad_norm": 0.06850667297840118,
28
+ "learning_rate": 3.694581280788178e-05,
29
+ "loss": 1.2931,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.02216803369541122,
34
+ "grad_norm": 0.4395073354244232,
35
+ "learning_rate": 4.926108374384237e-05,
36
+ "loss": 1.5698,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.027710042119264023,
41
+ "grad_norm": 0.077068030834198,
42
+ "learning_rate": 6.157635467980296e-05,
43
+ "loss": 1.0537,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 0.03325205054311683,
48
+ "grad_norm": 0.3282291293144226,
49
+ "learning_rate": 7.389162561576355e-05,
50
+ "loss": 0.9749,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 0.03879405896696963,
55
+ "grad_norm": 0.0593000203371048,
56
+ "learning_rate": 8.620689655172413e-05,
57
+ "loss": 0.9349,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 0.04433606739082244,
62
+ "grad_norm": 0.25612473487854004,
63
+ "learning_rate": 9.852216748768474e-05,
64
+ "loss": 0.8974,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 0.04987807581467524,
69
+ "grad_norm": 0.0757347121834755,
70
+ "learning_rate": 0.00011083743842364534,
71
+ "loss": 0.9081,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 0.055420084238528046,
76
+ "grad_norm": 0.14145499467849731,
77
+ "learning_rate": 0.00012315270935960593,
78
+ "loss": 0.8607,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 0.06096209266238085,
83
+ "grad_norm": 0.07710155844688416,
84
+ "learning_rate": 0.00013546798029556652,
85
+ "loss": 0.8973,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 0.06650410108623366,
90
+ "grad_norm": 0.14791467785835266,
91
+ "learning_rate": 0.0001477832512315271,
92
+ "loss": 0.7924,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 0.07204610951008646,
97
+ "grad_norm": 0.07742594182491302,
98
+ "learning_rate": 0.00016009852216748767,
99
+ "loss": 0.8698,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 0.07758811793393926,
104
+ "grad_norm": 0.14303487539291382,
105
+ "learning_rate": 0.00017241379310344826,
106
+ "loss": 0.786,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 0.08313012635779206,
111
+ "grad_norm": 0.0865108072757721,
112
+ "learning_rate": 0.00018472906403940888,
113
+ "loss": 0.8606,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 0.08867213478164487,
118
+ "grad_norm": 0.7533164024353027,
119
+ "learning_rate": 0.00019704433497536947,
120
+ "loss": 0.807,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 0.09421414320549767,
125
+ "grad_norm": 0.08325570821762085,
126
+ "learning_rate": 0.00019999896617927833,
127
+ "loss": 0.8635,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 0.09975615162935048,
132
+ "grad_norm": 0.1043543666601181,
133
+ "learning_rate": 0.0001999944557842899,
134
+ "loss": 0.7825,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 0.10529816005320328,
139
+ "grad_norm": 0.07949995994567871,
140
+ "learning_rate": 0.0001999863658806385,
141
+ "loss": 0.8379,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 0.11084016847705609,
146
+ "grad_norm": 0.12020070850849152,
147
+ "learning_rate": 0.00019997469675791905,
148
+ "loss": 0.768,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 0.11638217690090889,
153
+ "grad_norm": 0.0803595781326294,
154
+ "learning_rate": 0.00019995944883385196,
155
+ "loss": 0.8487,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 0.1219241853247617,
160
+ "grad_norm": 0.11509452760219574,
161
+ "learning_rate": 0.0001999406226542682,
162
+ "loss": 0.7787,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 0.1274661937486145,
167
+ "grad_norm": 0.07928384840488434,
168
+ "learning_rate": 0.00019991821889308987,
169
+ "loss": 0.8357,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 0.1330082021724673,
174
+ "grad_norm": 0.09423446655273438,
175
+ "learning_rate": 0.00019989223835230606,
176
+ "loss": 0.7564,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 0.1385502105963201,
181
+ "grad_norm": 0.0835939422249794,
182
+ "learning_rate": 0.000199862681961944,
183
+ "loss": 0.8568,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 0.1440922190201729,
188
+ "grad_norm": 0.09292898327112198,
189
+ "learning_rate": 0.0001998295507800359,
190
+ "loss": 0.7612,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 0.1496342274440257,
195
+ "grad_norm": 0.07704215496778488,
196
+ "learning_rate": 0.00019979284599258107,
197
+ "loss": 0.8263,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 0.15517623586787851,
202
+ "grad_norm": 0.10980474948883057,
203
+ "learning_rate": 0.0001997525689135034,
204
+ "loss": 0.7677,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 0.16071824429173132,
209
+ "grad_norm": 0.08016064018011093,
210
+ "learning_rate": 0.0001997087209846043,
211
+ "loss": 0.8344,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 0.16626025271558412,
216
+ "grad_norm": 0.0950881615281105,
217
+ "learning_rate": 0.0001996613037755113,
218
+ "loss": 0.769,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 0.17180226113943692,
223
+ "grad_norm": 0.07932984828948975,
224
+ "learning_rate": 0.00019961031898362152,
225
+ "loss": 0.8156,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 0.17734426956328975,
230
+ "grad_norm": 0.09336528927087784,
231
+ "learning_rate": 0.00019955576843404128,
232
+ "loss": 0.7767,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 0.18288627798714255,
237
+ "grad_norm": 0.08560346812009811,
238
+ "learning_rate": 0.00019949765407952042,
239
+ "loss": 0.8228,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 0.18842828641099535,
244
+ "grad_norm": 0.08475169539451599,
245
+ "learning_rate": 0.00019943597800038267,
246
+ "loss": 0.7669,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 0.19397029483484815,
251
+ "grad_norm": 0.09038034081459045,
252
+ "learning_rate": 0.00019937074240445105,
253
+ "loss": 0.8182,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 0.19951230325870095,
258
+ "grad_norm": 0.09195873886346817,
259
+ "learning_rate": 0.0001993019496269688,
260
+ "loss": 0.7598,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 0.20505431168255375,
265
+ "grad_norm": 0.08655796200037003,
266
+ "learning_rate": 0.0001992296021305159,
267
+ "loss": 0.8167,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 0.21059632010640655,
272
+ "grad_norm": 0.08353498578071594,
273
+ "learning_rate": 0.00019915370250492084,
274
+ "loss": 0.7486,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 0.21613832853025935,
279
+ "grad_norm": 0.09225723147392273,
280
+ "learning_rate": 0.0001990742534671679,
281
+ "loss": 0.8138,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 0.22168033695411218,
286
+ "grad_norm": 0.12104763090610504,
287
+ "learning_rate": 0.00019899125786129997,
288
+ "loss": 0.7153,
289
+ "step": 1000
290
+ }
291
+ ],
292
+ "logging_steps": 25,
293
+ "max_steps": 13533,
294
+ "num_input_tokens_seen": 0,
295
+ "num_train_epochs": 3,
296
+ "save_steps": 200,
297
+ "stateful_callbacks": {
298
+ "TrainerControl": {
299
+ "args": {
300
+ "should_epoch_stop": false,
301
+ "should_evaluate": false,
302
+ "should_log": false,
303
+ "should_save": true,
304
+ "should_training_stop": false
305
+ },
306
+ "attributes": {}
307
+ }
308
+ },
309
+ "total_flos": 4.78733273181696e+17,
310
+ "train_batch_size": 2,
311
+ "trial_name": null,
312
+ "trial_params": null
313
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a13672a0401c8d0a53efc112f53ebd65f36fa9003e87b1710879c58d881a1e1
3
+ size 5051
checkpoint-10000/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: meta-llama/Llama-2-13b-chat-hf
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.12.0
checkpoint-10000/adapter_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf",
5
+ "bias": "lora_only",
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": 0.001,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 64,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "q_proj",
24
+ "v_proj"
25
+ ],
26
+ "task_type": "CAUSAL_LM",
27
+ "use_dora": false,
28
+ "use_rslora": false
29
+ }
checkpoint-10000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65f8840cbbaa700a0942555614b55f1801df615f7179aad0ab8fb0118ea4e486
3
+ size 209736952
checkpoint-10000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8278c49ac2607b3f2bc56bf4fb612b22386249dc19b8a5af3f0b77c326945004
3
+ size 419529285
checkpoint-10000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fff43d80abc63a3828242eaac15ac1d7e14d669caebfbf67ac745dab3c866d81
3
+ size 14575
checkpoint-10000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5baba50c9446306722759cf46e5fbc1991fffa1d9047464636a8aa263559559
3
+ size 627
checkpoint-10000/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-10000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-10000/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-10000/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<<SYS>>\\n' + system_message + '\\n<</SYS>>\\n\\n' + message['content'] %}{% else %}{% set content = message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ bos_token + '[INST] ' + content.strip() + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ ' ' + content.strip() + ' ' + eos_token }}{% endif %}{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": false,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
checkpoint-10000/trainer_state.json ADDED
@@ -0,0 +1,2833 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.216803369541122,
5
+ "eval_steps": 500,
6
+ "global_step": 10000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.005542008423852805,
13
+ "grad_norm": 0.07243233174085617,
14
+ "learning_rate": 1.2315270935960592e-05,
15
+ "loss": 1.4594,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.01108401684770561,
20
+ "grad_norm": 0.40484485030174255,
21
+ "learning_rate": 2.4630541871921184e-05,
22
+ "loss": 2.2032,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.016626025271558414,
27
+ "grad_norm": 0.06850667297840118,
28
+ "learning_rate": 3.694581280788178e-05,
29
+ "loss": 1.2931,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.02216803369541122,
34
+ "grad_norm": 0.4395073354244232,
35
+ "learning_rate": 4.926108374384237e-05,
36
+ "loss": 1.5698,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.027710042119264023,
41
+ "grad_norm": 0.077068030834198,
42
+ "learning_rate": 6.157635467980296e-05,
43
+ "loss": 1.0537,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 0.03325205054311683,
48
+ "grad_norm": 0.3282291293144226,
49
+ "learning_rate": 7.389162561576355e-05,
50
+ "loss": 0.9749,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 0.03879405896696963,
55
+ "grad_norm": 0.0593000203371048,
56
+ "learning_rate": 8.620689655172413e-05,
57
+ "loss": 0.9349,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 0.04433606739082244,
62
+ "grad_norm": 0.25612473487854004,
63
+ "learning_rate": 9.852216748768474e-05,
64
+ "loss": 0.8974,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 0.04987807581467524,
69
+ "grad_norm": 0.0757347121834755,
70
+ "learning_rate": 0.00011083743842364534,
71
+ "loss": 0.9081,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 0.055420084238528046,
76
+ "grad_norm": 0.14145499467849731,
77
+ "learning_rate": 0.00012315270935960593,
78
+ "loss": 0.8607,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 0.06096209266238085,
83
+ "grad_norm": 0.07710155844688416,
84
+ "learning_rate": 0.00013546798029556652,
85
+ "loss": 0.8973,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 0.06650410108623366,
90
+ "grad_norm": 0.14791467785835266,
91
+ "learning_rate": 0.0001477832512315271,
92
+ "loss": 0.7924,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 0.07204610951008646,
97
+ "grad_norm": 0.07742594182491302,
98
+ "learning_rate": 0.00016009852216748767,
99
+ "loss": 0.8698,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 0.07758811793393926,
104
+ "grad_norm": 0.14303487539291382,
105
+ "learning_rate": 0.00017241379310344826,
106
+ "loss": 0.786,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 0.08313012635779206,
111
+ "grad_norm": 0.0865108072757721,
112
+ "learning_rate": 0.00018472906403940888,
113
+ "loss": 0.8606,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 0.08867213478164487,
118
+ "grad_norm": 0.7533164024353027,
119
+ "learning_rate": 0.00019704433497536947,
120
+ "loss": 0.807,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 0.09421414320549767,
125
+ "grad_norm": 0.08325570821762085,
126
+ "learning_rate": 0.00019999896617927833,
127
+ "loss": 0.8635,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 0.09975615162935048,
132
+ "grad_norm": 0.1043543666601181,
133
+ "learning_rate": 0.0001999944557842899,
134
+ "loss": 0.7825,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 0.10529816005320328,
139
+ "grad_norm": 0.07949995994567871,
140
+ "learning_rate": 0.0001999863658806385,
141
+ "loss": 0.8379,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 0.11084016847705609,
146
+ "grad_norm": 0.12020070850849152,
147
+ "learning_rate": 0.00019997469675791905,
148
+ "loss": 0.768,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 0.11638217690090889,
153
+ "grad_norm": 0.0803595781326294,
154
+ "learning_rate": 0.00019995944883385196,
155
+ "loss": 0.8487,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 0.1219241853247617,
160
+ "grad_norm": 0.11509452760219574,
161
+ "learning_rate": 0.0001999406226542682,
162
+ "loss": 0.7787,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 0.1274661937486145,
167
+ "grad_norm": 0.07928384840488434,
168
+ "learning_rate": 0.00019991821889308987,
169
+ "loss": 0.8357,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 0.1330082021724673,
174
+ "grad_norm": 0.09423446655273438,
175
+ "learning_rate": 0.00019989223835230606,
176
+ "loss": 0.7564,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 0.1385502105963201,
181
+ "grad_norm": 0.0835939422249794,
182
+ "learning_rate": 0.000199862681961944,
183
+ "loss": 0.8568,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 0.1440922190201729,
188
+ "grad_norm": 0.09292898327112198,
189
+ "learning_rate": 0.0001998295507800359,
190
+ "loss": 0.7612,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 0.1496342274440257,
195
+ "grad_norm": 0.07704215496778488,
196
+ "learning_rate": 0.00019979284599258107,
197
+ "loss": 0.8263,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 0.15517623586787851,
202
+ "grad_norm": 0.10980474948883057,
203
+ "learning_rate": 0.0001997525689135034,
204
+ "loss": 0.7677,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 0.16071824429173132,
209
+ "grad_norm": 0.08016064018011093,
210
+ "learning_rate": 0.0001997087209846043,
211
+ "loss": 0.8344,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 0.16626025271558412,
216
+ "grad_norm": 0.0950881615281105,
217
+ "learning_rate": 0.0001996613037755113,
218
+ "loss": 0.769,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 0.17180226113943692,
223
+ "grad_norm": 0.07932984828948975,
224
+ "learning_rate": 0.00019961031898362152,
225
+ "loss": 0.8156,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 0.17734426956328975,
230
+ "grad_norm": 0.09336528927087784,
231
+ "learning_rate": 0.00019955576843404128,
232
+ "loss": 0.7767,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 0.18288627798714255,
237
+ "grad_norm": 0.08560346812009811,
238
+ "learning_rate": 0.00019949765407952042,
239
+ "loss": 0.8228,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 0.18842828641099535,
244
+ "grad_norm": 0.08475169539451599,
245
+ "learning_rate": 0.00019943597800038267,
246
+ "loss": 0.7669,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 0.19397029483484815,
251
+ "grad_norm": 0.09038034081459045,
252
+ "learning_rate": 0.00019937074240445105,
253
+ "loss": 0.8182,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 0.19951230325870095,
258
+ "grad_norm": 0.09195873886346817,
259
+ "learning_rate": 0.0001993019496269688,
260
+ "loss": 0.7598,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 0.20505431168255375,
265
+ "grad_norm": 0.08655796200037003,
266
+ "learning_rate": 0.0001992296021305159,
267
+ "loss": 0.8167,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 0.21059632010640655,
272
+ "grad_norm": 0.08353498578071594,
273
+ "learning_rate": 0.00019915370250492084,
274
+ "loss": 0.7486,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 0.21613832853025935,
279
+ "grad_norm": 0.09225723147392273,
280
+ "learning_rate": 0.0001990742534671679,
281
+ "loss": 0.8138,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 0.22168033695411218,
286
+ "grad_norm": 0.12104763090610504,
287
+ "learning_rate": 0.00019899125786129997,
288
+ "loss": 0.7153,
289
+ "step": 1000
290
+ },
291
+ {
292
+ "epoch": 0.22722234537796498,
293
+ "grad_norm": 0.0815986767411232,
294
+ "learning_rate": 0.00019890471865831669,
295
+ "loss": 0.7983,
296
+ "step": 1025
297
+ },
298
+ {
299
+ "epoch": 0.23276435380181779,
300
+ "grad_norm": 0.08845670521259308,
301
+ "learning_rate": 0.00019881463895606805,
302
+ "loss": 0.7187,
303
+ "step": 1050
304
+ },
305
+ {
306
+ "epoch": 0.2383063622256706,
307
+ "grad_norm": 0.0821809321641922,
308
+ "learning_rate": 0.00019872102197914359,
309
+ "loss": 0.804,
310
+ "step": 1075
311
+ },
312
+ {
313
+ "epoch": 0.2438483706495234,
314
+ "grad_norm": 0.08711609989404678,
315
+ "learning_rate": 0.00019862387107875688,
316
+ "loss": 0.7795,
317
+ "step": 1100
318
+ },
319
+ {
320
+ "epoch": 0.2493903790733762,
321
+ "grad_norm": 0.08517508953809738,
322
+ "learning_rate": 0.00019852318973262567,
323
+ "loss": 0.7937,
324
+ "step": 1125
325
+ },
326
+ {
327
+ "epoch": 0.254932387497229,
328
+ "grad_norm": 0.10830071568489075,
329
+ "learning_rate": 0.00019841898154484726,
330
+ "loss": 0.7458,
331
+ "step": 1150
332
+ },
333
+ {
334
+ "epoch": 0.2604743959210818,
335
+ "grad_norm": 0.08541836589574814,
336
+ "learning_rate": 0.0001983112502457696,
337
+ "loss": 0.8131,
338
+ "step": 1175
339
+ },
340
+ {
341
+ "epoch": 0.2660164043449346,
342
+ "grad_norm": 0.08794037252664566,
343
+ "learning_rate": 0.00019819999969185762,
344
+ "loss": 0.7577,
345
+ "step": 1200
346
+ },
347
+ {
348
+ "epoch": 0.2715584127687874,
349
+ "grad_norm": 0.08078176528215408,
350
+ "learning_rate": 0.00019808523386555542,
351
+ "loss": 0.812,
352
+ "step": 1225
353
+ },
354
+ {
355
+ "epoch": 0.2771004211926402,
356
+ "grad_norm": 0.09263130277395248,
357
+ "learning_rate": 0.0001979669568751434,
358
+ "loss": 0.7582,
359
+ "step": 1250
360
+ },
361
+ {
362
+ "epoch": 0.282642429616493,
363
+ "grad_norm": 0.08198932558298111,
364
+ "learning_rate": 0.00019784517295459147,
365
+ "loss": 0.7958,
366
+ "step": 1275
367
+ },
368
+ {
369
+ "epoch": 0.2881844380403458,
370
+ "grad_norm": 0.07858102023601532,
371
+ "learning_rate": 0.00019771988646340725,
372
+ "loss": 0.7744,
373
+ "step": 1300
374
+ },
375
+ {
376
+ "epoch": 0.2937264464641986,
377
+ "grad_norm": 0.0851408839225769,
378
+ "learning_rate": 0.00019759110188648026,
379
+ "loss": 0.7913,
380
+ "step": 1325
381
+ },
382
+ {
383
+ "epoch": 0.2992684548880514,
384
+ "grad_norm": 0.09252189099788666,
385
+ "learning_rate": 0.00019745882383392116,
386
+ "loss": 0.7675,
387
+ "step": 1350
388
+ },
389
+ {
390
+ "epoch": 0.30481046331190426,
391
+ "grad_norm": 0.08306555449962616,
392
+ "learning_rate": 0.0001973230570408968,
393
+ "loss": 0.8059,
394
+ "step": 1375
395
+ },
396
+ {
397
+ "epoch": 0.31035247173575703,
398
+ "grad_norm": 0.0797729641199112,
399
+ "learning_rate": 0.0001971838063674608,
400
+ "loss": 0.7424,
401
+ "step": 1400
402
+ },
403
+ {
404
+ "epoch": 0.31589448015960986,
405
+ "grad_norm": 0.08266165107488632,
406
+ "learning_rate": 0.0001970410767983794,
407
+ "loss": 0.7847,
408
+ "step": 1425
409
+ },
410
+ {
411
+ "epoch": 0.32143648858346263,
412
+ "grad_norm": 0.09364205598831177,
413
+ "learning_rate": 0.00019689487344295322,
414
+ "loss": 0.6924,
415
+ "step": 1450
416
+ },
417
+ {
418
+ "epoch": 0.32697849700731546,
419
+ "grad_norm": 0.08461842685937881,
420
+ "learning_rate": 0.00019674520153483414,
421
+ "loss": 0.8007,
422
+ "step": 1475
423
+ },
424
+ {
425
+ "epoch": 0.33252050543116823,
426
+ "grad_norm": 0.0840207040309906,
427
+ "learning_rate": 0.00019659206643183813,
428
+ "loss": 0.7139,
429
+ "step": 1500
430
+ },
431
+ {
432
+ "epoch": 0.33806251385502106,
433
+ "grad_norm": 0.08344192802906036,
434
+ "learning_rate": 0.00019643547361575343,
435
+ "loss": 0.7982,
436
+ "step": 1525
437
+ },
438
+ {
439
+ "epoch": 0.34360452227887384,
440
+ "grad_norm": 0.07934779673814774,
441
+ "learning_rate": 0.0001962754286921442,
442
+ "loss": 0.7164,
443
+ "step": 1550
444
+ },
445
+ {
446
+ "epoch": 0.34914653070272667,
447
+ "grad_norm": 0.08716201782226562,
448
+ "learning_rate": 0.00019611193739015,
449
+ "loss": 0.7846,
450
+ "step": 1575
451
+ },
452
+ {
453
+ "epoch": 0.3546885391265795,
454
+ "grad_norm": 0.08384064584970474,
455
+ "learning_rate": 0.0001959450055622806,
456
+ "loss": 0.7416,
457
+ "step": 1600
458
+ },
459
+ {
460
+ "epoch": 0.36023054755043227,
461
+ "grad_norm": 0.08661937713623047,
462
+ "learning_rate": 0.0001957746391842066,
463
+ "loss": 0.8075,
464
+ "step": 1625
465
+ },
466
+ {
467
+ "epoch": 0.3657725559742851,
468
+ "grad_norm": 0.09327207505702972,
469
+ "learning_rate": 0.00019560084435454536,
470
+ "loss": 0.7596,
471
+ "step": 1650
472
+ },
473
+ {
474
+ "epoch": 0.37131456439813787,
475
+ "grad_norm": 0.08391096442937851,
476
+ "learning_rate": 0.00019542362729464273,
477
+ "loss": 0.7794,
478
+ "step": 1675
479
+ },
480
+ {
481
+ "epoch": 0.3768565728219907,
482
+ "grad_norm": 0.07694080471992493,
483
+ "learning_rate": 0.00019524299434835052,
484
+ "loss": 0.7424,
485
+ "step": 1700
486
+ },
487
+ {
488
+ "epoch": 0.38239858124584347,
489
+ "grad_norm": 0.08567491173744202,
490
+ "learning_rate": 0.00019505895198179912,
491
+ "loss": 0.7996,
492
+ "step": 1725
493
+ },
494
+ {
495
+ "epoch": 0.3879405896696963,
496
+ "grad_norm": 0.08828684687614441,
497
+ "learning_rate": 0.0001948715067831663,
498
+ "loss": 0.7394,
499
+ "step": 1750
500
+ },
501
+ {
502
+ "epoch": 0.39348259809354913,
503
+ "grad_norm": 0.08347714692354202,
504
+ "learning_rate": 0.00019468066546244117,
505
+ "loss": 0.7734,
506
+ "step": 1775
507
+ },
508
+ {
509
+ "epoch": 0.3990246065174019,
510
+ "grad_norm": 0.07736373692750931,
511
+ "learning_rate": 0.00019448643485118412,
512
+ "loss": 0.7134,
513
+ "step": 1800
514
+ },
515
+ {
516
+ "epoch": 0.40456661494125473,
517
+ "grad_norm": 0.0840897262096405,
518
+ "learning_rate": 0.00019428882190228216,
519
+ "loss": 0.787,
520
+ "step": 1825
521
+ },
522
+ {
523
+ "epoch": 0.4101086233651075,
524
+ "grad_norm": 0.08665871620178223,
525
+ "learning_rate": 0.0001940878336897001,
526
+ "loss": 0.7151,
527
+ "step": 1850
528
+ },
529
+ {
530
+ "epoch": 0.41565063178896033,
531
+ "grad_norm": 0.08358912914991379,
532
+ "learning_rate": 0.0001938834774082274,
533
+ "loss": 0.7982,
534
+ "step": 1875
535
+ },
536
+ {
537
+ "epoch": 0.4211926402128131,
538
+ "grad_norm": 0.07928963005542755,
539
+ "learning_rate": 0.0001936757603732203,
540
+ "loss": 0.7195,
541
+ "step": 1900
542
+ },
543
+ {
544
+ "epoch": 0.42673464863666594,
545
+ "grad_norm": 0.08886470645666122,
546
+ "learning_rate": 0.00019346469002034042,
547
+ "loss": 0.7762,
548
+ "step": 1925
549
+ },
550
+ {
551
+ "epoch": 0.4322766570605187,
552
+ "grad_norm": 0.1071886494755745,
553
+ "learning_rate": 0.00019325027390528822,
554
+ "loss": 0.7453,
555
+ "step": 1950
556
+ },
557
+ {
558
+ "epoch": 0.43781866548437154,
559
+ "grad_norm": 0.08474262803792953,
560
+ "learning_rate": 0.00019303251970353261,
561
+ "loss": 0.7839,
562
+ "step": 1975
563
+ },
564
+ {
565
+ "epoch": 0.44336067390822437,
566
+ "grad_norm": 0.08803894370794296,
567
+ "learning_rate": 0.0001928114352100363,
568
+ "loss": 0.7171,
569
+ "step": 2000
570
+ },
571
+ {
572
+ "epoch": 0.44890268233207714,
573
+ "grad_norm": 0.08429575711488724,
574
+ "learning_rate": 0.00019258702833897665,
575
+ "loss": 0.7781,
576
+ "step": 2025
577
+ },
578
+ {
579
+ "epoch": 0.45444469075592997,
580
+ "grad_norm": 0.08510231226682663,
581
+ "learning_rate": 0.00019235930712346248,
582
+ "loss": 0.6949,
583
+ "step": 2050
584
+ },
585
+ {
586
+ "epoch": 0.45998669917978274,
587
+ "grad_norm": 0.08167176693677902,
588
+ "learning_rate": 0.00019212827971524634,
589
+ "loss": 0.7722,
590
+ "step": 2075
591
+ },
592
+ {
593
+ "epoch": 0.46552870760363557,
594
+ "grad_norm": 0.06542418897151947,
595
+ "learning_rate": 0.00019189395438443278,
596
+ "loss": 0.7203,
597
+ "step": 2100
598
+ },
599
+ {
600
+ "epoch": 0.47107071602748835,
601
+ "grad_norm": 0.08293402194976807,
602
+ "learning_rate": 0.00019165633951918247,
603
+ "loss": 0.7735,
604
+ "step": 2125
605
+ },
606
+ {
607
+ "epoch": 0.4766127244513412,
608
+ "grad_norm": 0.0809284895658493,
609
+ "learning_rate": 0.00019141544362541162,
610
+ "loss": 0.7412,
611
+ "step": 2150
612
+ },
613
+ {
614
+ "epoch": 0.48215473287519395,
615
+ "grad_norm": 0.08212891221046448,
616
+ "learning_rate": 0.00019117127532648773,
617
+ "loss": 0.7629,
618
+ "step": 2175
619
+ },
620
+ {
621
+ "epoch": 0.4876967412990468,
622
+ "grad_norm": 0.08602219074964523,
623
+ "learning_rate": 0.0001909238433629208,
624
+ "loss": 0.6935,
625
+ "step": 2200
626
+ },
627
+ {
628
+ "epoch": 0.4932387497228996,
629
+ "grad_norm": 0.08529417216777802,
630
+ "learning_rate": 0.0001906731565920505,
631
+ "loss": 0.7915,
632
+ "step": 2225
633
+ },
634
+ {
635
+ "epoch": 0.4987807581467524,
636
+ "grad_norm": 0.08774964511394501,
637
+ "learning_rate": 0.00019041922398772897,
638
+ "loss": 0.7359,
639
+ "step": 2250
640
+ },
641
+ {
642
+ "epoch": 0.5043227665706052,
643
+ "grad_norm": 0.08649475872516632,
644
+ "learning_rate": 0.00019016205463999984,
645
+ "loss": 0.7696,
646
+ "step": 2275
647
+ },
648
+ {
649
+ "epoch": 0.509864774994458,
650
+ "grad_norm": 0.0878506749868393,
651
+ "learning_rate": 0.00018990165775477252,
652
+ "loss": 0.7365,
653
+ "step": 2300
654
+ },
655
+ {
656
+ "epoch": 0.5154067834183108,
657
+ "grad_norm": 0.09131711721420288,
658
+ "learning_rate": 0.0001896380426534929,
659
+ "loss": 0.7809,
660
+ "step": 2325
661
+ },
662
+ {
663
+ "epoch": 0.5209487918421636,
664
+ "grad_norm": 0.07379825413227081,
665
+ "learning_rate": 0.00018937121877280957,
666
+ "loss": 0.7029,
667
+ "step": 2350
668
+ },
669
+ {
670
+ "epoch": 0.5264908002660164,
671
+ "grad_norm": 0.08535836637020111,
672
+ "learning_rate": 0.00018910119566423598,
673
+ "loss": 0.7679,
674
+ "step": 2375
675
+ },
676
+ {
677
+ "epoch": 0.5320328086898692,
678
+ "grad_norm": 0.06719771772623062,
679
+ "learning_rate": 0.00018882798299380864,
680
+ "loss": 0.7121,
681
+ "step": 2400
682
+ },
683
+ {
684
+ "epoch": 0.537574817113722,
685
+ "grad_norm": 0.09019796550273895,
686
+ "learning_rate": 0.00018855159054174093,
687
+ "loss": 0.7754,
688
+ "step": 2425
689
+ },
690
+ {
691
+ "epoch": 0.5431168255375748,
692
+ "grad_norm": 0.08144286274909973,
693
+ "learning_rate": 0.0001882720282020732,
694
+ "loss": 0.7255,
695
+ "step": 2450
696
+ },
697
+ {
698
+ "epoch": 0.5486588339614277,
699
+ "grad_norm": 0.08412271738052368,
700
+ "learning_rate": 0.0001879893059823185,
701
+ "loss": 0.7722,
702
+ "step": 2475
703
+ },
704
+ {
705
+ "epoch": 0.5542008423852804,
706
+ "grad_norm": 0.09016039222478867,
707
+ "learning_rate": 0.0001877034340031042,
708
+ "loss": 0.7275,
709
+ "step": 2500
710
+ },
711
+ {
712
+ "epoch": 0.5597428508091332,
713
+ "grad_norm": 0.08850298821926117,
714
+ "learning_rate": 0.00018741442249781,
715
+ "loss": 0.7828,
716
+ "step": 2525
717
+ },
718
+ {
719
+ "epoch": 0.565284859232986,
720
+ "grad_norm": 0.06989564746618271,
721
+ "learning_rate": 0.00018712228181220128,
722
+ "loss": 0.7111,
723
+ "step": 2550
724
+ },
725
+ {
726
+ "epoch": 0.5708268676568389,
727
+ "grad_norm": 0.09214618802070618,
728
+ "learning_rate": 0.00018682702240405906,
729
+ "loss": 0.7752,
730
+ "step": 2575
731
+ },
732
+ {
733
+ "epoch": 0.5763688760806917,
734
+ "grad_norm": 0.07766986638307571,
735
+ "learning_rate": 0.0001865286548428054,
736
+ "loss": 0.7108,
737
+ "step": 2600
738
+ },
739
+ {
740
+ "epoch": 0.5819108845045444,
741
+ "grad_norm": 0.07919591665267944,
742
+ "learning_rate": 0.00018622718980912514,
743
+ "loss": 0.775,
744
+ "step": 2625
745
+ },
746
+ {
747
+ "epoch": 0.5874528929283972,
748
+ "grad_norm": 0.07524783164262772,
749
+ "learning_rate": 0.00018592263809458361,
750
+ "loss": 0.6941,
751
+ "step": 2650
752
+ },
753
+ {
754
+ "epoch": 0.5929949013522501,
755
+ "grad_norm": 0.08549198508262634,
756
+ "learning_rate": 0.00018561501060124024,
757
+ "loss": 0.7718,
758
+ "step": 2675
759
+ },
760
+ {
761
+ "epoch": 0.5985369097761029,
762
+ "grad_norm": 0.08182788640260696,
763
+ "learning_rate": 0.0001853043183412584,
764
+ "loss": 0.7072,
765
+ "step": 2700
766
+ },
767
+ {
768
+ "epoch": 0.6040789181999556,
769
+ "grad_norm": 0.084741972386837,
770
+ "learning_rate": 0.00018499057243651096,
771
+ "loss": 0.7478,
772
+ "step": 2725
773
+ },
774
+ {
775
+ "epoch": 0.6096209266238085,
776
+ "grad_norm": 0.06824459880590439,
777
+ "learning_rate": 0.0001846737841181825,
778
+ "loss": 0.7238,
779
+ "step": 2750
780
+ },
781
+ {
782
+ "epoch": 0.6151629350476613,
783
+ "grad_norm": 0.08315033465623856,
784
+ "learning_rate": 0.00018435396472636704,
785
+ "loss": 0.7597,
786
+ "step": 2775
787
+ },
788
+ {
789
+ "epoch": 0.6207049434715141,
790
+ "grad_norm": 0.07116558402776718,
791
+ "learning_rate": 0.00018403112570966216,
792
+ "loss": 0.7096,
793
+ "step": 2800
794
+ },
795
+ {
796
+ "epoch": 0.6262469518953668,
797
+ "grad_norm": 0.08500215411186218,
798
+ "learning_rate": 0.00018370527862475916,
799
+ "loss": 0.756,
800
+ "step": 2825
801
+ },
802
+ {
803
+ "epoch": 0.6317889603192197,
804
+ "grad_norm": 0.07979004830121994,
805
+ "learning_rate": 0.00018337643513602933,
806
+ "loss": 0.6886,
807
+ "step": 2850
808
+ },
809
+ {
810
+ "epoch": 0.6373309687430725,
811
+ "grad_norm": 0.08140358328819275,
812
+ "learning_rate": 0.00018304460701510652,
813
+ "loss": 0.7648,
814
+ "step": 2875
815
+ },
816
+ {
817
+ "epoch": 0.6428729771669253,
818
+ "grad_norm": 0.07779423147439957,
819
+ "learning_rate": 0.0001827098061404656,
820
+ "loss": 0.7222,
821
+ "step": 2900
822
+ },
823
+ {
824
+ "epoch": 0.6484149855907781,
825
+ "grad_norm": 0.08853591978549957,
826
+ "learning_rate": 0.0001823720444969974,
827
+ "loss": 0.7736,
828
+ "step": 2925
829
+ },
830
+ {
831
+ "epoch": 0.6539569940146309,
832
+ "grad_norm": 0.07350102066993713,
833
+ "learning_rate": 0.0001820313341755795,
834
+ "loss": 0.7256,
835
+ "step": 2950
836
+ },
837
+ {
838
+ "epoch": 0.6594990024384837,
839
+ "grad_norm": 0.08152145147323608,
840
+ "learning_rate": 0.0001816876873726436,
841
+ "loss": 0.7598,
842
+ "step": 2975
843
+ },
844
+ {
845
+ "epoch": 0.6650410108623365,
846
+ "grad_norm": 0.08045897632837296,
847
+ "learning_rate": 0.00018134111638973876,
848
+ "loss": 0.7275,
849
+ "step": 3000
850
+ },
851
+ {
852
+ "epoch": 0.6705830192861894,
853
+ "grad_norm": 0.08514434099197388,
854
+ "learning_rate": 0.00018099163363309123,
855
+ "loss": 0.7688,
856
+ "step": 3025
857
+ },
858
+ {
859
+ "epoch": 0.6761250277100421,
860
+ "grad_norm": 0.060850344598293304,
861
+ "learning_rate": 0.00018063925161316012,
862
+ "loss": 0.7019,
863
+ "step": 3050
864
+ },
865
+ {
866
+ "epoch": 0.6816670361338949,
867
+ "grad_norm": 0.08471492677927017,
868
+ "learning_rate": 0.00018028398294418977,
869
+ "loss": 0.7573,
870
+ "step": 3075
871
+ },
872
+ {
873
+ "epoch": 0.6872090445577477,
874
+ "grad_norm": 0.0642291009426117,
875
+ "learning_rate": 0.00017992584034375798,
876
+ "loss": 0.7108,
877
+ "step": 3100
878
+ },
879
+ {
880
+ "epoch": 0.6927510529816006,
881
+ "grad_norm": 0.09357668459415436,
882
+ "learning_rate": 0.000179564836632321,
883
+ "loss": 0.7478,
884
+ "step": 3125
885
+ },
886
+ {
887
+ "epoch": 0.6982930614054533,
888
+ "grad_norm": 0.07198700308799744,
889
+ "learning_rate": 0.00017920098473275445,
890
+ "loss": 0.6973,
891
+ "step": 3150
892
+ },
893
+ {
894
+ "epoch": 0.7038350698293061,
895
+ "grad_norm": 0.08420095592737198,
896
+ "learning_rate": 0.00017883429766989064,
897
+ "loss": 0.7487,
898
+ "step": 3175
899
+ },
900
+ {
901
+ "epoch": 0.709377078253159,
902
+ "grad_norm": 0.06639819592237473,
903
+ "learning_rate": 0.00017846478857005255,
904
+ "loss": 0.6741,
905
+ "step": 3200
906
+ },
907
+ {
908
+ "epoch": 0.7149190866770118,
909
+ "grad_norm": 0.08200914412736893,
910
+ "learning_rate": 0.00017809247066058378,
911
+ "loss": 0.7526,
912
+ "step": 3225
913
+ },
914
+ {
915
+ "epoch": 0.7204610951008645,
916
+ "grad_norm": 0.07311141490936279,
917
+ "learning_rate": 0.0001777173572693751,
918
+ "loss": 0.677,
919
+ "step": 3250
920
+ },
921
+ {
922
+ "epoch": 0.7260031035247173,
923
+ "grad_norm": 0.08722089231014252,
924
+ "learning_rate": 0.00017733946182438726,
925
+ "loss": 0.7585,
926
+ "step": 3275
927
+ },
928
+ {
929
+ "epoch": 0.7315451119485702,
930
+ "grad_norm": 0.06589449942111969,
931
+ "learning_rate": 0.00017695879785317048,
932
+ "loss": 0.708,
933
+ "step": 3300
934
+ },
935
+ {
936
+ "epoch": 0.737087120372423,
937
+ "grad_norm": 0.08262074738740921,
938
+ "learning_rate": 0.0001765753789823801,
939
+ "loss": 0.749,
940
+ "step": 3325
941
+ },
942
+ {
943
+ "epoch": 0.7426291287962757,
944
+ "grad_norm": 0.07514823973178864,
945
+ "learning_rate": 0.00017618921893728867,
946
+ "loss": 0.6918,
947
+ "step": 3350
948
+ },
949
+ {
950
+ "epoch": 0.7481711372201286,
951
+ "grad_norm": 0.08757175505161285,
952
+ "learning_rate": 0.00017580033154129503,
953
+ "loss": 0.7445,
954
+ "step": 3375
955
+ },
956
+ {
957
+ "epoch": 0.7537131456439814,
958
+ "grad_norm": 0.0716458335518837,
959
+ "learning_rate": 0.0001754087307154289,
960
+ "loss": 0.7122,
961
+ "step": 3400
962
+ },
963
+ {
964
+ "epoch": 0.7592551540678342,
965
+ "grad_norm": 0.08453212678432465,
966
+ "learning_rate": 0.00017501443047785296,
967
+ "loss": 0.7656,
968
+ "step": 3425
969
+ },
970
+ {
971
+ "epoch": 0.7647971624916869,
972
+ "grad_norm": 0.06761575490236282,
973
+ "learning_rate": 0.00017461744494336098,
974
+ "loss": 0.6673,
975
+ "step": 3450
976
+ },
977
+ {
978
+ "epoch": 0.7703391709155398,
979
+ "grad_norm": 0.08577297627925873,
980
+ "learning_rate": 0.0001742177883228724,
981
+ "loss": 0.7494,
982
+ "step": 3475
983
+ },
984
+ {
985
+ "epoch": 0.7758811793393926,
986
+ "grad_norm": 0.05691730976104736,
987
+ "learning_rate": 0.00017381547492292376,
988
+ "loss": 0.6972,
989
+ "step": 3500
990
+ },
991
+ {
992
+ "epoch": 0.7814231877632454,
993
+ "grad_norm": 0.09115194529294968,
994
+ "learning_rate": 0.00017341051914515656,
995
+ "loss": 0.7706,
996
+ "step": 3525
997
+ },
998
+ {
999
+ "epoch": 0.7869651961870983,
1000
+ "grad_norm": 0.07214304804801941,
1001
+ "learning_rate": 0.00017300293548580162,
1002
+ "loss": 0.6807,
1003
+ "step": 3550
1004
+ },
1005
+ {
1006
+ "epoch": 0.792507204610951,
1007
+ "grad_norm": 0.08448139578104019,
1008
+ "learning_rate": 0.00017259273853516028,
1009
+ "loss": 0.7661,
1010
+ "step": 3575
1011
+ },
1012
+ {
1013
+ "epoch": 0.7980492130348038,
1014
+ "grad_norm": 0.08282499015331268,
1015
+ "learning_rate": 0.00017217994297708195,
1016
+ "loss": 0.7391,
1017
+ "step": 3600
1018
+ },
1019
+ {
1020
+ "epoch": 0.8035912214586566,
1021
+ "grad_norm": 0.0804004818201065,
1022
+ "learning_rate": 0.00017176456358843875,
1023
+ "loss": 0.7402,
1024
+ "step": 3625
1025
+ },
1026
+ {
1027
+ "epoch": 0.8091332298825095,
1028
+ "grad_norm": 0.07265755534172058,
1029
+ "learning_rate": 0.00017134661523859622,
1030
+ "loss": 0.7019,
1031
+ "step": 3650
1032
+ },
1033
+ {
1034
+ "epoch": 0.8146752383063622,
1035
+ "grad_norm": 0.08803457766771317,
1036
+ "learning_rate": 0.00017092611288888125,
1037
+ "loss": 0.7572,
1038
+ "step": 3675
1039
+ },
1040
+ {
1041
+ "epoch": 0.820217246730215,
1042
+ "grad_norm": 0.0652441680431366,
1043
+ "learning_rate": 0.0001705030715920464,
1044
+ "loss": 0.706,
1045
+ "step": 3700
1046
+ },
1047
+ {
1048
+ "epoch": 0.8257592551540678,
1049
+ "grad_norm": 0.08185753971338272,
1050
+ "learning_rate": 0.0001700775064917312,
1051
+ "loss": 0.764,
1052
+ "step": 3725
1053
+ },
1054
+ {
1055
+ "epoch": 0.8313012635779207,
1056
+ "grad_norm": 0.0859500914812088,
1057
+ "learning_rate": 0.00016964943282191984,
1058
+ "loss": 0.6927,
1059
+ "step": 3750
1060
+ },
1061
+ {
1062
+ "epoch": 0.8368432720017734,
1063
+ "grad_norm": 0.09176376461982727,
1064
+ "learning_rate": 0.00016921886590639602,
1065
+ "loss": 0.7567,
1066
+ "step": 3775
1067
+ },
1068
+ {
1069
+ "epoch": 0.8423852804256262,
1070
+ "grad_norm": 0.0646485984325409,
1071
+ "learning_rate": 0.0001687858211581943,
1072
+ "loss": 0.6848,
1073
+ "step": 3800
1074
+ },
1075
+ {
1076
+ "epoch": 0.8479272888494791,
1077
+ "grad_norm": 0.08545655012130737,
1078
+ "learning_rate": 0.00016835031407904839,
1079
+ "loss": 0.7546,
1080
+ "step": 3825
1081
+ },
1082
+ {
1083
+ "epoch": 0.8534692972733319,
1084
+ "grad_norm": 0.06338818371295929,
1085
+ "learning_rate": 0.00016791236025883626,
1086
+ "loss": 0.6655,
1087
+ "step": 3850
1088
+ },
1089
+ {
1090
+ "epoch": 0.8590113056971846,
1091
+ "grad_norm": 0.08781229704618454,
1092
+ "learning_rate": 0.00016747197537502205,
1093
+ "loss": 0.7441,
1094
+ "step": 3875
1095
+ },
1096
+ {
1097
+ "epoch": 0.8645533141210374,
1098
+ "grad_norm": 0.06220358610153198,
1099
+ "learning_rate": 0.00016702917519209487,
1100
+ "loss": 0.6795,
1101
+ "step": 3900
1102
+ },
1103
+ {
1104
+ "epoch": 0.8700953225448903,
1105
+ "grad_norm": 0.08917712420225143,
1106
+ "learning_rate": 0.0001665839755610044,
1107
+ "loss": 0.7552,
1108
+ "step": 3925
1109
+ },
1110
+ {
1111
+ "epoch": 0.8756373309687431,
1112
+ "grad_norm": 0.06624036282300949,
1113
+ "learning_rate": 0.00016613639241859355,
1114
+ "loss": 0.6632,
1115
+ "step": 3950
1116
+ },
1117
+ {
1118
+ "epoch": 0.8811793393925959,
1119
+ "grad_norm": 0.08898719400167465,
1120
+ "learning_rate": 0.00016568644178702803,
1121
+ "loss": 0.757,
1122
+ "step": 3975
1123
+ },
1124
+ {
1125
+ "epoch": 0.8867213478164487,
1126
+ "grad_norm": 0.05095354840159416,
1127
+ "learning_rate": 0.0001652341397732227,
1128
+ "loss": 0.6992,
1129
+ "step": 4000
1130
+ },
1131
+ {
1132
+ "epoch": 0.8922633562403015,
1133
+ "grad_norm": 0.08842916786670685,
1134
+ "learning_rate": 0.0001647795025682649,
1135
+ "loss": 0.7504,
1136
+ "step": 4025
1137
+ },
1138
+ {
1139
+ "epoch": 0.8978053646641543,
1140
+ "grad_norm": 0.0758206844329834,
1141
+ "learning_rate": 0.00016432254644683516,
1142
+ "loss": 0.7081,
1143
+ "step": 4050
1144
+ },
1145
+ {
1146
+ "epoch": 0.903347373088007,
1147
+ "grad_norm": 0.0940496176481247,
1148
+ "learning_rate": 0.0001638632877666243,
1149
+ "loss": 0.746,
1150
+ "step": 4075
1151
+ },
1152
+ {
1153
+ "epoch": 0.9088893815118599,
1154
+ "grad_norm": 0.06626766920089722,
1155
+ "learning_rate": 0.00016340174296774804,
1156
+ "loss": 0.6647,
1157
+ "step": 4100
1158
+ },
1159
+ {
1160
+ "epoch": 0.9144313899357127,
1161
+ "grad_norm": 0.08919317275285721,
1162
+ "learning_rate": 0.00016293792857215844,
1163
+ "loss": 0.7516,
1164
+ "step": 4125
1165
+ },
1166
+ {
1167
+ "epoch": 0.9199733983595655,
1168
+ "grad_norm": 0.06990760564804077,
1169
+ "learning_rate": 0.00016247186118305252,
1170
+ "loss": 0.7011,
1171
+ "step": 4150
1172
+ },
1173
+ {
1174
+ "epoch": 0.9255154067834183,
1175
+ "grad_norm": 0.0870794802904129,
1176
+ "learning_rate": 0.00016200355748427782,
1177
+ "loss": 0.7529,
1178
+ "step": 4175
1179
+ },
1180
+ {
1181
+ "epoch": 0.9310574152072711,
1182
+ "grad_norm": 0.06882854551076889,
1183
+ "learning_rate": 0.00016153303423973526,
1184
+ "loss": 0.7005,
1185
+ "step": 4200
1186
+ },
1187
+ {
1188
+ "epoch": 0.9365994236311239,
1189
+ "grad_norm": 0.084992416203022,
1190
+ "learning_rate": 0.0001610603082927789,
1191
+ "loss": 0.7519,
1192
+ "step": 4225
1193
+ },
1194
+ {
1195
+ "epoch": 0.9421414320549767,
1196
+ "grad_norm": 0.0638299211859703,
1197
+ "learning_rate": 0.00016058539656561323,
1198
+ "loss": 0.716,
1199
+ "step": 4250
1200
+ },
1201
+ {
1202
+ "epoch": 0.9476834404788296,
1203
+ "grad_norm": 0.08899606764316559,
1204
+ "learning_rate": 0.00016010831605868715,
1205
+ "loss": 0.7257,
1206
+ "step": 4275
1207
+ },
1208
+ {
1209
+ "epoch": 0.9532254489026823,
1210
+ "grad_norm": 0.06550378352403641,
1211
+ "learning_rate": 0.00015962908385008565,
1212
+ "loss": 0.7174,
1213
+ "step": 4300
1214
+ },
1215
+ {
1216
+ "epoch": 0.9587674573265351,
1217
+ "grad_norm": 0.09001540392637253,
1218
+ "learning_rate": 0.00015914771709491828,
1219
+ "loss": 0.7271,
1220
+ "step": 4325
1221
+ },
1222
+ {
1223
+ "epoch": 0.9643094657503879,
1224
+ "grad_norm": 0.06641615182161331,
1225
+ "learning_rate": 0.000158664233024705,
1226
+ "loss": 0.69,
1227
+ "step": 4350
1228
+ },
1229
+ {
1230
+ "epoch": 0.9698514741742408,
1231
+ "grad_norm": 0.08917039632797241,
1232
+ "learning_rate": 0.0001581786489467596,
1233
+ "loss": 0.7483,
1234
+ "step": 4375
1235
+ },
1236
+ {
1237
+ "epoch": 0.9753934825980936,
1238
+ "grad_norm": 0.05995697155594826,
1239
+ "learning_rate": 0.00015769098224356992,
1240
+ "loss": 0.7033,
1241
+ "step": 4400
1242
+ },
1243
+ {
1244
+ "epoch": 0.9809354910219463,
1245
+ "grad_norm": 0.08998765051364899,
1246
+ "learning_rate": 0.00015720125037217572,
1247
+ "loss": 0.7462,
1248
+ "step": 4425
1249
+ },
1250
+ {
1251
+ "epoch": 0.9864774994457992,
1252
+ "grad_norm": 0.05868702754378319,
1253
+ "learning_rate": 0.00015670947086354376,
1254
+ "loss": 0.6654,
1255
+ "step": 4450
1256
+ },
1257
+ {
1258
+ "epoch": 0.992019507869652,
1259
+ "grad_norm": 0.0880926102399826,
1260
+ "learning_rate": 0.00015621566132194005,
1261
+ "loss": 0.752,
1262
+ "step": 4475
1263
+ },
1264
+ {
1265
+ "epoch": 0.9975615162935048,
1266
+ "grad_norm": 0.08538970351219177,
1267
+ "learning_rate": 0.00015571983942430005,
1268
+ "loss": 0.7338,
1269
+ "step": 4500
1270
+ },
1271
+ {
1272
+ "epoch": 1.0031035247173576,
1273
+ "grad_norm": 0.0827050730586052,
1274
+ "learning_rate": 0.0001552220229195956,
1275
+ "loss": 0.7174,
1276
+ "step": 4525
1277
+ },
1278
+ {
1279
+ "epoch": 1.0086455331412103,
1280
+ "grad_norm": 0.10867294669151306,
1281
+ "learning_rate": 0.00015472222962819955,
1282
+ "loss": 0.7637,
1283
+ "step": 4550
1284
+ },
1285
+ {
1286
+ "epoch": 1.0141875415650632,
1287
+ "grad_norm": 0.08738269656896591,
1288
+ "learning_rate": 0.00015422047744124802,
1289
+ "loss": 0.6247,
1290
+ "step": 4575
1291
+ },
1292
+ {
1293
+ "epoch": 1.019729549988916,
1294
+ "grad_norm": 0.12865987420082092,
1295
+ "learning_rate": 0.0001537167843199998,
1296
+ "loss": 0.7424,
1297
+ "step": 4600
1298
+ },
1299
+ {
1300
+ "epoch": 1.0252715584127687,
1301
+ "grad_norm": 0.08619695156812668,
1302
+ "learning_rate": 0.00015321116829519345,
1303
+ "loss": 0.6461,
1304
+ "step": 4625
1305
+ },
1306
+ {
1307
+ "epoch": 1.0308135668366216,
1308
+ "grad_norm": 0.11726492643356323,
1309
+ "learning_rate": 0.0001527036474664019,
1310
+ "loss": 0.7433,
1311
+ "step": 4650
1312
+ },
1313
+ {
1314
+ "epoch": 1.0363555752604743,
1315
+ "grad_norm": 0.08198727667331696,
1316
+ "learning_rate": 0.0001521942400013844,
1317
+ "loss": 0.6086,
1318
+ "step": 4675
1319
+ },
1320
+ {
1321
+ "epoch": 1.0418975836843272,
1322
+ "grad_norm": 0.11951526254415512,
1323
+ "learning_rate": 0.00015168296413543635,
1324
+ "loss": 0.7521,
1325
+ "step": 4700
1326
+ },
1327
+ {
1328
+ "epoch": 1.04743959210818,
1329
+ "grad_norm": 0.08714735507965088,
1330
+ "learning_rate": 0.0001511698381707363,
1331
+ "loss": 0.631,
1332
+ "step": 4725
1333
+ },
1334
+ {
1335
+ "epoch": 1.0529816005320327,
1336
+ "grad_norm": 0.13869455456733704,
1337
+ "learning_rate": 0.00015065488047569107,
1338
+ "loss": 0.7524,
1339
+ "step": 4750
1340
+ },
1341
+ {
1342
+ "epoch": 1.0585236089558856,
1343
+ "grad_norm": 0.08524268865585327,
1344
+ "learning_rate": 0.00015013810948427794,
1345
+ "loss": 0.6617,
1346
+ "step": 4775
1347
+ },
1348
+ {
1349
+ "epoch": 1.0640656173797385,
1350
+ "grad_norm": 0.11017199605703354,
1351
+ "learning_rate": 0.00014961954369538494,
1352
+ "loss": 0.7598,
1353
+ "step": 4800
1354
+ },
1355
+ {
1356
+ "epoch": 1.0696076258035911,
1357
+ "grad_norm": 0.0834374874830246,
1358
+ "learning_rate": 0.00014909920167214858,
1359
+ "loss": 0.627,
1360
+ "step": 4825
1361
+ },
1362
+ {
1363
+ "epoch": 1.075149634227444,
1364
+ "grad_norm": 0.1357167363166809,
1365
+ "learning_rate": 0.0001485771020412894,
1366
+ "loss": 0.7466,
1367
+ "step": 4850
1368
+ },
1369
+ {
1370
+ "epoch": 1.080691642651297,
1371
+ "grad_norm": 0.08910629153251648,
1372
+ "learning_rate": 0.00014805326349244503,
1373
+ "loss": 0.6238,
1374
+ "step": 4875
1375
+ },
1376
+ {
1377
+ "epoch": 1.0862336510751496,
1378
+ "grad_norm": 0.10706546157598495,
1379
+ "learning_rate": 0.00014752770477750144,
1380
+ "loss": 0.7533,
1381
+ "step": 4900
1382
+ },
1383
+ {
1384
+ "epoch": 1.0917756594990025,
1385
+ "grad_norm": 0.09201759845018387,
1386
+ "learning_rate": 0.00014700044470992136,
1387
+ "loss": 0.6521,
1388
+ "step": 4925
1389
+ },
1390
+ {
1391
+ "epoch": 1.0973176679228553,
1392
+ "grad_norm": 0.14048361778259277,
1393
+ "learning_rate": 0.00014647150216407106,
1394
+ "loss": 0.7412,
1395
+ "step": 4950
1396
+ },
1397
+ {
1398
+ "epoch": 1.102859676346708,
1399
+ "grad_norm": 0.08308299630880356,
1400
+ "learning_rate": 0.00014594089607454454,
1401
+ "loss": 0.6333,
1402
+ "step": 4975
1403
+ },
1404
+ {
1405
+ "epoch": 1.108401684770561,
1406
+ "grad_norm": 0.12057497352361679,
1407
+ "learning_rate": 0.00014540864543548582,
1408
+ "loss": 0.7538,
1409
+ "step": 5000
1410
+ },
1411
+ {
1412
+ "epoch": 1.1139436931944136,
1413
+ "grad_norm": 0.089565709233284,
1414
+ "learning_rate": 0.00014487476929990898,
1415
+ "loss": 0.6662,
1416
+ "step": 5025
1417
+ },
1418
+ {
1419
+ "epoch": 1.1194857016182664,
1420
+ "grad_norm": 0.12125346809625626,
1421
+ "learning_rate": 0.00014433928677901612,
1422
+ "loss": 0.7653,
1423
+ "step": 5050
1424
+ },
1425
+ {
1426
+ "epoch": 1.1250277100421193,
1427
+ "grad_norm": 0.08421044796705246,
1428
+ "learning_rate": 0.00014380221704151318,
1429
+ "loss": 0.615,
1430
+ "step": 5075
1431
+ },
1432
+ {
1433
+ "epoch": 1.130569718465972,
1434
+ "grad_norm": 0.12215881794691086,
1435
+ "learning_rate": 0.0001432635793129239,
1436
+ "loss": 0.7482,
1437
+ "step": 5100
1438
+ },
1439
+ {
1440
+ "epoch": 1.1361117268898249,
1441
+ "grad_norm": 0.08646813780069351,
1442
+ "learning_rate": 0.0001427233928749014,
1443
+ "loss": 0.6292,
1444
+ "step": 5125
1445
+ },
1446
+ {
1447
+ "epoch": 1.1416537353136778,
1448
+ "grad_norm": 0.11372750997543335,
1449
+ "learning_rate": 0.00014218167706453816,
1450
+ "loss": 0.7487,
1451
+ "step": 5150
1452
+ },
1453
+ {
1454
+ "epoch": 1.1471957437375304,
1455
+ "grad_norm": 0.08925063908100128,
1456
+ "learning_rate": 0.00014163845127367362,
1457
+ "loss": 0.6336,
1458
+ "step": 5175
1459
+ },
1460
+ {
1461
+ "epoch": 1.1527377521613833,
1462
+ "grad_norm": 0.12316026538610458,
1463
+ "learning_rate": 0.00014109373494820018,
1464
+ "loss": 0.7566,
1465
+ "step": 5200
1466
+ },
1467
+ {
1468
+ "epoch": 1.1582797605852362,
1469
+ "grad_norm": 0.08784055709838867,
1470
+ "learning_rate": 0.00014054754758736698,
1471
+ "loss": 0.6124,
1472
+ "step": 5225
1473
+ },
1474
+ {
1475
+ "epoch": 1.1638217690090888,
1476
+ "grad_norm": 0.11267346143722534,
1477
+ "learning_rate": 0.0001399999087430819,
1478
+ "loss": 0.7611,
1479
+ "step": 5250
1480
+ },
1481
+ {
1482
+ "epoch": 1.1693637774329417,
1483
+ "grad_norm": 0.08636374026536942,
1484
+ "learning_rate": 0.00013945083801921167,
1485
+ "loss": 0.6561,
1486
+ "step": 5275
1487
+ },
1488
+ {
1489
+ "epoch": 1.1749057858567946,
1490
+ "grad_norm": 0.13902850449085236,
1491
+ "learning_rate": 0.0001389003550708802,
1492
+ "loss": 0.7546,
1493
+ "step": 5300
1494
+ },
1495
+ {
1496
+ "epoch": 1.1804477942806473,
1497
+ "grad_norm": 0.09029743075370789,
1498
+ "learning_rate": 0.0001383484796037648,
1499
+ "loss": 0.6226,
1500
+ "step": 5325
1501
+ },
1502
+ {
1503
+ "epoch": 1.1859898027045002,
1504
+ "grad_norm": 0.10377778112888336,
1505
+ "learning_rate": 0.00013779523137339095,
1506
+ "loss": 0.7422,
1507
+ "step": 5350
1508
+ },
1509
+ {
1510
+ "epoch": 1.1915318111283528,
1511
+ "grad_norm": 0.08857985585927963,
1512
+ "learning_rate": 0.00013724063018442494,
1513
+ "loss": 0.6346,
1514
+ "step": 5375
1515
+ },
1516
+ {
1517
+ "epoch": 1.1970738195522057,
1518
+ "grad_norm": 0.15107670426368713,
1519
+ "learning_rate": 0.0001366846958899651,
1520
+ "loss": 0.7266,
1521
+ "step": 5400
1522
+ },
1523
+ {
1524
+ "epoch": 1.2026158279760586,
1525
+ "grad_norm": 0.09053236246109009,
1526
+ "learning_rate": 0.000136127448390831,
1527
+ "loss": 0.617,
1528
+ "step": 5425
1529
+ },
1530
+ {
1531
+ "epoch": 1.2081578363999113,
1532
+ "grad_norm": 0.11061497032642365,
1533
+ "learning_rate": 0.00013556890763485112,
1534
+ "loss": 0.7631,
1535
+ "step": 5450
1536
+ },
1537
+ {
1538
+ "epoch": 1.2136998448237641,
1539
+ "grad_norm": 0.08849512785673141,
1540
+ "learning_rate": 0.0001350090936161487,
1541
+ "loss": 0.5992,
1542
+ "step": 5475
1543
+ },
1544
+ {
1545
+ "epoch": 1.219241853247617,
1546
+ "grad_norm": 0.12802088260650635,
1547
+ "learning_rate": 0.00013444802637442606,
1548
+ "loss": 0.7389,
1549
+ "step": 5500
1550
+ },
1551
+ {
1552
+ "epoch": 1.2247838616714697,
1553
+ "grad_norm": 0.08612997084856033,
1554
+ "learning_rate": 0.0001338857259942473,
1555
+ "loss": 0.6173,
1556
+ "step": 5525
1557
+ },
1558
+ {
1559
+ "epoch": 1.2303258700953226,
1560
+ "grad_norm": 0.1339733898639679,
1561
+ "learning_rate": 0.0001333222126043192,
1562
+ "loss": 0.7473,
1563
+ "step": 5550
1564
+ },
1565
+ {
1566
+ "epoch": 1.2358678785191755,
1567
+ "grad_norm": 0.08974476903676987,
1568
+ "learning_rate": 0.00013275750637677073,
1569
+ "loss": 0.6224,
1570
+ "step": 5575
1571
+ },
1572
+ {
1573
+ "epoch": 1.2414098869430281,
1574
+ "grad_norm": 0.1431790143251419,
1575
+ "learning_rate": 0.00013219162752643103,
1576
+ "loss": 0.748,
1577
+ "step": 5600
1578
+ },
1579
+ {
1580
+ "epoch": 1.246951895366881,
1581
+ "grad_norm": 0.08545912057161331,
1582
+ "learning_rate": 0.0001316245963101056,
1583
+ "loss": 0.5969,
1584
+ "step": 5625
1585
+ },
1586
+ {
1587
+ "epoch": 1.2524939037907337,
1588
+ "grad_norm": 0.13927248120307922,
1589
+ "learning_rate": 0.00013105643302585137,
1590
+ "loss": 0.752,
1591
+ "step": 5650
1592
+ },
1593
+ {
1594
+ "epoch": 1.2580359122145865,
1595
+ "grad_norm": 0.08908078819513321,
1596
+ "learning_rate": 0.0001304871580122499,
1597
+ "loss": 0.6422,
1598
+ "step": 5675
1599
+ },
1600
+ {
1601
+ "epoch": 1.2635779206384394,
1602
+ "grad_norm": 0.13006670773029327,
1603
+ "learning_rate": 0.00012991679164767942,
1604
+ "loss": 0.7594,
1605
+ "step": 5700
1606
+ },
1607
+ {
1608
+ "epoch": 1.269119929062292,
1609
+ "grad_norm": 0.08754415810108185,
1610
+ "learning_rate": 0.00012934535434958538,
1611
+ "loss": 0.6121,
1612
+ "step": 5725
1613
+ },
1614
+ {
1615
+ "epoch": 1.274661937486145,
1616
+ "grad_norm": 0.15051035583019257,
1617
+ "learning_rate": 0.0001287728665737495,
1618
+ "loss": 0.7414,
1619
+ "step": 5750
1620
+ },
1621
+ {
1622
+ "epoch": 1.2802039459099979,
1623
+ "grad_norm": 0.08734755963087082,
1624
+ "learning_rate": 0.00012819934881355745,
1625
+ "loss": 0.6368,
1626
+ "step": 5775
1627
+ },
1628
+ {
1629
+ "epoch": 1.2857459543338505,
1630
+ "grad_norm": 0.17122209072113037,
1631
+ "learning_rate": 0.0001276248215992654,
1632
+ "loss": 0.7535,
1633
+ "step": 5800
1634
+ },
1635
+ {
1636
+ "epoch": 1.2912879627577034,
1637
+ "grad_norm": 0.08958180993795395,
1638
+ "learning_rate": 0.00012704930549726503,
1639
+ "loss": 0.5941,
1640
+ "step": 5825
1641
+ },
1642
+ {
1643
+ "epoch": 1.2968299711815563,
1644
+ "grad_norm": 0.10746924579143524,
1645
+ "learning_rate": 0.0001264728211093473,
1646
+ "loss": 0.7498,
1647
+ "step": 5850
1648
+ },
1649
+ {
1650
+ "epoch": 1.302371979605409,
1651
+ "grad_norm": 0.08696026355028152,
1652
+ "learning_rate": 0.00012589538907196486,
1653
+ "loss": 0.6451,
1654
+ "step": 5875
1655
+ },
1656
+ {
1657
+ "epoch": 1.3079139880292618,
1658
+ "grad_norm": 0.14526303112506866,
1659
+ "learning_rate": 0.0001253170300554936,
1660
+ "loss": 0.7427,
1661
+ "step": 5900
1662
+ },
1663
+ {
1664
+ "epoch": 1.3134559964531145,
1665
+ "grad_norm": 0.08736992627382278,
1666
+ "learning_rate": 0.0001247377647634924,
1667
+ "loss": 0.632,
1668
+ "step": 5925
1669
+ },
1670
+ {
1671
+ "epoch": 1.3189980048769674,
1672
+ "grad_norm": 0.13362443447113037,
1673
+ "learning_rate": 0.00012415761393196227,
1674
+ "loss": 0.7394,
1675
+ "step": 5950
1676
+ },
1677
+ {
1678
+ "epoch": 1.3245400133008203,
1679
+ "grad_norm": 0.09453903138637543,
1680
+ "learning_rate": 0.00012357659832860386,
1681
+ "loss": 0.6394,
1682
+ "step": 5975
1683
+ },
1684
+ {
1685
+ "epoch": 1.3300820217246732,
1686
+ "grad_norm": 0.1354217529296875,
1687
+ "learning_rate": 0.00012299473875207416,
1688
+ "loss": 0.727,
1689
+ "step": 6000
1690
+ },
1691
+ {
1692
+ "epoch": 1.3356240301485258,
1693
+ "grad_norm": 0.09107760339975357,
1694
+ "learning_rate": 0.0001224120560312419,
1695
+ "loss": 0.6197,
1696
+ "step": 6025
1697
+ },
1698
+ {
1699
+ "epoch": 1.3411660385723787,
1700
+ "grad_norm": 0.12501803040504456,
1701
+ "learning_rate": 0.00012182857102444203,
1702
+ "loss": 0.7475,
1703
+ "step": 6050
1704
+ },
1705
+ {
1706
+ "epoch": 1.3467080469962314,
1707
+ "grad_norm": 0.08888363093137741,
1708
+ "learning_rate": 0.00012124430461872886,
1709
+ "loss": 0.6108,
1710
+ "step": 6075
1711
+ },
1712
+ {
1713
+ "epoch": 1.3522500554200843,
1714
+ "grad_norm": 0.16767624020576477,
1715
+ "learning_rate": 0.00012065927772912863,
1716
+ "loss": 0.7408,
1717
+ "step": 6100
1718
+ },
1719
+ {
1720
+ "epoch": 1.3577920638439371,
1721
+ "grad_norm": 0.09112541377544403,
1722
+ "learning_rate": 0.00012007351129789062,
1723
+ "loss": 0.5868,
1724
+ "step": 6125
1725
+ },
1726
+ {
1727
+ "epoch": 1.3633340722677898,
1728
+ "grad_norm": 0.13539327681064606,
1729
+ "learning_rate": 0.0001194870262937375,
1730
+ "loss": 0.7505,
1731
+ "step": 6150
1732
+ },
1733
+ {
1734
+ "epoch": 1.3688760806916427,
1735
+ "grad_norm": 0.08977732807397842,
1736
+ "learning_rate": 0.00011889984371111475,
1737
+ "loss": 0.5985,
1738
+ "step": 6175
1739
+ },
1740
+ {
1741
+ "epoch": 1.3744180891154953,
1742
+ "grad_norm": 0.17703984677791595,
1743
+ "learning_rate": 0.00011831198456943924,
1744
+ "loss": 0.7334,
1745
+ "step": 6200
1746
+ },
1747
+ {
1748
+ "epoch": 1.3799600975393482,
1749
+ "grad_norm": 0.09067991375923157,
1750
+ "learning_rate": 0.00011772346991234651,
1751
+ "loss": 0.5874,
1752
+ "step": 6225
1753
+ },
1754
+ {
1755
+ "epoch": 1.3855021059632011,
1756
+ "grad_norm": 0.10922655463218689,
1757
+ "learning_rate": 0.00011713432080693772,
1758
+ "loss": 0.746,
1759
+ "step": 6250
1760
+ },
1761
+ {
1762
+ "epoch": 1.391044114387054,
1763
+ "grad_norm": 0.08802726864814758,
1764
+ "learning_rate": 0.00011654455834302535,
1765
+ "loss": 0.6084,
1766
+ "step": 6275
1767
+ },
1768
+ {
1769
+ "epoch": 1.3965861228109067,
1770
+ "grad_norm": 0.11013362556695938,
1771
+ "learning_rate": 0.00011595420363237844,
1772
+ "loss": 0.7431,
1773
+ "step": 6300
1774
+ },
1775
+ {
1776
+ "epoch": 1.4021281312347595,
1777
+ "grad_norm": 0.09353320300579071,
1778
+ "learning_rate": 0.00011536327780796661,
1779
+ "loss": 0.6504,
1780
+ "step": 6325
1781
+ },
1782
+ {
1783
+ "epoch": 1.4076701396586122,
1784
+ "grad_norm": 0.160513773560524,
1785
+ "learning_rate": 0.00011477180202320377,
1786
+ "loss": 0.7451,
1787
+ "step": 6350
1788
+ },
1789
+ {
1790
+ "epoch": 1.413212148082465,
1791
+ "grad_norm": 0.09337064623832703,
1792
+ "learning_rate": 0.0001141797974511907,
1793
+ "loss": 0.6435,
1794
+ "step": 6375
1795
+ },
1796
+ {
1797
+ "epoch": 1.418754156506318,
1798
+ "grad_norm": 0.12163395434617996,
1799
+ "learning_rate": 0.00011358728528395733,
1800
+ "loss": 0.7313,
1801
+ "step": 6400
1802
+ },
1803
+ {
1804
+ "epoch": 1.4242961649301706,
1805
+ "grad_norm": 0.08646170049905777,
1806
+ "learning_rate": 0.00011299428673170389,
1807
+ "loss": 0.6327,
1808
+ "step": 6425
1809
+ },
1810
+ {
1811
+ "epoch": 1.4298381733540235,
1812
+ "grad_norm": 0.13511331379413605,
1813
+ "learning_rate": 0.00011240082302204194,
1814
+ "loss": 0.7324,
1815
+ "step": 6450
1816
+ },
1817
+ {
1818
+ "epoch": 1.4353801817778762,
1819
+ "grad_norm": 0.08811099082231522,
1820
+ "learning_rate": 0.00011180691539923407,
1821
+ "loss": 0.6322,
1822
+ "step": 6475
1823
+ },
1824
+ {
1825
+ "epoch": 1.440922190201729,
1826
+ "grad_norm": 0.12366902828216553,
1827
+ "learning_rate": 0.00011121258512343391,
1828
+ "loss": 0.7303,
1829
+ "step": 6500
1830
+ },
1831
+ {
1832
+ "epoch": 1.446464198625582,
1833
+ "grad_norm": 0.09673753380775452,
1834
+ "learning_rate": 0.00011061785346992463,
1835
+ "loss": 0.6368,
1836
+ "step": 6525
1837
+ },
1838
+ {
1839
+ "epoch": 1.4520062070494348,
1840
+ "grad_norm": 0.12419258803129196,
1841
+ "learning_rate": 0.00011002274172835771,
1842
+ "loss": 0.7431,
1843
+ "step": 6550
1844
+ },
1845
+ {
1846
+ "epoch": 1.4575482154732875,
1847
+ "grad_norm": 0.09401362389326096,
1848
+ "learning_rate": 0.00010942727120199052,
1849
+ "loss": 0.6079,
1850
+ "step": 6575
1851
+ },
1852
+ {
1853
+ "epoch": 1.4630902238971404,
1854
+ "grad_norm": 0.10955937206745148,
1855
+ "learning_rate": 0.000108831463206924,
1856
+ "loss": 0.7296,
1857
+ "step": 6600
1858
+ },
1859
+ {
1860
+ "epoch": 1.468632232320993,
1861
+ "grad_norm": 0.09884931892156601,
1862
+ "learning_rate": 0.00010823533907133943,
1863
+ "loss": 0.6373,
1864
+ "step": 6625
1865
+ },
1866
+ {
1867
+ "epoch": 1.474174240744846,
1868
+ "grad_norm": 0.14859217405319214,
1869
+ "learning_rate": 0.00010763892013473495,
1870
+ "loss": 0.7199,
1871
+ "step": 6650
1872
+ },
1873
+ {
1874
+ "epoch": 1.4797162491686988,
1875
+ "grad_norm": 0.0963672623038292,
1876
+ "learning_rate": 0.00010704222774716177,
1877
+ "loss": 0.6156,
1878
+ "step": 6675
1879
+ },
1880
+ {
1881
+ "epoch": 1.4852582575925515,
1882
+ "grad_norm": 0.14681296050548553,
1883
+ "learning_rate": 0.00010644528326845988,
1884
+ "loss": 0.74,
1885
+ "step": 6700
1886
+ },
1887
+ {
1888
+ "epoch": 1.4908002660164044,
1889
+ "grad_norm": 0.09270428866147995,
1890
+ "learning_rate": 0.00010584810806749327,
1891
+ "loss": 0.6185,
1892
+ "step": 6725
1893
+ },
1894
+ {
1895
+ "epoch": 1.496342274440257,
1896
+ "grad_norm": 0.11598405987024307,
1897
+ "learning_rate": 0.00010525072352138526,
1898
+ "loss": 0.7463,
1899
+ "step": 6750
1900
+ },
1901
+ {
1902
+ "epoch": 1.50188428286411,
1903
+ "grad_norm": 0.09355468302965164,
1904
+ "learning_rate": 0.00010465315101475295,
1905
+ "loss": 0.5996,
1906
+ "step": 6775
1907
+ },
1908
+ {
1909
+ "epoch": 1.5074262912879628,
1910
+ "grad_norm": 0.15108546614646912,
1911
+ "learning_rate": 0.00010405541193894204,
1912
+ "loss": 0.7512,
1913
+ "step": 6800
1914
+ },
1915
+ {
1916
+ "epoch": 1.5129682997118157,
1917
+ "grad_norm": 0.08825406432151794,
1918
+ "learning_rate": 0.00010345752769126079,
1919
+ "loss": 0.6367,
1920
+ "step": 6825
1921
+ },
1922
+ {
1923
+ "epoch": 1.5185103081356683,
1924
+ "grad_norm": 0.11757966130971909,
1925
+ "learning_rate": 0.0001028595196742143,
1926
+ "loss": 0.7556,
1927
+ "step": 6850
1928
+ },
1929
+ {
1930
+ "epoch": 1.5240523165595212,
1931
+ "grad_norm": 0.08973203599452972,
1932
+ "learning_rate": 0.00010226140929473813,
1933
+ "loss": 0.6038,
1934
+ "step": 6875
1935
+ },
1936
+ {
1937
+ "epoch": 1.5295943249833739,
1938
+ "grad_norm": 0.1337171196937561,
1939
+ "learning_rate": 0.00010166321796343223,
1940
+ "loss": 0.7388,
1941
+ "step": 6900
1942
+ },
1943
+ {
1944
+ "epoch": 1.5351363334072268,
1945
+ "grad_norm": 0.08748678117990494,
1946
+ "learning_rate": 0.0001010649670937943,
1947
+ "loss": 0.6093,
1948
+ "step": 6925
1949
+ },
1950
+ {
1951
+ "epoch": 1.5406783418310797,
1952
+ "grad_norm": 0.1228007897734642,
1953
+ "learning_rate": 0.00010046667810145338,
1954
+ "loss": 0.7494,
1955
+ "step": 6950
1956
+ },
1957
+ {
1958
+ "epoch": 1.5462203502549325,
1959
+ "grad_norm": 0.09511099755764008,
1960
+ "learning_rate": 9.986837240340319e-05,
1961
+ "loss": 0.5998,
1962
+ "step": 6975
1963
+ },
1964
+ {
1965
+ "epoch": 1.5517623586787852,
1966
+ "grad_norm": 0.12583385407924652,
1967
+ "learning_rate": 9.927007141723548e-05,
1968
+ "loss": 0.7266,
1969
+ "step": 7000
1970
+ },
1971
+ {
1972
+ "epoch": 1.5573043671026379,
1973
+ "grad_norm": 0.08915423601865768,
1974
+ "learning_rate": 9.867179656037326e-05,
1975
+ "loss": 0.638,
1976
+ "step": 7025
1977
+ },
1978
+ {
1979
+ "epoch": 1.5628463755264907,
1980
+ "grad_norm": 0.1239473968744278,
1981
+ "learning_rate": 9.80735692493043e-05,
1982
+ "loss": 0.7473,
1983
+ "step": 7050
1984
+ },
1985
+ {
1986
+ "epoch": 1.5683883839503436,
1987
+ "grad_norm": 0.09568199515342712,
1988
+ "learning_rate": 9.747541089881424e-05,
1989
+ "loss": 0.6174,
1990
+ "step": 7075
1991
+ },
1992
+ {
1993
+ "epoch": 1.5739303923741965,
1994
+ "grad_norm": 0.13295117020606995,
1995
+ "learning_rate": 9.687734292122024e-05,
1996
+ "loss": 0.7278,
1997
+ "step": 7100
1998
+ },
1999
+ {
2000
+ "epoch": 1.5794724007980492,
2001
+ "grad_norm": 0.09703335911035538,
2002
+ "learning_rate": 9.627938672560432e-05,
2003
+ "loss": 0.6051,
2004
+ "step": 7125
2005
+ },
2006
+ {
2007
+ "epoch": 1.585014409221902,
2008
+ "grad_norm": 0.12026989459991455,
2009
+ "learning_rate": 9.568156371704705e-05,
2010
+ "loss": 0.7341,
2011
+ "step": 7150
2012
+ },
2013
+ {
2014
+ "epoch": 1.5905564176457547,
2015
+ "grad_norm": 0.09185943752527237,
2016
+ "learning_rate": 9.508389529586128e-05,
2017
+ "loss": 0.607,
2018
+ "step": 7175
2019
+ },
2020
+ {
2021
+ "epoch": 1.5960984260696076,
2022
+ "grad_norm": 0.16115328669548035,
2023
+ "learning_rate": 9.448640285682613e-05,
2024
+ "loss": 0.7321,
2025
+ "step": 7200
2026
+ },
2027
+ {
2028
+ "epoch": 1.6016404344934605,
2029
+ "grad_norm": 0.09262697398662567,
2030
+ "learning_rate": 9.388910778842103e-05,
2031
+ "loss": 0.6027,
2032
+ "step": 7225
2033
+ },
2034
+ {
2035
+ "epoch": 1.6071824429173134,
2036
+ "grad_norm": 0.17859314382076263,
2037
+ "learning_rate": 9.329203147206007e-05,
2038
+ "loss": 0.7498,
2039
+ "step": 7250
2040
+ },
2041
+ {
2042
+ "epoch": 1.612724451341166,
2043
+ "grad_norm": 0.09392908960580826,
2044
+ "learning_rate": 9.269519528132677e-05,
2045
+ "loss": 0.6035,
2046
+ "step": 7275
2047
+ },
2048
+ {
2049
+ "epoch": 1.6182664597650187,
2050
+ "grad_norm": 0.1226978749036789,
2051
+ "learning_rate": 9.209862058120879e-05,
2052
+ "loss": 0.753,
2053
+ "step": 7300
2054
+ },
2055
+ {
2056
+ "epoch": 1.6238084681888716,
2057
+ "grad_norm": 0.09812294691801071,
2058
+ "learning_rate": 9.15023287273332e-05,
2059
+ "loss": 0.631,
2060
+ "step": 7325
2061
+ },
2062
+ {
2063
+ "epoch": 1.6293504766127245,
2064
+ "grad_norm": 0.14241814613342285,
2065
+ "learning_rate": 9.0906341065202e-05,
2066
+ "loss": 0.7291,
2067
+ "step": 7350
2068
+ },
2069
+ {
2070
+ "epoch": 1.6348924850365774,
2071
+ "grad_norm": 0.09261428564786911,
2072
+ "learning_rate": 9.031067892942805e-05,
2073
+ "loss": 0.6204,
2074
+ "step": 7375
2075
+ },
2076
+ {
2077
+ "epoch": 1.64043449346043,
2078
+ "grad_norm": 0.1319173127412796,
2079
+ "learning_rate": 8.971536364297126e-05,
2080
+ "loss": 0.7422,
2081
+ "step": 7400
2082
+ },
2083
+ {
2084
+ "epoch": 1.645976501884283,
2085
+ "grad_norm": 0.09849465638399124,
2086
+ "learning_rate": 8.912041651637541e-05,
2087
+ "loss": 0.6212,
2088
+ "step": 7425
2089
+ },
2090
+ {
2091
+ "epoch": 1.6515185103081356,
2092
+ "grad_norm": 0.1365344524383545,
2093
+ "learning_rate": 8.852585884700519e-05,
2094
+ "loss": 0.735,
2095
+ "step": 7450
2096
+ },
2097
+ {
2098
+ "epoch": 1.6570605187319885,
2099
+ "grad_norm": 0.0948396623134613,
2100
+ "learning_rate": 8.79317119182839e-05,
2101
+ "loss": 0.6113,
2102
+ "step": 7475
2103
+ },
2104
+ {
2105
+ "epoch": 1.6626025271558413,
2106
+ "grad_norm": 0.14469577372074127,
2107
+ "learning_rate": 8.73379969989315e-05,
2108
+ "loss": 0.7577,
2109
+ "step": 7500
2110
+ },
2111
+ {
2112
+ "epoch": 1.6681445355796942,
2113
+ "grad_norm": 0.08975006639957428,
2114
+ "learning_rate": 8.674473534220326e-05,
2115
+ "loss": 0.6123,
2116
+ "step": 7525
2117
+ },
2118
+ {
2119
+ "epoch": 1.6736865440035469,
2120
+ "grad_norm": 0.18550659716129303,
2121
+ "learning_rate": 8.615194818512905e-05,
2122
+ "loss": 0.7173,
2123
+ "step": 7550
2124
+ },
2125
+ {
2126
+ "epoch": 1.6792285524273995,
2127
+ "grad_norm": 0.09581893682479858,
2128
+ "learning_rate": 8.555965674775295e-05,
2129
+ "loss": 0.6052,
2130
+ "step": 7575
2131
+ },
2132
+ {
2133
+ "epoch": 1.6847705608512524,
2134
+ "grad_norm": 0.15174435079097748,
2135
+ "learning_rate": 8.496788223237381e-05,
2136
+ "loss": 0.741,
2137
+ "step": 7600
2138
+ },
2139
+ {
2140
+ "epoch": 1.6903125692751053,
2141
+ "grad_norm": 0.09482391923666,
2142
+ "learning_rate": 8.43766458227861e-05,
2143
+ "loss": 0.6281,
2144
+ "step": 7625
2145
+ },
2146
+ {
2147
+ "epoch": 1.6958545776989582,
2148
+ "grad_norm": 0.12190216034650803,
2149
+ "learning_rate": 8.37859686835218e-05,
2150
+ "loss": 0.7489,
2151
+ "step": 7650
2152
+ },
2153
+ {
2154
+ "epoch": 1.7013965861228109,
2155
+ "grad_norm": 0.09474999457597733,
2156
+ "learning_rate": 8.319587195909251e-05,
2157
+ "loss": 0.6238,
2158
+ "step": 7675
2159
+ },
2160
+ {
2161
+ "epoch": 1.7069385945466637,
2162
+ "grad_norm": 0.13137421011924744,
2163
+ "learning_rate": 8.260637677323279e-05,
2164
+ "loss": 0.7256,
2165
+ "step": 7700
2166
+ },
2167
+ {
2168
+ "epoch": 1.7124806029705164,
2169
+ "grad_norm": 0.09658095240592957,
2170
+ "learning_rate": 8.201750422814379e-05,
2171
+ "loss": 0.6038,
2172
+ "step": 7725
2173
+ },
2174
+ {
2175
+ "epoch": 1.7180226113943693,
2176
+ "grad_norm": 0.15730302035808563,
2177
+ "learning_rate": 8.142927540373805e-05,
2178
+ "loss": 0.7255,
2179
+ "step": 7750
2180
+ },
2181
+ {
2182
+ "epoch": 1.7235646198182222,
2183
+ "grad_norm": 0.09805120527744293,
2184
+ "learning_rate": 8.084171135688467e-05,
2185
+ "loss": 0.6136,
2186
+ "step": 7775
2187
+ },
2188
+ {
2189
+ "epoch": 1.729106628242075,
2190
+ "grad_norm": 0.14167381823062897,
2191
+ "learning_rate": 8.02548331206558e-05,
2192
+ "loss": 0.7246,
2193
+ "step": 7800
2194
+ },
2195
+ {
2196
+ "epoch": 1.7346486366659277,
2197
+ "grad_norm": 0.09563518315553665,
2198
+ "learning_rate": 7.966866170357346e-05,
2199
+ "loss": 0.6199,
2200
+ "step": 7825
2201
+ },
2202
+ {
2203
+ "epoch": 1.7401906450897804,
2204
+ "grad_norm": 0.11947919428348541,
2205
+ "learning_rate": 7.908321808885766e-05,
2206
+ "loss": 0.7284,
2207
+ "step": 7850
2208
+ },
2209
+ {
2210
+ "epoch": 1.7457326535136333,
2211
+ "grad_norm": 0.09702473878860474,
2212
+ "learning_rate": 7.849852323367521e-05,
2213
+ "loss": 0.6343,
2214
+ "step": 7875
2215
+ },
2216
+ {
2217
+ "epoch": 1.7512746619374862,
2218
+ "grad_norm": 0.16644233465194702,
2219
+ "learning_rate": 7.791459806838957e-05,
2220
+ "loss": 0.7295,
2221
+ "step": 7900
2222
+ },
2223
+ {
2224
+ "epoch": 1.756816670361339,
2225
+ "grad_norm": 0.09687721729278564,
2226
+ "learning_rate": 7.733146349581144e-05,
2227
+ "loss": 0.6232,
2228
+ "step": 7925
2229
+ },
2230
+ {
2231
+ "epoch": 1.7623586787851917,
2232
+ "grad_norm": 0.1432383805513382,
2233
+ "learning_rate": 7.674914039045076e-05,
2234
+ "loss": 0.7351,
2235
+ "step": 7950
2236
+ },
2237
+ {
2238
+ "epoch": 1.7679006872090446,
2239
+ "grad_norm": 0.09745761752128601,
2240
+ "learning_rate": 7.61676495977692e-05,
2241
+ "loss": 0.5918,
2242
+ "step": 7975
2243
+ },
2244
+ {
2245
+ "epoch": 1.7734426956328972,
2246
+ "grad_norm": 0.1507551074028015,
2247
+ "learning_rate": 7.558701193343419e-05,
2248
+ "loss": 0.7384,
2249
+ "step": 8000
2250
+ },
2251
+ {
2252
+ "epoch": 1.7789847040567501,
2253
+ "grad_norm": 0.09434136003255844,
2254
+ "learning_rate": 7.500724818257351e-05,
2255
+ "loss": 0.5987,
2256
+ "step": 8025
2257
+ },
2258
+ {
2259
+ "epoch": 1.784526712480603,
2260
+ "grad_norm": 0.1321529746055603,
2261
+ "learning_rate": 7.442837909903156e-05,
2262
+ "loss": 0.7409,
2263
+ "step": 8050
2264
+ },
2265
+ {
2266
+ "epoch": 1.790068720904456,
2267
+ "grad_norm": 0.09421923011541367,
2268
+ "learning_rate": 7.385042540462615e-05,
2269
+ "loss": 0.6042,
2270
+ "step": 8075
2271
+ },
2272
+ {
2273
+ "epoch": 1.7956107293283086,
2274
+ "grad_norm": 0.1154310330748558,
2275
+ "learning_rate": 7.32734077884069e-05,
2276
+ "loss": 0.7319,
2277
+ "step": 8100
2278
+ },
2279
+ {
2280
+ "epoch": 1.8011527377521612,
2281
+ "grad_norm": 0.09810709953308105,
2282
+ "learning_rate": 7.272037071314008e-05,
2283
+ "loss": 0.6196,
2284
+ "step": 8125
2285
+ },
2286
+ {
2287
+ "epoch": 1.806694746176014,
2288
+ "grad_norm": 0.12382727861404419,
2289
+ "learning_rate": 7.2145247695974e-05,
2290
+ "loss": 0.7366,
2291
+ "step": 8150
2292
+ },
2293
+ {
2294
+ "epoch": 1.812236754599867,
2295
+ "grad_norm": 0.10054343193769455,
2296
+ "learning_rate": 7.157112179736207e-05,
2297
+ "loss": 0.638,
2298
+ "step": 8175
2299
+ },
2300
+ {
2301
+ "epoch": 1.8177787630237199,
2302
+ "grad_norm": 0.12496750056743622,
2303
+ "learning_rate": 7.099801356933004e-05,
2304
+ "loss": 0.7435,
2305
+ "step": 8200
2306
+ },
2307
+ {
2308
+ "epoch": 1.8233207714475728,
2309
+ "grad_norm": 0.09787677973508835,
2310
+ "learning_rate": 7.0425943527474e-05,
2311
+ "loss": 0.5983,
2312
+ "step": 8225
2313
+ },
2314
+ {
2315
+ "epoch": 1.8288627798714254,
2316
+ "grad_norm": 0.16083738207817078,
2317
+ "learning_rate": 6.985493215022605e-05,
2318
+ "loss": 0.719,
2319
+ "step": 8250
2320
+ },
2321
+ {
2322
+ "epoch": 1.834404788295278,
2323
+ "grad_norm": 0.09314879029989243,
2324
+ "learning_rate": 6.928499987812112e-05,
2325
+ "loss": 0.6156,
2326
+ "step": 8275
2327
+ },
2328
+ {
2329
+ "epoch": 1.839946796719131,
2330
+ "grad_norm": 0.1240190863609314,
2331
+ "learning_rate": 6.871616711306545e-05,
2332
+ "loss": 0.7312,
2333
+ "step": 8300
2334
+ },
2335
+ {
2336
+ "epoch": 1.8454888051429839,
2337
+ "grad_norm": 0.10087582468986511,
2338
+ "learning_rate": 6.814845421760602e-05,
2339
+ "loss": 0.5953,
2340
+ "step": 8325
2341
+ },
2342
+ {
2343
+ "epoch": 1.8510308135668367,
2344
+ "grad_norm": 0.15391846001148224,
2345
+ "learning_rate": 6.758188151420189e-05,
2346
+ "loss": 0.7372,
2347
+ "step": 8350
2348
+ },
2349
+ {
2350
+ "epoch": 1.8565728219906894,
2351
+ "grad_norm": 0.09913575649261475,
2352
+ "learning_rate": 6.701646928449646e-05,
2353
+ "loss": 0.5859,
2354
+ "step": 8375
2355
+ },
2356
+ {
2357
+ "epoch": 1.862114830414542,
2358
+ "grad_norm": 0.1197889968752861,
2359
+ "learning_rate": 6.645223776859166e-05,
2360
+ "loss": 0.7416,
2361
+ "step": 8400
2362
+ },
2363
+ {
2364
+ "epoch": 1.867656838838395,
2365
+ "grad_norm": 0.09789609163999557,
2366
+ "learning_rate": 6.588920716432329e-05,
2367
+ "loss": 0.6422,
2368
+ "step": 8425
2369
+ },
2370
+ {
2371
+ "epoch": 1.8731988472622478,
2372
+ "grad_norm": 0.134750634431839,
2373
+ "learning_rate": 6.532739762653804e-05,
2374
+ "loss": 0.7462,
2375
+ "step": 8450
2376
+ },
2377
+ {
2378
+ "epoch": 1.8787408556861007,
2379
+ "grad_norm": 0.09457212686538696,
2380
+ "learning_rate": 6.476682926637197e-05,
2381
+ "loss": 0.6161,
2382
+ "step": 8475
2383
+ },
2384
+ {
2385
+ "epoch": 1.8842828641099536,
2386
+ "grad_norm": 0.18783515691757202,
2387
+ "learning_rate": 6.420752215053065e-05,
2388
+ "loss": 0.7236,
2389
+ "step": 8500
2390
+ },
2391
+ {
2392
+ "epoch": 1.8898248725338063,
2393
+ "grad_norm": 0.09566831588745117,
2394
+ "learning_rate": 6.364949630057078e-05,
2395
+ "loss": 0.5882,
2396
+ "step": 8525
2397
+ },
2398
+ {
2399
+ "epoch": 1.895366880957659,
2400
+ "grad_norm": 0.12137682735919952,
2401
+ "learning_rate": 6.30927716921835e-05,
2402
+ "loss": 0.7284,
2403
+ "step": 8550
2404
+ },
2405
+ {
2406
+ "epoch": 1.9009088893815118,
2407
+ "grad_norm": 0.09804505854845047,
2408
+ "learning_rate": 6.25373682544793e-05,
2409
+ "loss": 0.6165,
2410
+ "step": 8575
2411
+ },
2412
+ {
2413
+ "epoch": 1.9064508978053647,
2414
+ "grad_norm": 0.20505575835704803,
2415
+ "learning_rate": 6.198330586927463e-05,
2416
+ "loss": 0.7224,
2417
+ "step": 8600
2418
+ },
2419
+ {
2420
+ "epoch": 1.9119929062292176,
2421
+ "grad_norm": 0.10258720070123672,
2422
+ "learning_rate": 6.14306043703802e-05,
2423
+ "loss": 0.5683,
2424
+ "step": 8625
2425
+ },
2426
+ {
2427
+ "epoch": 1.9175349146530702,
2428
+ "grad_norm": 0.14654862880706787,
2429
+ "learning_rate": 6.087928354289103e-05,
2430
+ "loss": 0.7336,
2431
+ "step": 8650
2432
+ },
2433
+ {
2434
+ "epoch": 1.9230769230769231,
2435
+ "grad_norm": 0.09862499684095383,
2436
+ "learning_rate": 6.0329363122478e-05,
2437
+ "loss": 0.5824,
2438
+ "step": 8675
2439
+ },
2440
+ {
2441
+ "epoch": 1.9286189315007758,
2442
+ "grad_norm": 0.135267972946167,
2443
+ "learning_rate": 5.978086279468163e-05,
2444
+ "loss": 0.744,
2445
+ "step": 8700
2446
+ },
2447
+ {
2448
+ "epoch": 1.9341609399246287,
2449
+ "grad_norm": 0.1033608540892601,
2450
+ "learning_rate": 5.923380219420729e-05,
2451
+ "loss": 0.6134,
2452
+ "step": 8725
2453
+ },
2454
+ {
2455
+ "epoch": 1.9397029483484816,
2456
+ "grad_norm": 0.1315843164920807,
2457
+ "learning_rate": 5.8688200904222266e-05,
2458
+ "loss": 0.7151,
2459
+ "step": 8750
2460
+ },
2461
+ {
2462
+ "epoch": 1.9452449567723344,
2463
+ "grad_norm": 0.09951479732990265,
2464
+ "learning_rate": 5.8144078455654846e-05,
2465
+ "loss": 0.622,
2466
+ "step": 8775
2467
+ },
2468
+ {
2469
+ "epoch": 1.950786965196187,
2470
+ "grad_norm": 0.2037511169910431,
2471
+ "learning_rate": 5.760145432649515e-05,
2472
+ "loss": 0.7239,
2473
+ "step": 8800
2474
+ },
2475
+ {
2476
+ "epoch": 1.9563289736200398,
2477
+ "grad_norm": 0.10025127977132797,
2478
+ "learning_rate": 5.706034794109778e-05,
2479
+ "loss": 0.5922,
2480
+ "step": 8825
2481
+ },
2482
+ {
2483
+ "epoch": 1.9618709820438927,
2484
+ "grad_norm": 0.18914011120796204,
2485
+ "learning_rate": 5.65207786694866e-05,
2486
+ "loss": 0.7205,
2487
+ "step": 8850
2488
+ },
2489
+ {
2490
+ "epoch": 1.9674129904677455,
2491
+ "grad_norm": 0.10025196522474289,
2492
+ "learning_rate": 5.5982765826661256e-05,
2493
+ "loss": 0.5814,
2494
+ "step": 8875
2495
+ },
2496
+ {
2497
+ "epoch": 1.9729549988915984,
2498
+ "grad_norm": 0.12357232719659805,
2499
+ "learning_rate": 5.544632867190591e-05,
2500
+ "loss": 0.7217,
2501
+ "step": 8900
2502
+ },
2503
+ {
2504
+ "epoch": 1.978497007315451,
2505
+ "grad_norm": 0.09909965842962265,
2506
+ "learning_rate": 5.491148640809962e-05,
2507
+ "loss": 0.6102,
2508
+ "step": 8925
2509
+ },
2510
+ {
2511
+ "epoch": 1.984039015739304,
2512
+ "grad_norm": 0.16131410002708435,
2513
+ "learning_rate": 5.437825818102902e-05,
2514
+ "loss": 0.7193,
2515
+ "step": 8950
2516
+ },
2517
+ {
2518
+ "epoch": 1.9895810241631566,
2519
+ "grad_norm": 0.10195192694664001,
2520
+ "learning_rate": 5.384666307870293e-05,
2521
+ "loss": 0.5989,
2522
+ "step": 8975
2523
+ },
2524
+ {
2525
+ "epoch": 1.9951230325870095,
2526
+ "grad_norm": 0.1269746869802475,
2527
+ "learning_rate": 5.331672013066922e-05,
2528
+ "loss": 0.7287,
2529
+ "step": 9000
2530
+ },
2531
+ {
2532
+ "epoch": 2.0006650410108624,
2533
+ "grad_norm": 0.09312586486339569,
2534
+ "learning_rate": 5.278844830733332e-05,
2535
+ "loss": 0.6024,
2536
+ "step": 9025
2537
+ },
2538
+ {
2539
+ "epoch": 2.0062070494347153,
2540
+ "grad_norm": 0.10967884957790375,
2541
+ "learning_rate": 5.226186651927938e-05,
2542
+ "loss": 0.7053,
2543
+ "step": 9050
2544
+ },
2545
+ {
2546
+ "epoch": 2.0117490578585677,
2547
+ "grad_norm": 0.09102078527212143,
2548
+ "learning_rate": 5.1736993616593165e-05,
2549
+ "loss": 0.5861,
2550
+ "step": 9075
2551
+ },
2552
+ {
2553
+ "epoch": 2.0172910662824206,
2554
+ "grad_norm": 0.10821503400802612,
2555
+ "learning_rate": 5.121384838818746e-05,
2556
+ "loss": 0.6865,
2557
+ "step": 9100
2558
+ },
2559
+ {
2560
+ "epoch": 2.0228330747062735,
2561
+ "grad_norm": 0.09317923337221146,
2562
+ "learning_rate": 5.0692449561129285e-05,
2563
+ "loss": 0.5912,
2564
+ "step": 9125
2565
+ },
2566
+ {
2567
+ "epoch": 2.0283750831301264,
2568
+ "grad_norm": 0.11409013718366623,
2569
+ "learning_rate": 5.017281579996961e-05,
2570
+ "loss": 0.6979,
2571
+ "step": 9150
2572
+ },
2573
+ {
2574
+ "epoch": 2.0339170915539793,
2575
+ "grad_norm": 0.09311998635530472,
2576
+ "learning_rate": 4.965496570607523e-05,
2577
+ "loss": 0.6235,
2578
+ "step": 9175
2579
+ },
2580
+ {
2581
+ "epoch": 2.039459099977832,
2582
+ "grad_norm": 0.12173454463481903,
2583
+ "learning_rate": 4.913891781696285e-05,
2584
+ "loss": 0.7002,
2585
+ "step": 9200
2586
+ },
2587
+ {
2588
+ "epoch": 2.0450011084016846,
2589
+ "grad_norm": 0.10086531937122345,
2590
+ "learning_rate": 4.8624690605635626e-05,
2591
+ "loss": 0.5569,
2592
+ "step": 9225
2593
+ },
2594
+ {
2595
+ "epoch": 2.0505431168255375,
2596
+ "grad_norm": 0.10775309801101685,
2597
+ "learning_rate": 4.811230247992165e-05,
2598
+ "loss": 0.6925,
2599
+ "step": 9250
2600
+ },
2601
+ {
2602
+ "epoch": 2.0560851252493904,
2603
+ "grad_norm": 0.096441850066185,
2604
+ "learning_rate": 4.760177178181521e-05,
2605
+ "loss": 0.5906,
2606
+ "step": 9275
2607
+ },
2608
+ {
2609
+ "epoch": 2.0616271336732432,
2610
+ "grad_norm": 0.11258859932422638,
2611
+ "learning_rate": 4.709311678682005e-05,
2612
+ "loss": 0.6883,
2613
+ "step": 9300
2614
+ },
2615
+ {
2616
+ "epoch": 2.067169142097096,
2617
+ "grad_norm": 0.0907958596944809,
2618
+ "learning_rate": 4.658635570329537e-05,
2619
+ "loss": 0.6069,
2620
+ "step": 9325
2621
+ },
2622
+ {
2623
+ "epoch": 2.0727111505209486,
2624
+ "grad_norm": 0.11246030032634735,
2625
+ "learning_rate": 4.608150667180378e-05,
2626
+ "loss": 0.6951,
2627
+ "step": 9350
2628
+ },
2629
+ {
2630
+ "epoch": 2.0782531589448014,
2631
+ "grad_norm": 0.09199715405702591,
2632
+ "learning_rate": 4.557858776446203e-05,
2633
+ "loss": 0.5949,
2634
+ "step": 9375
2635
+ },
2636
+ {
2637
+ "epoch": 2.0837951673686543,
2638
+ "grad_norm": 0.11862944066524506,
2639
+ "learning_rate": 4.50776169842941e-05,
2640
+ "loss": 0.6701,
2641
+ "step": 9400
2642
+ },
2643
+ {
2644
+ "epoch": 2.089337175792507,
2645
+ "grad_norm": 0.09631045907735825,
2646
+ "learning_rate": 4.457861226458678e-05,
2647
+ "loss": 0.6158,
2648
+ "step": 9425
2649
+ },
2650
+ {
2651
+ "epoch": 2.09487918421636,
2652
+ "grad_norm": 0.107430100440979,
2653
+ "learning_rate": 4.408159146824756e-05,
2654
+ "loss": 0.6998,
2655
+ "step": 9450
2656
+ },
2657
+ {
2658
+ "epoch": 2.100421192640213,
2659
+ "grad_norm": 0.0933394506573677,
2660
+ "learning_rate": 4.358657238716533e-05,
2661
+ "loss": 0.5998,
2662
+ "step": 9475
2663
+ },
2664
+ {
2665
+ "epoch": 2.1059632010640654,
2666
+ "grad_norm": 0.11312496662139893,
2667
+ "learning_rate": 4.309357274157338e-05,
2668
+ "loss": 0.6909,
2669
+ "step": 9500
2670
+ },
2671
+ {
2672
+ "epoch": 2.1115052094879183,
2673
+ "grad_norm": 0.09570565819740295,
2674
+ "learning_rate": 4.260261017941526e-05,
2675
+ "loss": 0.5805,
2676
+ "step": 9525
2677
+ },
2678
+ {
2679
+ "epoch": 2.117047217911771,
2680
+ "grad_norm": 0.11327219754457474,
2681
+ "learning_rate": 4.211370227571276e-05,
2682
+ "loss": 0.6968,
2683
+ "step": 9550
2684
+ },
2685
+ {
2686
+ "epoch": 2.122589226335624,
2687
+ "grad_norm": 0.09823332726955414,
2688
+ "learning_rate": 4.162686653193698e-05,
2689
+ "loss": 0.5965,
2690
+ "step": 9575
2691
+ },
2692
+ {
2693
+ "epoch": 2.128131234759477,
2694
+ "grad_norm": 0.11287786811590195,
2695
+ "learning_rate": 4.11421203753817e-05,
2696
+ "loss": 0.6936,
2697
+ "step": 9600
2698
+ },
2699
+ {
2700
+ "epoch": 2.1336732431833294,
2701
+ "grad_norm": 0.09058432281017303,
2702
+ "learning_rate": 4.065948115853973e-05,
2703
+ "loss": 0.61,
2704
+ "step": 9625
2705
+ },
2706
+ {
2707
+ "epoch": 2.1392152516071823,
2708
+ "grad_norm": 0.11232877522706985,
2709
+ "learning_rate": 4.017896615848149e-05,
2710
+ "loss": 0.6852,
2711
+ "step": 9650
2712
+ },
2713
+ {
2714
+ "epoch": 2.144757260031035,
2715
+ "grad_norm": 0.09588344395160675,
2716
+ "learning_rate": 3.9700592576236686e-05,
2717
+ "loss": 0.6083,
2718
+ "step": 9675
2719
+ },
2720
+ {
2721
+ "epoch": 2.150299268454888,
2722
+ "grad_norm": 0.11739671975374222,
2723
+ "learning_rate": 3.922437753617856e-05,
2724
+ "loss": 0.6997,
2725
+ "step": 9700
2726
+ },
2727
+ {
2728
+ "epoch": 2.155841276878741,
2729
+ "grad_norm": 0.09523261338472366,
2730
+ "learning_rate": 3.875033808541083e-05,
2731
+ "loss": 0.5693,
2732
+ "step": 9725
2733
+ },
2734
+ {
2735
+ "epoch": 2.161383285302594,
2736
+ "grad_norm": 0.11812377721071243,
2737
+ "learning_rate": 3.827849119315755e-05,
2738
+ "loss": 0.6907,
2739
+ "step": 9750
2740
+ },
2741
+ {
2742
+ "epoch": 2.1669252937264463,
2743
+ "grad_norm": 0.09816546738147736,
2744
+ "learning_rate": 3.780885375015549e-05,
2745
+ "loss": 0.5891,
2746
+ "step": 9775
2747
+ },
2748
+ {
2749
+ "epoch": 2.172467302150299,
2750
+ "grad_norm": 0.11008067429065704,
2751
+ "learning_rate": 3.734144256804978e-05,
2752
+ "loss": 0.691,
2753
+ "step": 9800
2754
+ },
2755
+ {
2756
+ "epoch": 2.178009310574152,
2757
+ "grad_norm": 0.09762485325336456,
2758
+ "learning_rate": 3.687627437879177e-05,
2759
+ "loss": 0.5914,
2760
+ "step": 9825
2761
+ },
2762
+ {
2763
+ "epoch": 2.183551318998005,
2764
+ "grad_norm": 0.11886761337518692,
2765
+ "learning_rate": 3.6413365834040326e-05,
2766
+ "loss": 0.6896,
2767
+ "step": 9850
2768
+ },
2769
+ {
2770
+ "epoch": 2.189093327421858,
2771
+ "grad_norm": 0.09332608431577682,
2772
+ "learning_rate": 3.595273350456557e-05,
2773
+ "loss": 0.5844,
2774
+ "step": 9875
2775
+ },
2776
+ {
2777
+ "epoch": 2.1946353358457107,
2778
+ "grad_norm": 0.1171206533908844,
2779
+ "learning_rate": 3.549439387965592e-05,
2780
+ "loss": 0.7006,
2781
+ "step": 9900
2782
+ },
2783
+ {
2784
+ "epoch": 2.200177344269563,
2785
+ "grad_norm": 0.09177059680223465,
2786
+ "learning_rate": 3.503836336652756e-05,
2787
+ "loss": 0.589,
2788
+ "step": 9925
2789
+ },
2790
+ {
2791
+ "epoch": 2.205719352693416,
2792
+ "grad_norm": 0.11849093437194824,
2793
+ "learning_rate": 3.4584658289737296e-05,
2794
+ "loss": 0.687,
2795
+ "step": 9950
2796
+ },
2797
+ {
2798
+ "epoch": 2.211261361117269,
2799
+ "grad_norm": 0.09985481947660446,
2800
+ "learning_rate": 3.4133294890598065e-05,
2801
+ "loss": 0.6102,
2802
+ "step": 9975
2803
+ },
2804
+ {
2805
+ "epoch": 2.216803369541122,
2806
+ "grad_norm": 0.11642364412546158,
2807
+ "learning_rate": 3.3684289326597726e-05,
2808
+ "loss": 0.6963,
2809
+ "step": 10000
2810
+ }
2811
+ ],
2812
+ "logging_steps": 25,
2813
+ "max_steps": 13533,
2814
+ "num_input_tokens_seen": 0,
2815
+ "num_train_epochs": 3,
2816
+ "save_steps": 200,
2817
+ "stateful_callbacks": {
2818
+ "TrainerControl": {
2819
+ "args": {
2820
+ "should_epoch_stop": false,
2821
+ "should_evaluate": false,
2822
+ "should_log": false,
2823
+ "should_save": true,
2824
+ "should_training_stop": false
2825
+ },
2826
+ "attributes": {}
2827
+ }
2828
+ },
2829
+ "total_flos": 4.800783998233467e+18,
2830
+ "train_batch_size": 2,
2831
+ "trial_name": null,
2832
+ "trial_params": null
2833
+ }
checkpoint-10000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a13672a0401c8d0a53efc112f53ebd65f36fa9003e87b1710879c58d881a1e1
3
+ size 5051
checkpoint-13000/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: meta-llama/Llama-2-13b-chat-hf
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.12.0
checkpoint-13000/adapter_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf",
5
+ "bias": "lora_only",
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": 0.001,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 64,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "q_proj",
24
+ "v_proj"
25
+ ],
26
+ "task_type": "CAUSAL_LM",
27
+ "use_dora": false,
28
+ "use_rslora": false
29
+ }
checkpoint-13000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:803cd263d4f529894fe732ea0e358436e92ff26b5a5500d5944ab9c8959d23e7
3
+ size 209736952
checkpoint-13000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d9c518731f7b2b57e6c39a76bdbbe149cd125abccaedddfdc7256f1b2adf6a6
3
+ size 419529285
checkpoint-13000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:970432e3c65f4daee20d7d893efba2303b069229276460cf9b79a13b5fea3e68
3
+ size 14575
checkpoint-13000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73ed378f484d9b731afa6442e520363c7c3802bc4262465a019c1ad46861f790
3
+ size 627
checkpoint-13000/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-13000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-13000/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-13000/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<<SYS>>\\n' + system_message + '\\n<</SYS>>\\n\\n' + message['content'] %}{% else %}{% set content = message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ bos_token + '[INST] ' + content.strip() + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ ' ' + content.strip() + ' ' + eos_token }}{% endif %}{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": false,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
checkpoint-13000/trainer_state.json ADDED
@@ -0,0 +1,3673 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.881844380403458,
5
+ "eval_steps": 500,
6
+ "global_step": 13000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.005542008423852805,
13
+ "grad_norm": 0.07243233174085617,
14
+ "learning_rate": 1.2315270935960592e-05,
15
+ "loss": 1.4594,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.01108401684770561,
20
+ "grad_norm": 0.40484485030174255,
21
+ "learning_rate": 2.4630541871921184e-05,
22
+ "loss": 2.2032,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.016626025271558414,
27
+ "grad_norm": 0.06850667297840118,
28
+ "learning_rate": 3.694581280788178e-05,
29
+ "loss": 1.2931,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.02216803369541122,
34
+ "grad_norm": 0.4395073354244232,
35
+ "learning_rate": 4.926108374384237e-05,
36
+ "loss": 1.5698,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.027710042119264023,
41
+ "grad_norm": 0.077068030834198,
42
+ "learning_rate": 6.157635467980296e-05,
43
+ "loss": 1.0537,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 0.03325205054311683,
48
+ "grad_norm": 0.3282291293144226,
49
+ "learning_rate": 7.389162561576355e-05,
50
+ "loss": 0.9749,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 0.03879405896696963,
55
+ "grad_norm": 0.0593000203371048,
56
+ "learning_rate": 8.620689655172413e-05,
57
+ "loss": 0.9349,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 0.04433606739082244,
62
+ "grad_norm": 0.25612473487854004,
63
+ "learning_rate": 9.852216748768474e-05,
64
+ "loss": 0.8974,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 0.04987807581467524,
69
+ "grad_norm": 0.0757347121834755,
70
+ "learning_rate": 0.00011083743842364534,
71
+ "loss": 0.9081,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 0.055420084238528046,
76
+ "grad_norm": 0.14145499467849731,
77
+ "learning_rate": 0.00012315270935960593,
78
+ "loss": 0.8607,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 0.06096209266238085,
83
+ "grad_norm": 0.07710155844688416,
84
+ "learning_rate": 0.00013546798029556652,
85
+ "loss": 0.8973,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 0.06650410108623366,
90
+ "grad_norm": 0.14791467785835266,
91
+ "learning_rate": 0.0001477832512315271,
92
+ "loss": 0.7924,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 0.07204610951008646,
97
+ "grad_norm": 0.07742594182491302,
98
+ "learning_rate": 0.00016009852216748767,
99
+ "loss": 0.8698,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 0.07758811793393926,
104
+ "grad_norm": 0.14303487539291382,
105
+ "learning_rate": 0.00017241379310344826,
106
+ "loss": 0.786,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 0.08313012635779206,
111
+ "grad_norm": 0.0865108072757721,
112
+ "learning_rate": 0.00018472906403940888,
113
+ "loss": 0.8606,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 0.08867213478164487,
118
+ "grad_norm": 0.7533164024353027,
119
+ "learning_rate": 0.00019704433497536947,
120
+ "loss": 0.807,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 0.09421414320549767,
125
+ "grad_norm": 0.08325570821762085,
126
+ "learning_rate": 0.00019999896617927833,
127
+ "loss": 0.8635,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 0.09975615162935048,
132
+ "grad_norm": 0.1043543666601181,
133
+ "learning_rate": 0.0001999944557842899,
134
+ "loss": 0.7825,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 0.10529816005320328,
139
+ "grad_norm": 0.07949995994567871,
140
+ "learning_rate": 0.0001999863658806385,
141
+ "loss": 0.8379,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 0.11084016847705609,
146
+ "grad_norm": 0.12020070850849152,
147
+ "learning_rate": 0.00019997469675791905,
148
+ "loss": 0.768,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 0.11638217690090889,
153
+ "grad_norm": 0.0803595781326294,
154
+ "learning_rate": 0.00019995944883385196,
155
+ "loss": 0.8487,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 0.1219241853247617,
160
+ "grad_norm": 0.11509452760219574,
161
+ "learning_rate": 0.0001999406226542682,
162
+ "loss": 0.7787,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 0.1274661937486145,
167
+ "grad_norm": 0.07928384840488434,
168
+ "learning_rate": 0.00019991821889308987,
169
+ "loss": 0.8357,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 0.1330082021724673,
174
+ "grad_norm": 0.09423446655273438,
175
+ "learning_rate": 0.00019989223835230606,
176
+ "loss": 0.7564,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 0.1385502105963201,
181
+ "grad_norm": 0.0835939422249794,
182
+ "learning_rate": 0.000199862681961944,
183
+ "loss": 0.8568,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 0.1440922190201729,
188
+ "grad_norm": 0.09292898327112198,
189
+ "learning_rate": 0.0001998295507800359,
190
+ "loss": 0.7612,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 0.1496342274440257,
195
+ "grad_norm": 0.07704215496778488,
196
+ "learning_rate": 0.00019979284599258107,
197
+ "loss": 0.8263,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 0.15517623586787851,
202
+ "grad_norm": 0.10980474948883057,
203
+ "learning_rate": 0.0001997525689135034,
204
+ "loss": 0.7677,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 0.16071824429173132,
209
+ "grad_norm": 0.08016064018011093,
210
+ "learning_rate": 0.0001997087209846043,
211
+ "loss": 0.8344,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 0.16626025271558412,
216
+ "grad_norm": 0.0950881615281105,
217
+ "learning_rate": 0.0001996613037755113,
218
+ "loss": 0.769,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 0.17180226113943692,
223
+ "grad_norm": 0.07932984828948975,
224
+ "learning_rate": 0.00019961031898362152,
225
+ "loss": 0.8156,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 0.17734426956328975,
230
+ "grad_norm": 0.09336528927087784,
231
+ "learning_rate": 0.00019955576843404128,
232
+ "loss": 0.7767,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 0.18288627798714255,
237
+ "grad_norm": 0.08560346812009811,
238
+ "learning_rate": 0.00019949765407952042,
239
+ "loss": 0.8228,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 0.18842828641099535,
244
+ "grad_norm": 0.08475169539451599,
245
+ "learning_rate": 0.00019943597800038267,
246
+ "loss": 0.7669,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 0.19397029483484815,
251
+ "grad_norm": 0.09038034081459045,
252
+ "learning_rate": 0.00019937074240445105,
253
+ "loss": 0.8182,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 0.19951230325870095,
258
+ "grad_norm": 0.09195873886346817,
259
+ "learning_rate": 0.0001993019496269688,
260
+ "loss": 0.7598,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 0.20505431168255375,
265
+ "grad_norm": 0.08655796200037003,
266
+ "learning_rate": 0.0001992296021305159,
267
+ "loss": 0.8167,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 0.21059632010640655,
272
+ "grad_norm": 0.08353498578071594,
273
+ "learning_rate": 0.00019915370250492084,
274
+ "loss": 0.7486,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 0.21613832853025935,
279
+ "grad_norm": 0.09225723147392273,
280
+ "learning_rate": 0.0001990742534671679,
281
+ "loss": 0.8138,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 0.22168033695411218,
286
+ "grad_norm": 0.12104763090610504,
287
+ "learning_rate": 0.00019899125786129997,
288
+ "loss": 0.7153,
289
+ "step": 1000
290
+ },
291
+ {
292
+ "epoch": 0.22722234537796498,
293
+ "grad_norm": 0.0815986767411232,
294
+ "learning_rate": 0.00019890471865831669,
295
+ "loss": 0.7983,
296
+ "step": 1025
297
+ },
298
+ {
299
+ "epoch": 0.23276435380181779,
300
+ "grad_norm": 0.08845670521259308,
301
+ "learning_rate": 0.00019881463895606805,
302
+ "loss": 0.7187,
303
+ "step": 1050
304
+ },
305
+ {
306
+ "epoch": 0.2383063622256706,
307
+ "grad_norm": 0.0821809321641922,
308
+ "learning_rate": 0.00019872102197914359,
309
+ "loss": 0.804,
310
+ "step": 1075
311
+ },
312
+ {
313
+ "epoch": 0.2438483706495234,
314
+ "grad_norm": 0.08711609989404678,
315
+ "learning_rate": 0.00019862387107875688,
316
+ "loss": 0.7795,
317
+ "step": 1100
318
+ },
319
+ {
320
+ "epoch": 0.2493903790733762,
321
+ "grad_norm": 0.08517508953809738,
322
+ "learning_rate": 0.00019852318973262567,
323
+ "loss": 0.7937,
324
+ "step": 1125
325
+ },
326
+ {
327
+ "epoch": 0.254932387497229,
328
+ "grad_norm": 0.10830071568489075,
329
+ "learning_rate": 0.00019841898154484726,
330
+ "loss": 0.7458,
331
+ "step": 1150
332
+ },
333
+ {
334
+ "epoch": 0.2604743959210818,
335
+ "grad_norm": 0.08541836589574814,
336
+ "learning_rate": 0.0001983112502457696,
337
+ "loss": 0.8131,
338
+ "step": 1175
339
+ },
340
+ {
341
+ "epoch": 0.2660164043449346,
342
+ "grad_norm": 0.08794037252664566,
343
+ "learning_rate": 0.00019819999969185762,
344
+ "loss": 0.7577,
345
+ "step": 1200
346
+ },
347
+ {
348
+ "epoch": 0.2715584127687874,
349
+ "grad_norm": 0.08078176528215408,
350
+ "learning_rate": 0.00019808523386555542,
351
+ "loss": 0.812,
352
+ "step": 1225
353
+ },
354
+ {
355
+ "epoch": 0.2771004211926402,
356
+ "grad_norm": 0.09263130277395248,
357
+ "learning_rate": 0.0001979669568751434,
358
+ "loss": 0.7582,
359
+ "step": 1250
360
+ },
361
+ {
362
+ "epoch": 0.282642429616493,
363
+ "grad_norm": 0.08198932558298111,
364
+ "learning_rate": 0.00019784517295459147,
365
+ "loss": 0.7958,
366
+ "step": 1275
367
+ },
368
+ {
369
+ "epoch": 0.2881844380403458,
370
+ "grad_norm": 0.07858102023601532,
371
+ "learning_rate": 0.00019771988646340725,
372
+ "loss": 0.7744,
373
+ "step": 1300
374
+ },
375
+ {
376
+ "epoch": 0.2937264464641986,
377
+ "grad_norm": 0.0851408839225769,
378
+ "learning_rate": 0.00019759110188648026,
379
+ "loss": 0.7913,
380
+ "step": 1325
381
+ },
382
+ {
383
+ "epoch": 0.2992684548880514,
384
+ "grad_norm": 0.09252189099788666,
385
+ "learning_rate": 0.00019745882383392116,
386
+ "loss": 0.7675,
387
+ "step": 1350
388
+ },
389
+ {
390
+ "epoch": 0.30481046331190426,
391
+ "grad_norm": 0.08306555449962616,
392
+ "learning_rate": 0.0001973230570408968,
393
+ "loss": 0.8059,
394
+ "step": 1375
395
+ },
396
+ {
397
+ "epoch": 0.31035247173575703,
398
+ "grad_norm": 0.0797729641199112,
399
+ "learning_rate": 0.0001971838063674608,
400
+ "loss": 0.7424,
401
+ "step": 1400
402
+ },
403
+ {
404
+ "epoch": 0.31589448015960986,
405
+ "grad_norm": 0.08266165107488632,
406
+ "learning_rate": 0.0001970410767983794,
407
+ "loss": 0.7847,
408
+ "step": 1425
409
+ },
410
+ {
411
+ "epoch": 0.32143648858346263,
412
+ "grad_norm": 0.09364205598831177,
413
+ "learning_rate": 0.00019689487344295322,
414
+ "loss": 0.6924,
415
+ "step": 1450
416
+ },
417
+ {
418
+ "epoch": 0.32697849700731546,
419
+ "grad_norm": 0.08461842685937881,
420
+ "learning_rate": 0.00019674520153483414,
421
+ "loss": 0.8007,
422
+ "step": 1475
423
+ },
424
+ {
425
+ "epoch": 0.33252050543116823,
426
+ "grad_norm": 0.0840207040309906,
427
+ "learning_rate": 0.00019659206643183813,
428
+ "loss": 0.7139,
429
+ "step": 1500
430
+ },
431
+ {
432
+ "epoch": 0.33806251385502106,
433
+ "grad_norm": 0.08344192802906036,
434
+ "learning_rate": 0.00019643547361575343,
435
+ "loss": 0.7982,
436
+ "step": 1525
437
+ },
438
+ {
439
+ "epoch": 0.34360452227887384,
440
+ "grad_norm": 0.07934779673814774,
441
+ "learning_rate": 0.0001962754286921442,
442
+ "loss": 0.7164,
443
+ "step": 1550
444
+ },
445
+ {
446
+ "epoch": 0.34914653070272667,
447
+ "grad_norm": 0.08716201782226562,
448
+ "learning_rate": 0.00019611193739015,
449
+ "loss": 0.7846,
450
+ "step": 1575
451
+ },
452
+ {
453
+ "epoch": 0.3546885391265795,
454
+ "grad_norm": 0.08384064584970474,
455
+ "learning_rate": 0.0001959450055622806,
456
+ "loss": 0.7416,
457
+ "step": 1600
458
+ },
459
+ {
460
+ "epoch": 0.36023054755043227,
461
+ "grad_norm": 0.08661937713623047,
462
+ "learning_rate": 0.0001957746391842066,
463
+ "loss": 0.8075,
464
+ "step": 1625
465
+ },
466
+ {
467
+ "epoch": 0.3657725559742851,
468
+ "grad_norm": 0.09327207505702972,
469
+ "learning_rate": 0.00019560084435454536,
470
+ "loss": 0.7596,
471
+ "step": 1650
472
+ },
473
+ {
474
+ "epoch": 0.37131456439813787,
475
+ "grad_norm": 0.08391096442937851,
476
+ "learning_rate": 0.00019542362729464273,
477
+ "loss": 0.7794,
478
+ "step": 1675
479
+ },
480
+ {
481
+ "epoch": 0.3768565728219907,
482
+ "grad_norm": 0.07694080471992493,
483
+ "learning_rate": 0.00019524299434835052,
484
+ "loss": 0.7424,
485
+ "step": 1700
486
+ },
487
+ {
488
+ "epoch": 0.38239858124584347,
489
+ "grad_norm": 0.08567491173744202,
490
+ "learning_rate": 0.00019505895198179912,
491
+ "loss": 0.7996,
492
+ "step": 1725
493
+ },
494
+ {
495
+ "epoch": 0.3879405896696963,
496
+ "grad_norm": 0.08828684687614441,
497
+ "learning_rate": 0.0001948715067831663,
498
+ "loss": 0.7394,
499
+ "step": 1750
500
+ },
501
+ {
502
+ "epoch": 0.39348259809354913,
503
+ "grad_norm": 0.08347714692354202,
504
+ "learning_rate": 0.00019468066546244117,
505
+ "loss": 0.7734,
506
+ "step": 1775
507
+ },
508
+ {
509
+ "epoch": 0.3990246065174019,
510
+ "grad_norm": 0.07736373692750931,
511
+ "learning_rate": 0.00019448643485118412,
512
+ "loss": 0.7134,
513
+ "step": 1800
514
+ },
515
+ {
516
+ "epoch": 0.40456661494125473,
517
+ "grad_norm": 0.0840897262096405,
518
+ "learning_rate": 0.00019428882190228216,
519
+ "loss": 0.787,
520
+ "step": 1825
521
+ },
522
+ {
523
+ "epoch": 0.4101086233651075,
524
+ "grad_norm": 0.08665871620178223,
525
+ "learning_rate": 0.0001940878336897001,
526
+ "loss": 0.7151,
527
+ "step": 1850
528
+ },
529
+ {
530
+ "epoch": 0.41565063178896033,
531
+ "grad_norm": 0.08358912914991379,
532
+ "learning_rate": 0.0001938834774082274,
533
+ "loss": 0.7982,
534
+ "step": 1875
535
+ },
536
+ {
537
+ "epoch": 0.4211926402128131,
538
+ "grad_norm": 0.07928963005542755,
539
+ "learning_rate": 0.0001936757603732203,
540
+ "loss": 0.7195,
541
+ "step": 1900
542
+ },
543
+ {
544
+ "epoch": 0.42673464863666594,
545
+ "grad_norm": 0.08886470645666122,
546
+ "learning_rate": 0.00019346469002034042,
547
+ "loss": 0.7762,
548
+ "step": 1925
549
+ },
550
+ {
551
+ "epoch": 0.4322766570605187,
552
+ "grad_norm": 0.1071886494755745,
553
+ "learning_rate": 0.00019325027390528822,
554
+ "loss": 0.7453,
555
+ "step": 1950
556
+ },
557
+ {
558
+ "epoch": 0.43781866548437154,
559
+ "grad_norm": 0.08474262803792953,
560
+ "learning_rate": 0.00019303251970353261,
561
+ "loss": 0.7839,
562
+ "step": 1975
563
+ },
564
+ {
565
+ "epoch": 0.44336067390822437,
566
+ "grad_norm": 0.08803894370794296,
567
+ "learning_rate": 0.0001928114352100363,
568
+ "loss": 0.7171,
569
+ "step": 2000
570
+ },
571
+ {
572
+ "epoch": 0.44890268233207714,
573
+ "grad_norm": 0.08429575711488724,
574
+ "learning_rate": 0.00019258702833897665,
575
+ "loss": 0.7781,
576
+ "step": 2025
577
+ },
578
+ {
579
+ "epoch": 0.45444469075592997,
580
+ "grad_norm": 0.08510231226682663,
581
+ "learning_rate": 0.00019235930712346248,
582
+ "loss": 0.6949,
583
+ "step": 2050
584
+ },
585
+ {
586
+ "epoch": 0.45998669917978274,
587
+ "grad_norm": 0.08167176693677902,
588
+ "learning_rate": 0.00019212827971524634,
589
+ "loss": 0.7722,
590
+ "step": 2075
591
+ },
592
+ {
593
+ "epoch": 0.46552870760363557,
594
+ "grad_norm": 0.06542418897151947,
595
+ "learning_rate": 0.00019189395438443278,
596
+ "loss": 0.7203,
597
+ "step": 2100
598
+ },
599
+ {
600
+ "epoch": 0.47107071602748835,
601
+ "grad_norm": 0.08293402194976807,
602
+ "learning_rate": 0.00019165633951918247,
603
+ "loss": 0.7735,
604
+ "step": 2125
605
+ },
606
+ {
607
+ "epoch": 0.4766127244513412,
608
+ "grad_norm": 0.0809284895658493,
609
+ "learning_rate": 0.00019141544362541162,
610
+ "loss": 0.7412,
611
+ "step": 2150
612
+ },
613
+ {
614
+ "epoch": 0.48215473287519395,
615
+ "grad_norm": 0.08212891221046448,
616
+ "learning_rate": 0.00019117127532648773,
617
+ "loss": 0.7629,
618
+ "step": 2175
619
+ },
620
+ {
621
+ "epoch": 0.4876967412990468,
622
+ "grad_norm": 0.08602219074964523,
623
+ "learning_rate": 0.0001909238433629208,
624
+ "loss": 0.6935,
625
+ "step": 2200
626
+ },
627
+ {
628
+ "epoch": 0.4932387497228996,
629
+ "grad_norm": 0.08529417216777802,
630
+ "learning_rate": 0.0001906731565920505,
631
+ "loss": 0.7915,
632
+ "step": 2225
633
+ },
634
+ {
635
+ "epoch": 0.4987807581467524,
636
+ "grad_norm": 0.08774964511394501,
637
+ "learning_rate": 0.00019041922398772897,
638
+ "loss": 0.7359,
639
+ "step": 2250
640
+ },
641
+ {
642
+ "epoch": 0.5043227665706052,
643
+ "grad_norm": 0.08649475872516632,
644
+ "learning_rate": 0.00019016205463999984,
645
+ "loss": 0.7696,
646
+ "step": 2275
647
+ },
648
+ {
649
+ "epoch": 0.509864774994458,
650
+ "grad_norm": 0.0878506749868393,
651
+ "learning_rate": 0.00018990165775477252,
652
+ "loss": 0.7365,
653
+ "step": 2300
654
+ },
655
+ {
656
+ "epoch": 0.5154067834183108,
657
+ "grad_norm": 0.09131711721420288,
658
+ "learning_rate": 0.0001896380426534929,
659
+ "loss": 0.7809,
660
+ "step": 2325
661
+ },
662
+ {
663
+ "epoch": 0.5209487918421636,
664
+ "grad_norm": 0.07379825413227081,
665
+ "learning_rate": 0.00018937121877280957,
666
+ "loss": 0.7029,
667
+ "step": 2350
668
+ },
669
+ {
670
+ "epoch": 0.5264908002660164,
671
+ "grad_norm": 0.08535836637020111,
672
+ "learning_rate": 0.00018910119566423598,
673
+ "loss": 0.7679,
674
+ "step": 2375
675
+ },
676
+ {
677
+ "epoch": 0.5320328086898692,
678
+ "grad_norm": 0.06719771772623062,
679
+ "learning_rate": 0.00018882798299380864,
680
+ "loss": 0.7121,
681
+ "step": 2400
682
+ },
683
+ {
684
+ "epoch": 0.537574817113722,
685
+ "grad_norm": 0.09019796550273895,
686
+ "learning_rate": 0.00018855159054174093,
687
+ "loss": 0.7754,
688
+ "step": 2425
689
+ },
690
+ {
691
+ "epoch": 0.5431168255375748,
692
+ "grad_norm": 0.08144286274909973,
693
+ "learning_rate": 0.0001882720282020732,
694
+ "loss": 0.7255,
695
+ "step": 2450
696
+ },
697
+ {
698
+ "epoch": 0.5486588339614277,
699
+ "grad_norm": 0.08412271738052368,
700
+ "learning_rate": 0.0001879893059823185,
701
+ "loss": 0.7722,
702
+ "step": 2475
703
+ },
704
+ {
705
+ "epoch": 0.5542008423852804,
706
+ "grad_norm": 0.09016039222478867,
707
+ "learning_rate": 0.0001877034340031042,
708
+ "loss": 0.7275,
709
+ "step": 2500
710
+ },
711
+ {
712
+ "epoch": 0.5597428508091332,
713
+ "grad_norm": 0.08850298821926117,
714
+ "learning_rate": 0.00018741442249781,
715
+ "loss": 0.7828,
716
+ "step": 2525
717
+ },
718
+ {
719
+ "epoch": 0.565284859232986,
720
+ "grad_norm": 0.06989564746618271,
721
+ "learning_rate": 0.00018712228181220128,
722
+ "loss": 0.7111,
723
+ "step": 2550
724
+ },
725
+ {
726
+ "epoch": 0.5708268676568389,
727
+ "grad_norm": 0.09214618802070618,
728
+ "learning_rate": 0.00018682702240405906,
729
+ "loss": 0.7752,
730
+ "step": 2575
731
+ },
732
+ {
733
+ "epoch": 0.5763688760806917,
734
+ "grad_norm": 0.07766986638307571,
735
+ "learning_rate": 0.0001865286548428054,
736
+ "loss": 0.7108,
737
+ "step": 2600
738
+ },
739
+ {
740
+ "epoch": 0.5819108845045444,
741
+ "grad_norm": 0.07919591665267944,
742
+ "learning_rate": 0.00018622718980912514,
743
+ "loss": 0.775,
744
+ "step": 2625
745
+ },
746
+ {
747
+ "epoch": 0.5874528929283972,
748
+ "grad_norm": 0.07524783164262772,
749
+ "learning_rate": 0.00018592263809458361,
750
+ "loss": 0.6941,
751
+ "step": 2650
752
+ },
753
+ {
754
+ "epoch": 0.5929949013522501,
755
+ "grad_norm": 0.08549198508262634,
756
+ "learning_rate": 0.00018561501060124024,
757
+ "loss": 0.7718,
758
+ "step": 2675
759
+ },
760
+ {
761
+ "epoch": 0.5985369097761029,
762
+ "grad_norm": 0.08182788640260696,
763
+ "learning_rate": 0.0001853043183412584,
764
+ "loss": 0.7072,
765
+ "step": 2700
766
+ },
767
+ {
768
+ "epoch": 0.6040789181999556,
769
+ "grad_norm": 0.084741972386837,
770
+ "learning_rate": 0.00018499057243651096,
771
+ "loss": 0.7478,
772
+ "step": 2725
773
+ },
774
+ {
775
+ "epoch": 0.6096209266238085,
776
+ "grad_norm": 0.06824459880590439,
777
+ "learning_rate": 0.0001846737841181825,
778
+ "loss": 0.7238,
779
+ "step": 2750
780
+ },
781
+ {
782
+ "epoch": 0.6151629350476613,
783
+ "grad_norm": 0.08315033465623856,
784
+ "learning_rate": 0.00018435396472636704,
785
+ "loss": 0.7597,
786
+ "step": 2775
787
+ },
788
+ {
789
+ "epoch": 0.6207049434715141,
790
+ "grad_norm": 0.07116558402776718,
791
+ "learning_rate": 0.00018403112570966216,
792
+ "loss": 0.7096,
793
+ "step": 2800
794
+ },
795
+ {
796
+ "epoch": 0.6262469518953668,
797
+ "grad_norm": 0.08500215411186218,
798
+ "learning_rate": 0.00018370527862475916,
799
+ "loss": 0.756,
800
+ "step": 2825
801
+ },
802
+ {
803
+ "epoch": 0.6317889603192197,
804
+ "grad_norm": 0.07979004830121994,
805
+ "learning_rate": 0.00018337643513602933,
806
+ "loss": 0.6886,
807
+ "step": 2850
808
+ },
809
+ {
810
+ "epoch": 0.6373309687430725,
811
+ "grad_norm": 0.08140358328819275,
812
+ "learning_rate": 0.00018304460701510652,
813
+ "loss": 0.7648,
814
+ "step": 2875
815
+ },
816
+ {
817
+ "epoch": 0.6428729771669253,
818
+ "grad_norm": 0.07779423147439957,
819
+ "learning_rate": 0.0001827098061404656,
820
+ "loss": 0.7222,
821
+ "step": 2900
822
+ },
823
+ {
824
+ "epoch": 0.6484149855907781,
825
+ "grad_norm": 0.08853591978549957,
826
+ "learning_rate": 0.0001823720444969974,
827
+ "loss": 0.7736,
828
+ "step": 2925
829
+ },
830
+ {
831
+ "epoch": 0.6539569940146309,
832
+ "grad_norm": 0.07350102066993713,
833
+ "learning_rate": 0.0001820313341755795,
834
+ "loss": 0.7256,
835
+ "step": 2950
836
+ },
837
+ {
838
+ "epoch": 0.6594990024384837,
839
+ "grad_norm": 0.08152145147323608,
840
+ "learning_rate": 0.0001816876873726436,
841
+ "loss": 0.7598,
842
+ "step": 2975
843
+ },
844
+ {
845
+ "epoch": 0.6650410108623365,
846
+ "grad_norm": 0.08045897632837296,
847
+ "learning_rate": 0.00018134111638973876,
848
+ "loss": 0.7275,
849
+ "step": 3000
850
+ },
851
+ {
852
+ "epoch": 0.6705830192861894,
853
+ "grad_norm": 0.08514434099197388,
854
+ "learning_rate": 0.00018099163363309123,
855
+ "loss": 0.7688,
856
+ "step": 3025
857
+ },
858
+ {
859
+ "epoch": 0.6761250277100421,
860
+ "grad_norm": 0.060850344598293304,
861
+ "learning_rate": 0.00018063925161316012,
862
+ "loss": 0.7019,
863
+ "step": 3050
864
+ },
865
+ {
866
+ "epoch": 0.6816670361338949,
867
+ "grad_norm": 0.08471492677927017,
868
+ "learning_rate": 0.00018028398294418977,
869
+ "loss": 0.7573,
870
+ "step": 3075
871
+ },
872
+ {
873
+ "epoch": 0.6872090445577477,
874
+ "grad_norm": 0.0642291009426117,
875
+ "learning_rate": 0.00017992584034375798,
876
+ "loss": 0.7108,
877
+ "step": 3100
878
+ },
879
+ {
880
+ "epoch": 0.6927510529816006,
881
+ "grad_norm": 0.09357668459415436,
882
+ "learning_rate": 0.000179564836632321,
883
+ "loss": 0.7478,
884
+ "step": 3125
885
+ },
886
+ {
887
+ "epoch": 0.6982930614054533,
888
+ "grad_norm": 0.07198700308799744,
889
+ "learning_rate": 0.00017920098473275445,
890
+ "loss": 0.6973,
891
+ "step": 3150
892
+ },
893
+ {
894
+ "epoch": 0.7038350698293061,
895
+ "grad_norm": 0.08420095592737198,
896
+ "learning_rate": 0.00017883429766989064,
897
+ "loss": 0.7487,
898
+ "step": 3175
899
+ },
900
+ {
901
+ "epoch": 0.709377078253159,
902
+ "grad_norm": 0.06639819592237473,
903
+ "learning_rate": 0.00017846478857005255,
904
+ "loss": 0.6741,
905
+ "step": 3200
906
+ },
907
+ {
908
+ "epoch": 0.7149190866770118,
909
+ "grad_norm": 0.08200914412736893,
910
+ "learning_rate": 0.00017809247066058378,
911
+ "loss": 0.7526,
912
+ "step": 3225
913
+ },
914
+ {
915
+ "epoch": 0.7204610951008645,
916
+ "grad_norm": 0.07311141490936279,
917
+ "learning_rate": 0.0001777173572693751,
918
+ "loss": 0.677,
919
+ "step": 3250
920
+ },
921
+ {
922
+ "epoch": 0.7260031035247173,
923
+ "grad_norm": 0.08722089231014252,
924
+ "learning_rate": 0.00017733946182438726,
925
+ "loss": 0.7585,
926
+ "step": 3275
927
+ },
928
+ {
929
+ "epoch": 0.7315451119485702,
930
+ "grad_norm": 0.06589449942111969,
931
+ "learning_rate": 0.00017695879785317048,
932
+ "loss": 0.708,
933
+ "step": 3300
934
+ },
935
+ {
936
+ "epoch": 0.737087120372423,
937
+ "grad_norm": 0.08262074738740921,
938
+ "learning_rate": 0.0001765753789823801,
939
+ "loss": 0.749,
940
+ "step": 3325
941
+ },
942
+ {
943
+ "epoch": 0.7426291287962757,
944
+ "grad_norm": 0.07514823973178864,
945
+ "learning_rate": 0.00017618921893728867,
946
+ "loss": 0.6918,
947
+ "step": 3350
948
+ },
949
+ {
950
+ "epoch": 0.7481711372201286,
951
+ "grad_norm": 0.08757175505161285,
952
+ "learning_rate": 0.00017580033154129503,
953
+ "loss": 0.7445,
954
+ "step": 3375
955
+ },
956
+ {
957
+ "epoch": 0.7537131456439814,
958
+ "grad_norm": 0.0716458335518837,
959
+ "learning_rate": 0.0001754087307154289,
960
+ "loss": 0.7122,
961
+ "step": 3400
962
+ },
963
+ {
964
+ "epoch": 0.7592551540678342,
965
+ "grad_norm": 0.08453212678432465,
966
+ "learning_rate": 0.00017501443047785296,
967
+ "loss": 0.7656,
968
+ "step": 3425
969
+ },
970
+ {
971
+ "epoch": 0.7647971624916869,
972
+ "grad_norm": 0.06761575490236282,
973
+ "learning_rate": 0.00017461744494336098,
974
+ "loss": 0.6673,
975
+ "step": 3450
976
+ },
977
+ {
978
+ "epoch": 0.7703391709155398,
979
+ "grad_norm": 0.08577297627925873,
980
+ "learning_rate": 0.0001742177883228724,
981
+ "loss": 0.7494,
982
+ "step": 3475
983
+ },
984
+ {
985
+ "epoch": 0.7758811793393926,
986
+ "grad_norm": 0.05691730976104736,
987
+ "learning_rate": 0.00017381547492292376,
988
+ "loss": 0.6972,
989
+ "step": 3500
990
+ },
991
+ {
992
+ "epoch": 0.7814231877632454,
993
+ "grad_norm": 0.09115194529294968,
994
+ "learning_rate": 0.00017341051914515656,
995
+ "loss": 0.7706,
996
+ "step": 3525
997
+ },
998
+ {
999
+ "epoch": 0.7869651961870983,
1000
+ "grad_norm": 0.07214304804801941,
1001
+ "learning_rate": 0.00017300293548580162,
1002
+ "loss": 0.6807,
1003
+ "step": 3550
1004
+ },
1005
+ {
1006
+ "epoch": 0.792507204610951,
1007
+ "grad_norm": 0.08448139578104019,
1008
+ "learning_rate": 0.00017259273853516028,
1009
+ "loss": 0.7661,
1010
+ "step": 3575
1011
+ },
1012
+ {
1013
+ "epoch": 0.7980492130348038,
1014
+ "grad_norm": 0.08282499015331268,
1015
+ "learning_rate": 0.00017217994297708195,
1016
+ "loss": 0.7391,
1017
+ "step": 3600
1018
+ },
1019
+ {
1020
+ "epoch": 0.8035912214586566,
1021
+ "grad_norm": 0.0804004818201065,
1022
+ "learning_rate": 0.00017176456358843875,
1023
+ "loss": 0.7402,
1024
+ "step": 3625
1025
+ },
1026
+ {
1027
+ "epoch": 0.8091332298825095,
1028
+ "grad_norm": 0.07265755534172058,
1029
+ "learning_rate": 0.00017134661523859622,
1030
+ "loss": 0.7019,
1031
+ "step": 3650
1032
+ },
1033
+ {
1034
+ "epoch": 0.8146752383063622,
1035
+ "grad_norm": 0.08803457766771317,
1036
+ "learning_rate": 0.00017092611288888125,
1037
+ "loss": 0.7572,
1038
+ "step": 3675
1039
+ },
1040
+ {
1041
+ "epoch": 0.820217246730215,
1042
+ "grad_norm": 0.0652441680431366,
1043
+ "learning_rate": 0.0001705030715920464,
1044
+ "loss": 0.706,
1045
+ "step": 3700
1046
+ },
1047
+ {
1048
+ "epoch": 0.8257592551540678,
1049
+ "grad_norm": 0.08185753971338272,
1050
+ "learning_rate": 0.0001700775064917312,
1051
+ "loss": 0.764,
1052
+ "step": 3725
1053
+ },
1054
+ {
1055
+ "epoch": 0.8313012635779207,
1056
+ "grad_norm": 0.0859500914812088,
1057
+ "learning_rate": 0.00016964943282191984,
1058
+ "loss": 0.6927,
1059
+ "step": 3750
1060
+ },
1061
+ {
1062
+ "epoch": 0.8368432720017734,
1063
+ "grad_norm": 0.09176376461982727,
1064
+ "learning_rate": 0.00016921886590639602,
1065
+ "loss": 0.7567,
1066
+ "step": 3775
1067
+ },
1068
+ {
1069
+ "epoch": 0.8423852804256262,
1070
+ "grad_norm": 0.0646485984325409,
1071
+ "learning_rate": 0.0001687858211581943,
1072
+ "loss": 0.6848,
1073
+ "step": 3800
1074
+ },
1075
+ {
1076
+ "epoch": 0.8479272888494791,
1077
+ "grad_norm": 0.08545655012130737,
1078
+ "learning_rate": 0.00016835031407904839,
1079
+ "loss": 0.7546,
1080
+ "step": 3825
1081
+ },
1082
+ {
1083
+ "epoch": 0.8534692972733319,
1084
+ "grad_norm": 0.06338818371295929,
1085
+ "learning_rate": 0.00016791236025883626,
1086
+ "loss": 0.6655,
1087
+ "step": 3850
1088
+ },
1089
+ {
1090
+ "epoch": 0.8590113056971846,
1091
+ "grad_norm": 0.08781229704618454,
1092
+ "learning_rate": 0.00016747197537502205,
1093
+ "loss": 0.7441,
1094
+ "step": 3875
1095
+ },
1096
+ {
1097
+ "epoch": 0.8645533141210374,
1098
+ "grad_norm": 0.06220358610153198,
1099
+ "learning_rate": 0.00016702917519209487,
1100
+ "loss": 0.6795,
1101
+ "step": 3900
1102
+ },
1103
+ {
1104
+ "epoch": 0.8700953225448903,
1105
+ "grad_norm": 0.08917712420225143,
1106
+ "learning_rate": 0.0001665839755610044,
1107
+ "loss": 0.7552,
1108
+ "step": 3925
1109
+ },
1110
+ {
1111
+ "epoch": 0.8756373309687431,
1112
+ "grad_norm": 0.06624036282300949,
1113
+ "learning_rate": 0.00016613639241859355,
1114
+ "loss": 0.6632,
1115
+ "step": 3950
1116
+ },
1117
+ {
1118
+ "epoch": 0.8811793393925959,
1119
+ "grad_norm": 0.08898719400167465,
1120
+ "learning_rate": 0.00016568644178702803,
1121
+ "loss": 0.757,
1122
+ "step": 3975
1123
+ },
1124
+ {
1125
+ "epoch": 0.8867213478164487,
1126
+ "grad_norm": 0.05095354840159416,
1127
+ "learning_rate": 0.0001652341397732227,
1128
+ "loss": 0.6992,
1129
+ "step": 4000
1130
+ },
1131
+ {
1132
+ "epoch": 0.8922633562403015,
1133
+ "grad_norm": 0.08842916786670685,
1134
+ "learning_rate": 0.0001647795025682649,
1135
+ "loss": 0.7504,
1136
+ "step": 4025
1137
+ },
1138
+ {
1139
+ "epoch": 0.8978053646641543,
1140
+ "grad_norm": 0.0758206844329834,
1141
+ "learning_rate": 0.00016432254644683516,
1142
+ "loss": 0.7081,
1143
+ "step": 4050
1144
+ },
1145
+ {
1146
+ "epoch": 0.903347373088007,
1147
+ "grad_norm": 0.0940496176481247,
1148
+ "learning_rate": 0.0001638632877666243,
1149
+ "loss": 0.746,
1150
+ "step": 4075
1151
+ },
1152
+ {
1153
+ "epoch": 0.9088893815118599,
1154
+ "grad_norm": 0.06626766920089722,
1155
+ "learning_rate": 0.00016340174296774804,
1156
+ "loss": 0.6647,
1157
+ "step": 4100
1158
+ },
1159
+ {
1160
+ "epoch": 0.9144313899357127,
1161
+ "grad_norm": 0.08919317275285721,
1162
+ "learning_rate": 0.00016293792857215844,
1163
+ "loss": 0.7516,
1164
+ "step": 4125
1165
+ },
1166
+ {
1167
+ "epoch": 0.9199733983595655,
1168
+ "grad_norm": 0.06990760564804077,
1169
+ "learning_rate": 0.00016247186118305252,
1170
+ "loss": 0.7011,
1171
+ "step": 4150
1172
+ },
1173
+ {
1174
+ "epoch": 0.9255154067834183,
1175
+ "grad_norm": 0.0870794802904129,
1176
+ "learning_rate": 0.00016200355748427782,
1177
+ "loss": 0.7529,
1178
+ "step": 4175
1179
+ },
1180
+ {
1181
+ "epoch": 0.9310574152072711,
1182
+ "grad_norm": 0.06882854551076889,
1183
+ "learning_rate": 0.00016153303423973526,
1184
+ "loss": 0.7005,
1185
+ "step": 4200
1186
+ },
1187
+ {
1188
+ "epoch": 0.9365994236311239,
1189
+ "grad_norm": 0.084992416203022,
1190
+ "learning_rate": 0.0001610603082927789,
1191
+ "loss": 0.7519,
1192
+ "step": 4225
1193
+ },
1194
+ {
1195
+ "epoch": 0.9421414320549767,
1196
+ "grad_norm": 0.0638299211859703,
1197
+ "learning_rate": 0.00016058539656561323,
1198
+ "loss": 0.716,
1199
+ "step": 4250
1200
+ },
1201
+ {
1202
+ "epoch": 0.9476834404788296,
1203
+ "grad_norm": 0.08899606764316559,
1204
+ "learning_rate": 0.00016010831605868715,
1205
+ "loss": 0.7257,
1206
+ "step": 4275
1207
+ },
1208
+ {
1209
+ "epoch": 0.9532254489026823,
1210
+ "grad_norm": 0.06550378352403641,
1211
+ "learning_rate": 0.00015962908385008565,
1212
+ "loss": 0.7174,
1213
+ "step": 4300
1214
+ },
1215
+ {
1216
+ "epoch": 0.9587674573265351,
1217
+ "grad_norm": 0.09001540392637253,
1218
+ "learning_rate": 0.00015914771709491828,
1219
+ "loss": 0.7271,
1220
+ "step": 4325
1221
+ },
1222
+ {
1223
+ "epoch": 0.9643094657503879,
1224
+ "grad_norm": 0.06641615182161331,
1225
+ "learning_rate": 0.000158664233024705,
1226
+ "loss": 0.69,
1227
+ "step": 4350
1228
+ },
1229
+ {
1230
+ "epoch": 0.9698514741742408,
1231
+ "grad_norm": 0.08917039632797241,
1232
+ "learning_rate": 0.0001581786489467596,
1233
+ "loss": 0.7483,
1234
+ "step": 4375
1235
+ },
1236
+ {
1237
+ "epoch": 0.9753934825980936,
1238
+ "grad_norm": 0.05995697155594826,
1239
+ "learning_rate": 0.00015769098224356992,
1240
+ "loss": 0.7033,
1241
+ "step": 4400
1242
+ },
1243
+ {
1244
+ "epoch": 0.9809354910219463,
1245
+ "grad_norm": 0.08998765051364899,
1246
+ "learning_rate": 0.00015720125037217572,
1247
+ "loss": 0.7462,
1248
+ "step": 4425
1249
+ },
1250
+ {
1251
+ "epoch": 0.9864774994457992,
1252
+ "grad_norm": 0.05868702754378319,
1253
+ "learning_rate": 0.00015670947086354376,
1254
+ "loss": 0.6654,
1255
+ "step": 4450
1256
+ },
1257
+ {
1258
+ "epoch": 0.992019507869652,
1259
+ "grad_norm": 0.0880926102399826,
1260
+ "learning_rate": 0.00015621566132194005,
1261
+ "loss": 0.752,
1262
+ "step": 4475
1263
+ },
1264
+ {
1265
+ "epoch": 0.9975615162935048,
1266
+ "grad_norm": 0.08538970351219177,
1267
+ "learning_rate": 0.00015571983942430005,
1268
+ "loss": 0.7338,
1269
+ "step": 4500
1270
+ },
1271
+ {
1272
+ "epoch": 1.0031035247173576,
1273
+ "grad_norm": 0.0827050730586052,
1274
+ "learning_rate": 0.0001552220229195956,
1275
+ "loss": 0.7174,
1276
+ "step": 4525
1277
+ },
1278
+ {
1279
+ "epoch": 1.0086455331412103,
1280
+ "grad_norm": 0.10867294669151306,
1281
+ "learning_rate": 0.00015472222962819955,
1282
+ "loss": 0.7637,
1283
+ "step": 4550
1284
+ },
1285
+ {
1286
+ "epoch": 1.0141875415650632,
1287
+ "grad_norm": 0.08738269656896591,
1288
+ "learning_rate": 0.00015422047744124802,
1289
+ "loss": 0.6247,
1290
+ "step": 4575
1291
+ },
1292
+ {
1293
+ "epoch": 1.019729549988916,
1294
+ "grad_norm": 0.12865987420082092,
1295
+ "learning_rate": 0.0001537167843199998,
1296
+ "loss": 0.7424,
1297
+ "step": 4600
1298
+ },
1299
+ {
1300
+ "epoch": 1.0252715584127687,
1301
+ "grad_norm": 0.08619695156812668,
1302
+ "learning_rate": 0.00015321116829519345,
1303
+ "loss": 0.6461,
1304
+ "step": 4625
1305
+ },
1306
+ {
1307
+ "epoch": 1.0308135668366216,
1308
+ "grad_norm": 0.11726492643356323,
1309
+ "learning_rate": 0.0001527036474664019,
1310
+ "loss": 0.7433,
1311
+ "step": 4650
1312
+ },
1313
+ {
1314
+ "epoch": 1.0363555752604743,
1315
+ "grad_norm": 0.08198727667331696,
1316
+ "learning_rate": 0.0001521942400013844,
1317
+ "loss": 0.6086,
1318
+ "step": 4675
1319
+ },
1320
+ {
1321
+ "epoch": 1.0418975836843272,
1322
+ "grad_norm": 0.11951526254415512,
1323
+ "learning_rate": 0.00015168296413543635,
1324
+ "loss": 0.7521,
1325
+ "step": 4700
1326
+ },
1327
+ {
1328
+ "epoch": 1.04743959210818,
1329
+ "grad_norm": 0.08714735507965088,
1330
+ "learning_rate": 0.0001511698381707363,
1331
+ "loss": 0.631,
1332
+ "step": 4725
1333
+ },
1334
+ {
1335
+ "epoch": 1.0529816005320327,
1336
+ "grad_norm": 0.13869455456733704,
1337
+ "learning_rate": 0.00015065488047569107,
1338
+ "loss": 0.7524,
1339
+ "step": 4750
1340
+ },
1341
+ {
1342
+ "epoch": 1.0585236089558856,
1343
+ "grad_norm": 0.08524268865585327,
1344
+ "learning_rate": 0.00015013810948427794,
1345
+ "loss": 0.6617,
1346
+ "step": 4775
1347
+ },
1348
+ {
1349
+ "epoch": 1.0640656173797385,
1350
+ "grad_norm": 0.11017199605703354,
1351
+ "learning_rate": 0.00014961954369538494,
1352
+ "loss": 0.7598,
1353
+ "step": 4800
1354
+ },
1355
+ {
1356
+ "epoch": 1.0696076258035911,
1357
+ "grad_norm": 0.0834374874830246,
1358
+ "learning_rate": 0.00014909920167214858,
1359
+ "loss": 0.627,
1360
+ "step": 4825
1361
+ },
1362
+ {
1363
+ "epoch": 1.075149634227444,
1364
+ "grad_norm": 0.1357167363166809,
1365
+ "learning_rate": 0.0001485771020412894,
1366
+ "loss": 0.7466,
1367
+ "step": 4850
1368
+ },
1369
+ {
1370
+ "epoch": 1.080691642651297,
1371
+ "grad_norm": 0.08910629153251648,
1372
+ "learning_rate": 0.00014805326349244503,
1373
+ "loss": 0.6238,
1374
+ "step": 4875
1375
+ },
1376
+ {
1377
+ "epoch": 1.0862336510751496,
1378
+ "grad_norm": 0.10706546157598495,
1379
+ "learning_rate": 0.00014752770477750144,
1380
+ "loss": 0.7533,
1381
+ "step": 4900
1382
+ },
1383
+ {
1384
+ "epoch": 1.0917756594990025,
1385
+ "grad_norm": 0.09201759845018387,
1386
+ "learning_rate": 0.00014700044470992136,
1387
+ "loss": 0.6521,
1388
+ "step": 4925
1389
+ },
1390
+ {
1391
+ "epoch": 1.0973176679228553,
1392
+ "grad_norm": 0.14048361778259277,
1393
+ "learning_rate": 0.00014647150216407106,
1394
+ "loss": 0.7412,
1395
+ "step": 4950
1396
+ },
1397
+ {
1398
+ "epoch": 1.102859676346708,
1399
+ "grad_norm": 0.08308299630880356,
1400
+ "learning_rate": 0.00014594089607454454,
1401
+ "loss": 0.6333,
1402
+ "step": 4975
1403
+ },
1404
+ {
1405
+ "epoch": 1.108401684770561,
1406
+ "grad_norm": 0.12057497352361679,
1407
+ "learning_rate": 0.00014540864543548582,
1408
+ "loss": 0.7538,
1409
+ "step": 5000
1410
+ },
1411
+ {
1412
+ "epoch": 1.1139436931944136,
1413
+ "grad_norm": 0.089565709233284,
1414
+ "learning_rate": 0.00014487476929990898,
1415
+ "loss": 0.6662,
1416
+ "step": 5025
1417
+ },
1418
+ {
1419
+ "epoch": 1.1194857016182664,
1420
+ "grad_norm": 0.12125346809625626,
1421
+ "learning_rate": 0.00014433928677901612,
1422
+ "loss": 0.7653,
1423
+ "step": 5050
1424
+ },
1425
+ {
1426
+ "epoch": 1.1250277100421193,
1427
+ "grad_norm": 0.08421044796705246,
1428
+ "learning_rate": 0.00014380221704151318,
1429
+ "loss": 0.615,
1430
+ "step": 5075
1431
+ },
1432
+ {
1433
+ "epoch": 1.130569718465972,
1434
+ "grad_norm": 0.12215881794691086,
1435
+ "learning_rate": 0.0001432635793129239,
1436
+ "loss": 0.7482,
1437
+ "step": 5100
1438
+ },
1439
+ {
1440
+ "epoch": 1.1361117268898249,
1441
+ "grad_norm": 0.08646813780069351,
1442
+ "learning_rate": 0.0001427233928749014,
1443
+ "loss": 0.6292,
1444
+ "step": 5125
1445
+ },
1446
+ {
1447
+ "epoch": 1.1416537353136778,
1448
+ "grad_norm": 0.11372750997543335,
1449
+ "learning_rate": 0.00014218167706453816,
1450
+ "loss": 0.7487,
1451
+ "step": 5150
1452
+ },
1453
+ {
1454
+ "epoch": 1.1471957437375304,
1455
+ "grad_norm": 0.08925063908100128,
1456
+ "learning_rate": 0.00014163845127367362,
1457
+ "loss": 0.6336,
1458
+ "step": 5175
1459
+ },
1460
+ {
1461
+ "epoch": 1.1527377521613833,
1462
+ "grad_norm": 0.12316026538610458,
1463
+ "learning_rate": 0.00014109373494820018,
1464
+ "loss": 0.7566,
1465
+ "step": 5200
1466
+ },
1467
+ {
1468
+ "epoch": 1.1582797605852362,
1469
+ "grad_norm": 0.08784055709838867,
1470
+ "learning_rate": 0.00014054754758736698,
1471
+ "loss": 0.6124,
1472
+ "step": 5225
1473
+ },
1474
+ {
1475
+ "epoch": 1.1638217690090888,
1476
+ "grad_norm": 0.11267346143722534,
1477
+ "learning_rate": 0.0001399999087430819,
1478
+ "loss": 0.7611,
1479
+ "step": 5250
1480
+ },
1481
+ {
1482
+ "epoch": 1.1693637774329417,
1483
+ "grad_norm": 0.08636374026536942,
1484
+ "learning_rate": 0.00013945083801921167,
1485
+ "loss": 0.6561,
1486
+ "step": 5275
1487
+ },
1488
+ {
1489
+ "epoch": 1.1749057858567946,
1490
+ "grad_norm": 0.13902850449085236,
1491
+ "learning_rate": 0.0001389003550708802,
1492
+ "loss": 0.7546,
1493
+ "step": 5300
1494
+ },
1495
+ {
1496
+ "epoch": 1.1804477942806473,
1497
+ "grad_norm": 0.09029743075370789,
1498
+ "learning_rate": 0.0001383484796037648,
1499
+ "loss": 0.6226,
1500
+ "step": 5325
1501
+ },
1502
+ {
1503
+ "epoch": 1.1859898027045002,
1504
+ "grad_norm": 0.10377778112888336,
1505
+ "learning_rate": 0.00013779523137339095,
1506
+ "loss": 0.7422,
1507
+ "step": 5350
1508
+ },
1509
+ {
1510
+ "epoch": 1.1915318111283528,
1511
+ "grad_norm": 0.08857985585927963,
1512
+ "learning_rate": 0.00013724063018442494,
1513
+ "loss": 0.6346,
1514
+ "step": 5375
1515
+ },
1516
+ {
1517
+ "epoch": 1.1970738195522057,
1518
+ "grad_norm": 0.15107670426368713,
1519
+ "learning_rate": 0.0001366846958899651,
1520
+ "loss": 0.7266,
1521
+ "step": 5400
1522
+ },
1523
+ {
1524
+ "epoch": 1.2026158279760586,
1525
+ "grad_norm": 0.09053236246109009,
1526
+ "learning_rate": 0.000136127448390831,
1527
+ "loss": 0.617,
1528
+ "step": 5425
1529
+ },
1530
+ {
1531
+ "epoch": 1.2081578363999113,
1532
+ "grad_norm": 0.11061497032642365,
1533
+ "learning_rate": 0.00013556890763485112,
1534
+ "loss": 0.7631,
1535
+ "step": 5450
1536
+ },
1537
+ {
1538
+ "epoch": 1.2136998448237641,
1539
+ "grad_norm": 0.08849512785673141,
1540
+ "learning_rate": 0.0001350090936161487,
1541
+ "loss": 0.5992,
1542
+ "step": 5475
1543
+ },
1544
+ {
1545
+ "epoch": 1.219241853247617,
1546
+ "grad_norm": 0.12802088260650635,
1547
+ "learning_rate": 0.00013444802637442606,
1548
+ "loss": 0.7389,
1549
+ "step": 5500
1550
+ },
1551
+ {
1552
+ "epoch": 1.2247838616714697,
1553
+ "grad_norm": 0.08612997084856033,
1554
+ "learning_rate": 0.0001338857259942473,
1555
+ "loss": 0.6173,
1556
+ "step": 5525
1557
+ },
1558
+ {
1559
+ "epoch": 1.2303258700953226,
1560
+ "grad_norm": 0.1339733898639679,
1561
+ "learning_rate": 0.0001333222126043192,
1562
+ "loss": 0.7473,
1563
+ "step": 5550
1564
+ },
1565
+ {
1566
+ "epoch": 1.2358678785191755,
1567
+ "grad_norm": 0.08974476903676987,
1568
+ "learning_rate": 0.00013275750637677073,
1569
+ "loss": 0.6224,
1570
+ "step": 5575
1571
+ },
1572
+ {
1573
+ "epoch": 1.2414098869430281,
1574
+ "grad_norm": 0.1431790143251419,
1575
+ "learning_rate": 0.00013219162752643103,
1576
+ "loss": 0.748,
1577
+ "step": 5600
1578
+ },
1579
+ {
1580
+ "epoch": 1.246951895366881,
1581
+ "grad_norm": 0.08545912057161331,
1582
+ "learning_rate": 0.0001316245963101056,
1583
+ "loss": 0.5969,
1584
+ "step": 5625
1585
+ },
1586
+ {
1587
+ "epoch": 1.2524939037907337,
1588
+ "grad_norm": 0.13927248120307922,
1589
+ "learning_rate": 0.00013105643302585137,
1590
+ "loss": 0.752,
1591
+ "step": 5650
1592
+ },
1593
+ {
1594
+ "epoch": 1.2580359122145865,
1595
+ "grad_norm": 0.08908078819513321,
1596
+ "learning_rate": 0.0001304871580122499,
1597
+ "loss": 0.6422,
1598
+ "step": 5675
1599
+ },
1600
+ {
1601
+ "epoch": 1.2635779206384394,
1602
+ "grad_norm": 0.13006670773029327,
1603
+ "learning_rate": 0.00012991679164767942,
1604
+ "loss": 0.7594,
1605
+ "step": 5700
1606
+ },
1607
+ {
1608
+ "epoch": 1.269119929062292,
1609
+ "grad_norm": 0.08754415810108185,
1610
+ "learning_rate": 0.00012934535434958538,
1611
+ "loss": 0.6121,
1612
+ "step": 5725
1613
+ },
1614
+ {
1615
+ "epoch": 1.274661937486145,
1616
+ "grad_norm": 0.15051035583019257,
1617
+ "learning_rate": 0.0001287728665737495,
1618
+ "loss": 0.7414,
1619
+ "step": 5750
1620
+ },
1621
+ {
1622
+ "epoch": 1.2802039459099979,
1623
+ "grad_norm": 0.08734755963087082,
1624
+ "learning_rate": 0.00012819934881355745,
1625
+ "loss": 0.6368,
1626
+ "step": 5775
1627
+ },
1628
+ {
1629
+ "epoch": 1.2857459543338505,
1630
+ "grad_norm": 0.17122209072113037,
1631
+ "learning_rate": 0.0001276248215992654,
1632
+ "loss": 0.7535,
1633
+ "step": 5800
1634
+ },
1635
+ {
1636
+ "epoch": 1.2912879627577034,
1637
+ "grad_norm": 0.08958180993795395,
1638
+ "learning_rate": 0.00012704930549726503,
1639
+ "loss": 0.5941,
1640
+ "step": 5825
1641
+ },
1642
+ {
1643
+ "epoch": 1.2968299711815563,
1644
+ "grad_norm": 0.10746924579143524,
1645
+ "learning_rate": 0.0001264728211093473,
1646
+ "loss": 0.7498,
1647
+ "step": 5850
1648
+ },
1649
+ {
1650
+ "epoch": 1.302371979605409,
1651
+ "grad_norm": 0.08696026355028152,
1652
+ "learning_rate": 0.00012589538907196486,
1653
+ "loss": 0.6451,
1654
+ "step": 5875
1655
+ },
1656
+ {
1657
+ "epoch": 1.3079139880292618,
1658
+ "grad_norm": 0.14526303112506866,
1659
+ "learning_rate": 0.0001253170300554936,
1660
+ "loss": 0.7427,
1661
+ "step": 5900
1662
+ },
1663
+ {
1664
+ "epoch": 1.3134559964531145,
1665
+ "grad_norm": 0.08736992627382278,
1666
+ "learning_rate": 0.0001247377647634924,
1667
+ "loss": 0.632,
1668
+ "step": 5925
1669
+ },
1670
+ {
1671
+ "epoch": 1.3189980048769674,
1672
+ "grad_norm": 0.13362443447113037,
1673
+ "learning_rate": 0.00012415761393196227,
1674
+ "loss": 0.7394,
1675
+ "step": 5950
1676
+ },
1677
+ {
1678
+ "epoch": 1.3245400133008203,
1679
+ "grad_norm": 0.09453903138637543,
1680
+ "learning_rate": 0.00012357659832860386,
1681
+ "loss": 0.6394,
1682
+ "step": 5975
1683
+ },
1684
+ {
1685
+ "epoch": 1.3300820217246732,
1686
+ "grad_norm": 0.1354217529296875,
1687
+ "learning_rate": 0.00012299473875207416,
1688
+ "loss": 0.727,
1689
+ "step": 6000
1690
+ },
1691
+ {
1692
+ "epoch": 1.3356240301485258,
1693
+ "grad_norm": 0.09107760339975357,
1694
+ "learning_rate": 0.0001224120560312419,
1695
+ "loss": 0.6197,
1696
+ "step": 6025
1697
+ },
1698
+ {
1699
+ "epoch": 1.3411660385723787,
1700
+ "grad_norm": 0.12501803040504456,
1701
+ "learning_rate": 0.00012182857102444203,
1702
+ "loss": 0.7475,
1703
+ "step": 6050
1704
+ },
1705
+ {
1706
+ "epoch": 1.3467080469962314,
1707
+ "grad_norm": 0.08888363093137741,
1708
+ "learning_rate": 0.00012124430461872886,
1709
+ "loss": 0.6108,
1710
+ "step": 6075
1711
+ },
1712
+ {
1713
+ "epoch": 1.3522500554200843,
1714
+ "grad_norm": 0.16767624020576477,
1715
+ "learning_rate": 0.00012065927772912863,
1716
+ "loss": 0.7408,
1717
+ "step": 6100
1718
+ },
1719
+ {
1720
+ "epoch": 1.3577920638439371,
1721
+ "grad_norm": 0.09112541377544403,
1722
+ "learning_rate": 0.00012007351129789062,
1723
+ "loss": 0.5868,
1724
+ "step": 6125
1725
+ },
1726
+ {
1727
+ "epoch": 1.3633340722677898,
1728
+ "grad_norm": 0.13539327681064606,
1729
+ "learning_rate": 0.0001194870262937375,
1730
+ "loss": 0.7505,
1731
+ "step": 6150
1732
+ },
1733
+ {
1734
+ "epoch": 1.3688760806916427,
1735
+ "grad_norm": 0.08977732807397842,
1736
+ "learning_rate": 0.00011889984371111475,
1737
+ "loss": 0.5985,
1738
+ "step": 6175
1739
+ },
1740
+ {
1741
+ "epoch": 1.3744180891154953,
1742
+ "grad_norm": 0.17703984677791595,
1743
+ "learning_rate": 0.00011831198456943924,
1744
+ "loss": 0.7334,
1745
+ "step": 6200
1746
+ },
1747
+ {
1748
+ "epoch": 1.3799600975393482,
1749
+ "grad_norm": 0.09067991375923157,
1750
+ "learning_rate": 0.00011772346991234651,
1751
+ "loss": 0.5874,
1752
+ "step": 6225
1753
+ },
1754
+ {
1755
+ "epoch": 1.3855021059632011,
1756
+ "grad_norm": 0.10922655463218689,
1757
+ "learning_rate": 0.00011713432080693772,
1758
+ "loss": 0.746,
1759
+ "step": 6250
1760
+ },
1761
+ {
1762
+ "epoch": 1.391044114387054,
1763
+ "grad_norm": 0.08802726864814758,
1764
+ "learning_rate": 0.00011654455834302535,
1765
+ "loss": 0.6084,
1766
+ "step": 6275
1767
+ },
1768
+ {
1769
+ "epoch": 1.3965861228109067,
1770
+ "grad_norm": 0.11013362556695938,
1771
+ "learning_rate": 0.00011595420363237844,
1772
+ "loss": 0.7431,
1773
+ "step": 6300
1774
+ },
1775
+ {
1776
+ "epoch": 1.4021281312347595,
1777
+ "grad_norm": 0.09353320300579071,
1778
+ "learning_rate": 0.00011536327780796661,
1779
+ "loss": 0.6504,
1780
+ "step": 6325
1781
+ },
1782
+ {
1783
+ "epoch": 1.4076701396586122,
1784
+ "grad_norm": 0.160513773560524,
1785
+ "learning_rate": 0.00011477180202320377,
1786
+ "loss": 0.7451,
1787
+ "step": 6350
1788
+ },
1789
+ {
1790
+ "epoch": 1.413212148082465,
1791
+ "grad_norm": 0.09337064623832703,
1792
+ "learning_rate": 0.0001141797974511907,
1793
+ "loss": 0.6435,
1794
+ "step": 6375
1795
+ },
1796
+ {
1797
+ "epoch": 1.418754156506318,
1798
+ "grad_norm": 0.12163395434617996,
1799
+ "learning_rate": 0.00011358728528395733,
1800
+ "loss": 0.7313,
1801
+ "step": 6400
1802
+ },
1803
+ {
1804
+ "epoch": 1.4242961649301706,
1805
+ "grad_norm": 0.08646170049905777,
1806
+ "learning_rate": 0.00011299428673170389,
1807
+ "loss": 0.6327,
1808
+ "step": 6425
1809
+ },
1810
+ {
1811
+ "epoch": 1.4298381733540235,
1812
+ "grad_norm": 0.13511331379413605,
1813
+ "learning_rate": 0.00011240082302204194,
1814
+ "loss": 0.7324,
1815
+ "step": 6450
1816
+ },
1817
+ {
1818
+ "epoch": 1.4353801817778762,
1819
+ "grad_norm": 0.08811099082231522,
1820
+ "learning_rate": 0.00011180691539923407,
1821
+ "loss": 0.6322,
1822
+ "step": 6475
1823
+ },
1824
+ {
1825
+ "epoch": 1.440922190201729,
1826
+ "grad_norm": 0.12366902828216553,
1827
+ "learning_rate": 0.00011121258512343391,
1828
+ "loss": 0.7303,
1829
+ "step": 6500
1830
+ },
1831
+ {
1832
+ "epoch": 1.446464198625582,
1833
+ "grad_norm": 0.09673753380775452,
1834
+ "learning_rate": 0.00011061785346992463,
1835
+ "loss": 0.6368,
1836
+ "step": 6525
1837
+ },
1838
+ {
1839
+ "epoch": 1.4520062070494348,
1840
+ "grad_norm": 0.12419258803129196,
1841
+ "learning_rate": 0.00011002274172835771,
1842
+ "loss": 0.7431,
1843
+ "step": 6550
1844
+ },
1845
+ {
1846
+ "epoch": 1.4575482154732875,
1847
+ "grad_norm": 0.09401362389326096,
1848
+ "learning_rate": 0.00010942727120199052,
1849
+ "loss": 0.6079,
1850
+ "step": 6575
1851
+ },
1852
+ {
1853
+ "epoch": 1.4630902238971404,
1854
+ "grad_norm": 0.10955937206745148,
1855
+ "learning_rate": 0.000108831463206924,
1856
+ "loss": 0.7296,
1857
+ "step": 6600
1858
+ },
1859
+ {
1860
+ "epoch": 1.468632232320993,
1861
+ "grad_norm": 0.09884931892156601,
1862
+ "learning_rate": 0.00010823533907133943,
1863
+ "loss": 0.6373,
1864
+ "step": 6625
1865
+ },
1866
+ {
1867
+ "epoch": 1.474174240744846,
1868
+ "grad_norm": 0.14859217405319214,
1869
+ "learning_rate": 0.00010763892013473495,
1870
+ "loss": 0.7199,
1871
+ "step": 6650
1872
+ },
1873
+ {
1874
+ "epoch": 1.4797162491686988,
1875
+ "grad_norm": 0.0963672623038292,
1876
+ "learning_rate": 0.00010704222774716177,
1877
+ "loss": 0.6156,
1878
+ "step": 6675
1879
+ },
1880
+ {
1881
+ "epoch": 1.4852582575925515,
1882
+ "grad_norm": 0.14681296050548553,
1883
+ "learning_rate": 0.00010644528326845988,
1884
+ "loss": 0.74,
1885
+ "step": 6700
1886
+ },
1887
+ {
1888
+ "epoch": 1.4908002660164044,
1889
+ "grad_norm": 0.09270428866147995,
1890
+ "learning_rate": 0.00010584810806749327,
1891
+ "loss": 0.6185,
1892
+ "step": 6725
1893
+ },
1894
+ {
1895
+ "epoch": 1.496342274440257,
1896
+ "grad_norm": 0.11598405987024307,
1897
+ "learning_rate": 0.00010525072352138526,
1898
+ "loss": 0.7463,
1899
+ "step": 6750
1900
+ },
1901
+ {
1902
+ "epoch": 1.50188428286411,
1903
+ "grad_norm": 0.09355468302965164,
1904
+ "learning_rate": 0.00010465315101475295,
1905
+ "loss": 0.5996,
1906
+ "step": 6775
1907
+ },
1908
+ {
1909
+ "epoch": 1.5074262912879628,
1910
+ "grad_norm": 0.15108546614646912,
1911
+ "learning_rate": 0.00010405541193894204,
1912
+ "loss": 0.7512,
1913
+ "step": 6800
1914
+ },
1915
+ {
1916
+ "epoch": 1.5129682997118157,
1917
+ "grad_norm": 0.08825406432151794,
1918
+ "learning_rate": 0.00010345752769126079,
1919
+ "loss": 0.6367,
1920
+ "step": 6825
1921
+ },
1922
+ {
1923
+ "epoch": 1.5185103081356683,
1924
+ "grad_norm": 0.11757966130971909,
1925
+ "learning_rate": 0.0001028595196742143,
1926
+ "loss": 0.7556,
1927
+ "step": 6850
1928
+ },
1929
+ {
1930
+ "epoch": 1.5240523165595212,
1931
+ "grad_norm": 0.08973203599452972,
1932
+ "learning_rate": 0.00010226140929473813,
1933
+ "loss": 0.6038,
1934
+ "step": 6875
1935
+ },
1936
+ {
1937
+ "epoch": 1.5295943249833739,
1938
+ "grad_norm": 0.1337171196937561,
1939
+ "learning_rate": 0.00010166321796343223,
1940
+ "loss": 0.7388,
1941
+ "step": 6900
1942
+ },
1943
+ {
1944
+ "epoch": 1.5351363334072268,
1945
+ "grad_norm": 0.08748678117990494,
1946
+ "learning_rate": 0.0001010649670937943,
1947
+ "loss": 0.6093,
1948
+ "step": 6925
1949
+ },
1950
+ {
1951
+ "epoch": 1.5406783418310797,
1952
+ "grad_norm": 0.1228007897734642,
1953
+ "learning_rate": 0.00010046667810145338,
1954
+ "loss": 0.7494,
1955
+ "step": 6950
1956
+ },
1957
+ {
1958
+ "epoch": 1.5462203502549325,
1959
+ "grad_norm": 0.09511099755764008,
1960
+ "learning_rate": 9.986837240340319e-05,
1961
+ "loss": 0.5998,
1962
+ "step": 6975
1963
+ },
1964
+ {
1965
+ "epoch": 1.5517623586787852,
1966
+ "grad_norm": 0.12583385407924652,
1967
+ "learning_rate": 9.927007141723548e-05,
1968
+ "loss": 0.7266,
1969
+ "step": 7000
1970
+ },
1971
+ {
1972
+ "epoch": 1.5573043671026379,
1973
+ "grad_norm": 0.08915423601865768,
1974
+ "learning_rate": 9.867179656037326e-05,
1975
+ "loss": 0.638,
1976
+ "step": 7025
1977
+ },
1978
+ {
1979
+ "epoch": 1.5628463755264907,
1980
+ "grad_norm": 0.1239473968744278,
1981
+ "learning_rate": 9.80735692493043e-05,
1982
+ "loss": 0.7473,
1983
+ "step": 7050
1984
+ },
1985
+ {
1986
+ "epoch": 1.5683883839503436,
1987
+ "grad_norm": 0.09568199515342712,
1988
+ "learning_rate": 9.747541089881424e-05,
1989
+ "loss": 0.6174,
1990
+ "step": 7075
1991
+ },
1992
+ {
1993
+ "epoch": 1.5739303923741965,
1994
+ "grad_norm": 0.13295117020606995,
1995
+ "learning_rate": 9.687734292122024e-05,
1996
+ "loss": 0.7278,
1997
+ "step": 7100
1998
+ },
1999
+ {
2000
+ "epoch": 1.5794724007980492,
2001
+ "grad_norm": 0.09703335911035538,
2002
+ "learning_rate": 9.627938672560432e-05,
2003
+ "loss": 0.6051,
2004
+ "step": 7125
2005
+ },
2006
+ {
2007
+ "epoch": 1.585014409221902,
2008
+ "grad_norm": 0.12026989459991455,
2009
+ "learning_rate": 9.568156371704705e-05,
2010
+ "loss": 0.7341,
2011
+ "step": 7150
2012
+ },
2013
+ {
2014
+ "epoch": 1.5905564176457547,
2015
+ "grad_norm": 0.09185943752527237,
2016
+ "learning_rate": 9.508389529586128e-05,
2017
+ "loss": 0.607,
2018
+ "step": 7175
2019
+ },
2020
+ {
2021
+ "epoch": 1.5960984260696076,
2022
+ "grad_norm": 0.16115328669548035,
2023
+ "learning_rate": 9.448640285682613e-05,
2024
+ "loss": 0.7321,
2025
+ "step": 7200
2026
+ },
2027
+ {
2028
+ "epoch": 1.6016404344934605,
2029
+ "grad_norm": 0.09262697398662567,
2030
+ "learning_rate": 9.388910778842103e-05,
2031
+ "loss": 0.6027,
2032
+ "step": 7225
2033
+ },
2034
+ {
2035
+ "epoch": 1.6071824429173134,
2036
+ "grad_norm": 0.17859314382076263,
2037
+ "learning_rate": 9.329203147206007e-05,
2038
+ "loss": 0.7498,
2039
+ "step": 7250
2040
+ },
2041
+ {
2042
+ "epoch": 1.612724451341166,
2043
+ "grad_norm": 0.09392908960580826,
2044
+ "learning_rate": 9.269519528132677e-05,
2045
+ "loss": 0.6035,
2046
+ "step": 7275
2047
+ },
2048
+ {
2049
+ "epoch": 1.6182664597650187,
2050
+ "grad_norm": 0.1226978749036789,
2051
+ "learning_rate": 9.209862058120879e-05,
2052
+ "loss": 0.753,
2053
+ "step": 7300
2054
+ },
2055
+ {
2056
+ "epoch": 1.6238084681888716,
2057
+ "grad_norm": 0.09812294691801071,
2058
+ "learning_rate": 9.15023287273332e-05,
2059
+ "loss": 0.631,
2060
+ "step": 7325
2061
+ },
2062
+ {
2063
+ "epoch": 1.6293504766127245,
2064
+ "grad_norm": 0.14241814613342285,
2065
+ "learning_rate": 9.0906341065202e-05,
2066
+ "loss": 0.7291,
2067
+ "step": 7350
2068
+ },
2069
+ {
2070
+ "epoch": 1.6348924850365774,
2071
+ "grad_norm": 0.09261428564786911,
2072
+ "learning_rate": 9.031067892942805e-05,
2073
+ "loss": 0.6204,
2074
+ "step": 7375
2075
+ },
2076
+ {
2077
+ "epoch": 1.64043449346043,
2078
+ "grad_norm": 0.1319173127412796,
2079
+ "learning_rate": 8.971536364297126e-05,
2080
+ "loss": 0.7422,
2081
+ "step": 7400
2082
+ },
2083
+ {
2084
+ "epoch": 1.645976501884283,
2085
+ "grad_norm": 0.09849465638399124,
2086
+ "learning_rate": 8.912041651637541e-05,
2087
+ "loss": 0.6212,
2088
+ "step": 7425
2089
+ },
2090
+ {
2091
+ "epoch": 1.6515185103081356,
2092
+ "grad_norm": 0.1365344524383545,
2093
+ "learning_rate": 8.852585884700519e-05,
2094
+ "loss": 0.735,
2095
+ "step": 7450
2096
+ },
2097
+ {
2098
+ "epoch": 1.6570605187319885,
2099
+ "grad_norm": 0.0948396623134613,
2100
+ "learning_rate": 8.79317119182839e-05,
2101
+ "loss": 0.6113,
2102
+ "step": 7475
2103
+ },
2104
+ {
2105
+ "epoch": 1.6626025271558413,
2106
+ "grad_norm": 0.14469577372074127,
2107
+ "learning_rate": 8.73379969989315e-05,
2108
+ "loss": 0.7577,
2109
+ "step": 7500
2110
+ },
2111
+ {
2112
+ "epoch": 1.6681445355796942,
2113
+ "grad_norm": 0.08975006639957428,
2114
+ "learning_rate": 8.674473534220326e-05,
2115
+ "loss": 0.6123,
2116
+ "step": 7525
2117
+ },
2118
+ {
2119
+ "epoch": 1.6736865440035469,
2120
+ "grad_norm": 0.18550659716129303,
2121
+ "learning_rate": 8.615194818512905e-05,
2122
+ "loss": 0.7173,
2123
+ "step": 7550
2124
+ },
2125
+ {
2126
+ "epoch": 1.6792285524273995,
2127
+ "grad_norm": 0.09581893682479858,
2128
+ "learning_rate": 8.555965674775295e-05,
2129
+ "loss": 0.6052,
2130
+ "step": 7575
2131
+ },
2132
+ {
2133
+ "epoch": 1.6847705608512524,
2134
+ "grad_norm": 0.15174435079097748,
2135
+ "learning_rate": 8.496788223237381e-05,
2136
+ "loss": 0.741,
2137
+ "step": 7600
2138
+ },
2139
+ {
2140
+ "epoch": 1.6903125692751053,
2141
+ "grad_norm": 0.09482391923666,
2142
+ "learning_rate": 8.43766458227861e-05,
2143
+ "loss": 0.6281,
2144
+ "step": 7625
2145
+ },
2146
+ {
2147
+ "epoch": 1.6958545776989582,
2148
+ "grad_norm": 0.12190216034650803,
2149
+ "learning_rate": 8.37859686835218e-05,
2150
+ "loss": 0.7489,
2151
+ "step": 7650
2152
+ },
2153
+ {
2154
+ "epoch": 1.7013965861228109,
2155
+ "grad_norm": 0.09474999457597733,
2156
+ "learning_rate": 8.319587195909251e-05,
2157
+ "loss": 0.6238,
2158
+ "step": 7675
2159
+ },
2160
+ {
2161
+ "epoch": 1.7069385945466637,
2162
+ "grad_norm": 0.13137421011924744,
2163
+ "learning_rate": 8.260637677323279e-05,
2164
+ "loss": 0.7256,
2165
+ "step": 7700
2166
+ },
2167
+ {
2168
+ "epoch": 1.7124806029705164,
2169
+ "grad_norm": 0.09658095240592957,
2170
+ "learning_rate": 8.201750422814379e-05,
2171
+ "loss": 0.6038,
2172
+ "step": 7725
2173
+ },
2174
+ {
2175
+ "epoch": 1.7180226113943693,
2176
+ "grad_norm": 0.15730302035808563,
2177
+ "learning_rate": 8.142927540373805e-05,
2178
+ "loss": 0.7255,
2179
+ "step": 7750
2180
+ },
2181
+ {
2182
+ "epoch": 1.7235646198182222,
2183
+ "grad_norm": 0.09805120527744293,
2184
+ "learning_rate": 8.084171135688467e-05,
2185
+ "loss": 0.6136,
2186
+ "step": 7775
2187
+ },
2188
+ {
2189
+ "epoch": 1.729106628242075,
2190
+ "grad_norm": 0.14167381823062897,
2191
+ "learning_rate": 8.02548331206558e-05,
2192
+ "loss": 0.7246,
2193
+ "step": 7800
2194
+ },
2195
+ {
2196
+ "epoch": 1.7346486366659277,
2197
+ "grad_norm": 0.09563518315553665,
2198
+ "learning_rate": 7.966866170357346e-05,
2199
+ "loss": 0.6199,
2200
+ "step": 7825
2201
+ },
2202
+ {
2203
+ "epoch": 1.7401906450897804,
2204
+ "grad_norm": 0.11947919428348541,
2205
+ "learning_rate": 7.908321808885766e-05,
2206
+ "loss": 0.7284,
2207
+ "step": 7850
2208
+ },
2209
+ {
2210
+ "epoch": 1.7457326535136333,
2211
+ "grad_norm": 0.09702473878860474,
2212
+ "learning_rate": 7.849852323367521e-05,
2213
+ "loss": 0.6343,
2214
+ "step": 7875
2215
+ },
2216
+ {
2217
+ "epoch": 1.7512746619374862,
2218
+ "grad_norm": 0.16644233465194702,
2219
+ "learning_rate": 7.791459806838957e-05,
2220
+ "loss": 0.7295,
2221
+ "step": 7900
2222
+ },
2223
+ {
2224
+ "epoch": 1.756816670361339,
2225
+ "grad_norm": 0.09687721729278564,
2226
+ "learning_rate": 7.733146349581144e-05,
2227
+ "loss": 0.6232,
2228
+ "step": 7925
2229
+ },
2230
+ {
2231
+ "epoch": 1.7623586787851917,
2232
+ "grad_norm": 0.1432383805513382,
2233
+ "learning_rate": 7.674914039045076e-05,
2234
+ "loss": 0.7351,
2235
+ "step": 7950
2236
+ },
2237
+ {
2238
+ "epoch": 1.7679006872090446,
2239
+ "grad_norm": 0.09745761752128601,
2240
+ "learning_rate": 7.61676495977692e-05,
2241
+ "loss": 0.5918,
2242
+ "step": 7975
2243
+ },
2244
+ {
2245
+ "epoch": 1.7734426956328972,
2246
+ "grad_norm": 0.1507551074028015,
2247
+ "learning_rate": 7.558701193343419e-05,
2248
+ "loss": 0.7384,
2249
+ "step": 8000
2250
+ },
2251
+ {
2252
+ "epoch": 1.7789847040567501,
2253
+ "grad_norm": 0.09434136003255844,
2254
+ "learning_rate": 7.500724818257351e-05,
2255
+ "loss": 0.5987,
2256
+ "step": 8025
2257
+ },
2258
+ {
2259
+ "epoch": 1.784526712480603,
2260
+ "grad_norm": 0.1321529746055603,
2261
+ "learning_rate": 7.442837909903156e-05,
2262
+ "loss": 0.7409,
2263
+ "step": 8050
2264
+ },
2265
+ {
2266
+ "epoch": 1.790068720904456,
2267
+ "grad_norm": 0.09421923011541367,
2268
+ "learning_rate": 7.385042540462615e-05,
2269
+ "loss": 0.6042,
2270
+ "step": 8075
2271
+ },
2272
+ {
2273
+ "epoch": 1.7956107293283086,
2274
+ "grad_norm": 0.1154310330748558,
2275
+ "learning_rate": 7.32734077884069e-05,
2276
+ "loss": 0.7319,
2277
+ "step": 8100
2278
+ },
2279
+ {
2280
+ "epoch": 1.8011527377521612,
2281
+ "grad_norm": 0.09810709953308105,
2282
+ "learning_rate": 7.272037071314008e-05,
2283
+ "loss": 0.6196,
2284
+ "step": 8125
2285
+ },
2286
+ {
2287
+ "epoch": 1.806694746176014,
2288
+ "grad_norm": 0.12382727861404419,
2289
+ "learning_rate": 7.2145247695974e-05,
2290
+ "loss": 0.7366,
2291
+ "step": 8150
2292
+ },
2293
+ {
2294
+ "epoch": 1.812236754599867,
2295
+ "grad_norm": 0.10054343193769455,
2296
+ "learning_rate": 7.157112179736207e-05,
2297
+ "loss": 0.638,
2298
+ "step": 8175
2299
+ },
2300
+ {
2301
+ "epoch": 1.8177787630237199,
2302
+ "grad_norm": 0.12496750056743622,
2303
+ "learning_rate": 7.099801356933004e-05,
2304
+ "loss": 0.7435,
2305
+ "step": 8200
2306
+ },
2307
+ {
2308
+ "epoch": 1.8233207714475728,
2309
+ "grad_norm": 0.09787677973508835,
2310
+ "learning_rate": 7.0425943527474e-05,
2311
+ "loss": 0.5983,
2312
+ "step": 8225
2313
+ },
2314
+ {
2315
+ "epoch": 1.8288627798714254,
2316
+ "grad_norm": 0.16083738207817078,
2317
+ "learning_rate": 6.985493215022605e-05,
2318
+ "loss": 0.719,
2319
+ "step": 8250
2320
+ },
2321
+ {
2322
+ "epoch": 1.834404788295278,
2323
+ "grad_norm": 0.09314879029989243,
2324
+ "learning_rate": 6.928499987812112e-05,
2325
+ "loss": 0.6156,
2326
+ "step": 8275
2327
+ },
2328
+ {
2329
+ "epoch": 1.839946796719131,
2330
+ "grad_norm": 0.1240190863609314,
2331
+ "learning_rate": 6.871616711306545e-05,
2332
+ "loss": 0.7312,
2333
+ "step": 8300
2334
+ },
2335
+ {
2336
+ "epoch": 1.8454888051429839,
2337
+ "grad_norm": 0.10087582468986511,
2338
+ "learning_rate": 6.814845421760602e-05,
2339
+ "loss": 0.5953,
2340
+ "step": 8325
2341
+ },
2342
+ {
2343
+ "epoch": 1.8510308135668367,
2344
+ "grad_norm": 0.15391846001148224,
2345
+ "learning_rate": 6.758188151420189e-05,
2346
+ "loss": 0.7372,
2347
+ "step": 8350
2348
+ },
2349
+ {
2350
+ "epoch": 1.8565728219906894,
2351
+ "grad_norm": 0.09913575649261475,
2352
+ "learning_rate": 6.701646928449646e-05,
2353
+ "loss": 0.5859,
2354
+ "step": 8375
2355
+ },
2356
+ {
2357
+ "epoch": 1.862114830414542,
2358
+ "grad_norm": 0.1197889968752861,
2359
+ "learning_rate": 6.645223776859166e-05,
2360
+ "loss": 0.7416,
2361
+ "step": 8400
2362
+ },
2363
+ {
2364
+ "epoch": 1.867656838838395,
2365
+ "grad_norm": 0.09789609163999557,
2366
+ "learning_rate": 6.588920716432329e-05,
2367
+ "loss": 0.6422,
2368
+ "step": 8425
2369
+ },
2370
+ {
2371
+ "epoch": 1.8731988472622478,
2372
+ "grad_norm": 0.134750634431839,
2373
+ "learning_rate": 6.532739762653804e-05,
2374
+ "loss": 0.7462,
2375
+ "step": 8450
2376
+ },
2377
+ {
2378
+ "epoch": 1.8787408556861007,
2379
+ "grad_norm": 0.09457212686538696,
2380
+ "learning_rate": 6.476682926637197e-05,
2381
+ "loss": 0.6161,
2382
+ "step": 8475
2383
+ },
2384
+ {
2385
+ "epoch": 1.8842828641099536,
2386
+ "grad_norm": 0.18783515691757202,
2387
+ "learning_rate": 6.420752215053065e-05,
2388
+ "loss": 0.7236,
2389
+ "step": 8500
2390
+ },
2391
+ {
2392
+ "epoch": 1.8898248725338063,
2393
+ "grad_norm": 0.09566831588745117,
2394
+ "learning_rate": 6.364949630057078e-05,
2395
+ "loss": 0.5882,
2396
+ "step": 8525
2397
+ },
2398
+ {
2399
+ "epoch": 1.895366880957659,
2400
+ "grad_norm": 0.12137682735919952,
2401
+ "learning_rate": 6.30927716921835e-05,
2402
+ "loss": 0.7284,
2403
+ "step": 8550
2404
+ },
2405
+ {
2406
+ "epoch": 1.9009088893815118,
2407
+ "grad_norm": 0.09804505854845047,
2408
+ "learning_rate": 6.25373682544793e-05,
2409
+ "loss": 0.6165,
2410
+ "step": 8575
2411
+ },
2412
+ {
2413
+ "epoch": 1.9064508978053647,
2414
+ "grad_norm": 0.20505575835704803,
2415
+ "learning_rate": 6.198330586927463e-05,
2416
+ "loss": 0.7224,
2417
+ "step": 8600
2418
+ },
2419
+ {
2420
+ "epoch": 1.9119929062292176,
2421
+ "grad_norm": 0.10258720070123672,
2422
+ "learning_rate": 6.14306043703802e-05,
2423
+ "loss": 0.5683,
2424
+ "step": 8625
2425
+ },
2426
+ {
2427
+ "epoch": 1.9175349146530702,
2428
+ "grad_norm": 0.14654862880706787,
2429
+ "learning_rate": 6.087928354289103e-05,
2430
+ "loss": 0.7336,
2431
+ "step": 8650
2432
+ },
2433
+ {
2434
+ "epoch": 1.9230769230769231,
2435
+ "grad_norm": 0.09862499684095383,
2436
+ "learning_rate": 6.0329363122478e-05,
2437
+ "loss": 0.5824,
2438
+ "step": 8675
2439
+ },
2440
+ {
2441
+ "epoch": 1.9286189315007758,
2442
+ "grad_norm": 0.135267972946167,
2443
+ "learning_rate": 5.978086279468163e-05,
2444
+ "loss": 0.744,
2445
+ "step": 8700
2446
+ },
2447
+ {
2448
+ "epoch": 1.9341609399246287,
2449
+ "grad_norm": 0.1033608540892601,
2450
+ "learning_rate": 5.923380219420729e-05,
2451
+ "loss": 0.6134,
2452
+ "step": 8725
2453
+ },
2454
+ {
2455
+ "epoch": 1.9397029483484816,
2456
+ "grad_norm": 0.1315843164920807,
2457
+ "learning_rate": 5.8688200904222266e-05,
2458
+ "loss": 0.7151,
2459
+ "step": 8750
2460
+ },
2461
+ {
2462
+ "epoch": 1.9452449567723344,
2463
+ "grad_norm": 0.09951479732990265,
2464
+ "learning_rate": 5.8144078455654846e-05,
2465
+ "loss": 0.622,
2466
+ "step": 8775
2467
+ },
2468
+ {
2469
+ "epoch": 1.950786965196187,
2470
+ "grad_norm": 0.2037511169910431,
2471
+ "learning_rate": 5.760145432649515e-05,
2472
+ "loss": 0.7239,
2473
+ "step": 8800
2474
+ },
2475
+ {
2476
+ "epoch": 1.9563289736200398,
2477
+ "grad_norm": 0.10025127977132797,
2478
+ "learning_rate": 5.706034794109778e-05,
2479
+ "loss": 0.5922,
2480
+ "step": 8825
2481
+ },
2482
+ {
2483
+ "epoch": 1.9618709820438927,
2484
+ "grad_norm": 0.18914011120796204,
2485
+ "learning_rate": 5.65207786694866e-05,
2486
+ "loss": 0.7205,
2487
+ "step": 8850
2488
+ },
2489
+ {
2490
+ "epoch": 1.9674129904677455,
2491
+ "grad_norm": 0.10025196522474289,
2492
+ "learning_rate": 5.5982765826661256e-05,
2493
+ "loss": 0.5814,
2494
+ "step": 8875
2495
+ },
2496
+ {
2497
+ "epoch": 1.9729549988915984,
2498
+ "grad_norm": 0.12357232719659805,
2499
+ "learning_rate": 5.544632867190591e-05,
2500
+ "loss": 0.7217,
2501
+ "step": 8900
2502
+ },
2503
+ {
2504
+ "epoch": 1.978497007315451,
2505
+ "grad_norm": 0.09909965842962265,
2506
+ "learning_rate": 5.491148640809962e-05,
2507
+ "loss": 0.6102,
2508
+ "step": 8925
2509
+ },
2510
+ {
2511
+ "epoch": 1.984039015739304,
2512
+ "grad_norm": 0.16131410002708435,
2513
+ "learning_rate": 5.437825818102902e-05,
2514
+ "loss": 0.7193,
2515
+ "step": 8950
2516
+ },
2517
+ {
2518
+ "epoch": 1.9895810241631566,
2519
+ "grad_norm": 0.10195192694664001,
2520
+ "learning_rate": 5.384666307870293e-05,
2521
+ "loss": 0.5989,
2522
+ "step": 8975
2523
+ },
2524
+ {
2525
+ "epoch": 1.9951230325870095,
2526
+ "grad_norm": 0.1269746869802475,
2527
+ "learning_rate": 5.331672013066922e-05,
2528
+ "loss": 0.7287,
2529
+ "step": 9000
2530
+ },
2531
+ {
2532
+ "epoch": 2.0006650410108624,
2533
+ "grad_norm": 0.09312586486339569,
2534
+ "learning_rate": 5.278844830733332e-05,
2535
+ "loss": 0.6024,
2536
+ "step": 9025
2537
+ },
2538
+ {
2539
+ "epoch": 2.0062070494347153,
2540
+ "grad_norm": 0.10967884957790375,
2541
+ "learning_rate": 5.226186651927938e-05,
2542
+ "loss": 0.7053,
2543
+ "step": 9050
2544
+ },
2545
+ {
2546
+ "epoch": 2.0117490578585677,
2547
+ "grad_norm": 0.09102078527212143,
2548
+ "learning_rate": 5.1736993616593165e-05,
2549
+ "loss": 0.5861,
2550
+ "step": 9075
2551
+ },
2552
+ {
2553
+ "epoch": 2.0172910662824206,
2554
+ "grad_norm": 0.10821503400802612,
2555
+ "learning_rate": 5.121384838818746e-05,
2556
+ "loss": 0.6865,
2557
+ "step": 9100
2558
+ },
2559
+ {
2560
+ "epoch": 2.0228330747062735,
2561
+ "grad_norm": 0.09317923337221146,
2562
+ "learning_rate": 5.0692449561129285e-05,
2563
+ "loss": 0.5912,
2564
+ "step": 9125
2565
+ },
2566
+ {
2567
+ "epoch": 2.0283750831301264,
2568
+ "grad_norm": 0.11409013718366623,
2569
+ "learning_rate": 5.017281579996961e-05,
2570
+ "loss": 0.6979,
2571
+ "step": 9150
2572
+ },
2573
+ {
2574
+ "epoch": 2.0339170915539793,
2575
+ "grad_norm": 0.09311998635530472,
2576
+ "learning_rate": 4.965496570607523e-05,
2577
+ "loss": 0.6235,
2578
+ "step": 9175
2579
+ },
2580
+ {
2581
+ "epoch": 2.039459099977832,
2582
+ "grad_norm": 0.12173454463481903,
2583
+ "learning_rate": 4.913891781696285e-05,
2584
+ "loss": 0.7002,
2585
+ "step": 9200
2586
+ },
2587
+ {
2588
+ "epoch": 2.0450011084016846,
2589
+ "grad_norm": 0.10086531937122345,
2590
+ "learning_rate": 4.8624690605635626e-05,
2591
+ "loss": 0.5569,
2592
+ "step": 9225
2593
+ },
2594
+ {
2595
+ "epoch": 2.0505431168255375,
2596
+ "grad_norm": 0.10775309801101685,
2597
+ "learning_rate": 4.811230247992165e-05,
2598
+ "loss": 0.6925,
2599
+ "step": 9250
2600
+ },
2601
+ {
2602
+ "epoch": 2.0560851252493904,
2603
+ "grad_norm": 0.096441850066185,
2604
+ "learning_rate": 4.760177178181521e-05,
2605
+ "loss": 0.5906,
2606
+ "step": 9275
2607
+ },
2608
+ {
2609
+ "epoch": 2.0616271336732432,
2610
+ "grad_norm": 0.11258859932422638,
2611
+ "learning_rate": 4.709311678682005e-05,
2612
+ "loss": 0.6883,
2613
+ "step": 9300
2614
+ },
2615
+ {
2616
+ "epoch": 2.067169142097096,
2617
+ "grad_norm": 0.0907958596944809,
2618
+ "learning_rate": 4.658635570329537e-05,
2619
+ "loss": 0.6069,
2620
+ "step": 9325
2621
+ },
2622
+ {
2623
+ "epoch": 2.0727111505209486,
2624
+ "grad_norm": 0.11246030032634735,
2625
+ "learning_rate": 4.608150667180378e-05,
2626
+ "loss": 0.6951,
2627
+ "step": 9350
2628
+ },
2629
+ {
2630
+ "epoch": 2.0782531589448014,
2631
+ "grad_norm": 0.09199715405702591,
2632
+ "learning_rate": 4.557858776446203e-05,
2633
+ "loss": 0.5949,
2634
+ "step": 9375
2635
+ },
2636
+ {
2637
+ "epoch": 2.0837951673686543,
2638
+ "grad_norm": 0.11862944066524506,
2639
+ "learning_rate": 4.50776169842941e-05,
2640
+ "loss": 0.6701,
2641
+ "step": 9400
2642
+ },
2643
+ {
2644
+ "epoch": 2.089337175792507,
2645
+ "grad_norm": 0.09631045907735825,
2646
+ "learning_rate": 4.457861226458678e-05,
2647
+ "loss": 0.6158,
2648
+ "step": 9425
2649
+ },
2650
+ {
2651
+ "epoch": 2.09487918421636,
2652
+ "grad_norm": 0.107430100440979,
2653
+ "learning_rate": 4.408159146824756e-05,
2654
+ "loss": 0.6998,
2655
+ "step": 9450
2656
+ },
2657
+ {
2658
+ "epoch": 2.100421192640213,
2659
+ "grad_norm": 0.0933394506573677,
2660
+ "learning_rate": 4.358657238716533e-05,
2661
+ "loss": 0.5998,
2662
+ "step": 9475
2663
+ },
2664
+ {
2665
+ "epoch": 2.1059632010640654,
2666
+ "grad_norm": 0.11312496662139893,
2667
+ "learning_rate": 4.309357274157338e-05,
2668
+ "loss": 0.6909,
2669
+ "step": 9500
2670
+ },
2671
+ {
2672
+ "epoch": 2.1115052094879183,
2673
+ "grad_norm": 0.09570565819740295,
2674
+ "learning_rate": 4.260261017941526e-05,
2675
+ "loss": 0.5805,
2676
+ "step": 9525
2677
+ },
2678
+ {
2679
+ "epoch": 2.117047217911771,
2680
+ "grad_norm": 0.11327219754457474,
2681
+ "learning_rate": 4.211370227571276e-05,
2682
+ "loss": 0.6968,
2683
+ "step": 9550
2684
+ },
2685
+ {
2686
+ "epoch": 2.122589226335624,
2687
+ "grad_norm": 0.09823332726955414,
2688
+ "learning_rate": 4.162686653193698e-05,
2689
+ "loss": 0.5965,
2690
+ "step": 9575
2691
+ },
2692
+ {
2693
+ "epoch": 2.128131234759477,
2694
+ "grad_norm": 0.11287786811590195,
2695
+ "learning_rate": 4.11421203753817e-05,
2696
+ "loss": 0.6936,
2697
+ "step": 9600
2698
+ },
2699
+ {
2700
+ "epoch": 2.1336732431833294,
2701
+ "grad_norm": 0.09058432281017303,
2702
+ "learning_rate": 4.065948115853973e-05,
2703
+ "loss": 0.61,
2704
+ "step": 9625
2705
+ },
2706
+ {
2707
+ "epoch": 2.1392152516071823,
2708
+ "grad_norm": 0.11232877522706985,
2709
+ "learning_rate": 4.017896615848149e-05,
2710
+ "loss": 0.6852,
2711
+ "step": 9650
2712
+ },
2713
+ {
2714
+ "epoch": 2.144757260031035,
2715
+ "grad_norm": 0.09588344395160675,
2716
+ "learning_rate": 3.9700592576236686e-05,
2717
+ "loss": 0.6083,
2718
+ "step": 9675
2719
+ },
2720
+ {
2721
+ "epoch": 2.150299268454888,
2722
+ "grad_norm": 0.11739671975374222,
2723
+ "learning_rate": 3.922437753617856e-05,
2724
+ "loss": 0.6997,
2725
+ "step": 9700
2726
+ },
2727
+ {
2728
+ "epoch": 2.155841276878741,
2729
+ "grad_norm": 0.09523261338472366,
2730
+ "learning_rate": 3.875033808541083e-05,
2731
+ "loss": 0.5693,
2732
+ "step": 9725
2733
+ },
2734
+ {
2735
+ "epoch": 2.161383285302594,
2736
+ "grad_norm": 0.11812377721071243,
2737
+ "learning_rate": 3.827849119315755e-05,
2738
+ "loss": 0.6907,
2739
+ "step": 9750
2740
+ },
2741
+ {
2742
+ "epoch": 2.1669252937264463,
2743
+ "grad_norm": 0.09816546738147736,
2744
+ "learning_rate": 3.780885375015549e-05,
2745
+ "loss": 0.5891,
2746
+ "step": 9775
2747
+ },
2748
+ {
2749
+ "epoch": 2.172467302150299,
2750
+ "grad_norm": 0.11008067429065704,
2751
+ "learning_rate": 3.734144256804978e-05,
2752
+ "loss": 0.691,
2753
+ "step": 9800
2754
+ },
2755
+ {
2756
+ "epoch": 2.178009310574152,
2757
+ "grad_norm": 0.09762485325336456,
2758
+ "learning_rate": 3.687627437879177e-05,
2759
+ "loss": 0.5914,
2760
+ "step": 9825
2761
+ },
2762
+ {
2763
+ "epoch": 2.183551318998005,
2764
+ "grad_norm": 0.11886761337518692,
2765
+ "learning_rate": 3.6413365834040326e-05,
2766
+ "loss": 0.6896,
2767
+ "step": 9850
2768
+ },
2769
+ {
2770
+ "epoch": 2.189093327421858,
2771
+ "grad_norm": 0.09332608431577682,
2772
+ "learning_rate": 3.595273350456557e-05,
2773
+ "loss": 0.5844,
2774
+ "step": 9875
2775
+ },
2776
+ {
2777
+ "epoch": 2.1946353358457107,
2778
+ "grad_norm": 0.1171206533908844,
2779
+ "learning_rate": 3.549439387965592e-05,
2780
+ "loss": 0.7006,
2781
+ "step": 9900
2782
+ },
2783
+ {
2784
+ "epoch": 2.200177344269563,
2785
+ "grad_norm": 0.09177059680223465,
2786
+ "learning_rate": 3.503836336652756e-05,
2787
+ "loss": 0.589,
2788
+ "step": 9925
2789
+ },
2790
+ {
2791
+ "epoch": 2.205719352693416,
2792
+ "grad_norm": 0.11849093437194824,
2793
+ "learning_rate": 3.4584658289737296e-05,
2794
+ "loss": 0.687,
2795
+ "step": 9950
2796
+ },
2797
+ {
2798
+ "epoch": 2.211261361117269,
2799
+ "grad_norm": 0.09985481947660446,
2800
+ "learning_rate": 3.4133294890598065e-05,
2801
+ "loss": 0.6102,
2802
+ "step": 9975
2803
+ },
2804
+ {
2805
+ "epoch": 2.216803369541122,
2806
+ "grad_norm": 0.11642364412546158,
2807
+ "learning_rate": 3.3684289326597726e-05,
2808
+ "loss": 0.6963,
2809
+ "step": 10000
2810
+ },
2811
+ {
2812
+ "epoch": 2.2223453779649747,
2813
+ "grad_norm": 0.08828485757112503,
2814
+ "learning_rate": 3.323765767082042e-05,
2815
+ "loss": 0.5929,
2816
+ "step": 10025
2817
+ },
2818
+ {
2819
+ "epoch": 2.227887386388827,
2820
+ "grad_norm": 0.11624643206596375,
2821
+ "learning_rate": 3.2793415911371386e-05,
2822
+ "loss": 0.7003,
2823
+ "step": 10050
2824
+ },
2825
+ {
2826
+ "epoch": 2.23342939481268,
2827
+ "grad_norm": 0.09697126597166061,
2828
+ "learning_rate": 3.235157995080451e-05,
2829
+ "loss": 0.5912,
2830
+ "step": 10075
2831
+ },
2832
+ {
2833
+ "epoch": 2.238971403236533,
2834
+ "grad_norm": 0.11310283094644547,
2835
+ "learning_rate": 3.191216560555326e-05,
2836
+ "loss": 0.672,
2837
+ "step": 10100
2838
+ },
2839
+ {
2840
+ "epoch": 2.2445134116603858,
2841
+ "grad_norm": 0.09560606628656387,
2842
+ "learning_rate": 3.147518860536422e-05,
2843
+ "loss": 0.5817,
2844
+ "step": 10125
2845
+ },
2846
+ {
2847
+ "epoch": 2.2500554200842386,
2848
+ "grad_norm": 0.12067476660013199,
2849
+ "learning_rate": 3.105799826047285e-05,
2850
+ "loss": 0.6943,
2851
+ "step": 10150
2852
+ },
2853
+ {
2854
+ "epoch": 2.255597428508091,
2855
+ "grad_norm": 0.09630808234214783,
2856
+ "learning_rate": 3.062584375085582e-05,
2857
+ "loss": 0.5833,
2858
+ "step": 10175
2859
+ },
2860
+ {
2861
+ "epoch": 2.261139436931944,
2862
+ "grad_norm": 0.11377058923244476,
2863
+ "learning_rate": 3.0196172632856158e-05,
2864
+ "loss": 0.6931,
2865
+ "step": 10200
2866
+ },
2867
+ {
2868
+ "epoch": 2.266681445355797,
2869
+ "grad_norm": 0.09476067125797272,
2870
+ "learning_rate": 2.9769000287441484e-05,
2871
+ "loss": 0.5616,
2872
+ "step": 10225
2873
+ },
2874
+ {
2875
+ "epoch": 2.2722234537796497,
2876
+ "grad_norm": 0.11672156304121017,
2877
+ "learning_rate": 2.9344342006130754e-05,
2878
+ "loss": 0.684,
2879
+ "step": 10250
2880
+ },
2881
+ {
2882
+ "epoch": 2.2777654622035026,
2883
+ "grad_norm": 0.09658654034137726,
2884
+ "learning_rate": 2.8922212990446716e-05,
2885
+ "loss": 0.5865,
2886
+ "step": 10275
2887
+ },
2888
+ {
2889
+ "epoch": 2.2833074706273555,
2890
+ "grad_norm": 0.11913245916366577,
2891
+ "learning_rate": 2.8502628351371842e-05,
2892
+ "loss": 0.6895,
2893
+ "step": 10300
2894
+ },
2895
+ {
2896
+ "epoch": 2.2888494790512084,
2897
+ "grad_norm": 0.0930788666009903,
2898
+ "learning_rate": 2.8102234790536696e-05,
2899
+ "loss": 0.6014,
2900
+ "step": 10325
2901
+ },
2902
+ {
2903
+ "epoch": 2.294391487475061,
2904
+ "grad_norm": 0.12074844539165497,
2905
+ "learning_rate": 2.7687680614304466e-05,
2906
+ "loss": 0.6919,
2907
+ "step": 10350
2908
+ },
2909
+ {
2910
+ "epoch": 2.2999334958989137,
2911
+ "grad_norm": 0.0964696928858757,
2912
+ "learning_rate": 2.72757150073249e-05,
2913
+ "loss": 0.5843,
2914
+ "step": 10375
2915
+ },
2916
+ {
2917
+ "epoch": 2.3054755043227666,
2918
+ "grad_norm": 0.11635053157806396,
2919
+ "learning_rate": 2.6866352716760167e-05,
2920
+ "loss": 0.697,
2921
+ "step": 10400
2922
+ },
2923
+ {
2924
+ "epoch": 2.3110175127466195,
2925
+ "grad_norm": 0.09311337023973465,
2926
+ "learning_rate": 2.6459608396581404e-05,
2927
+ "loss": 0.5885,
2928
+ "step": 10425
2929
+ },
2930
+ {
2931
+ "epoch": 2.3165595211704724,
2932
+ "grad_norm": 0.11619989573955536,
2933
+ "learning_rate": 2.6055496607044018e-05,
2934
+ "loss": 0.6825,
2935
+ "step": 10450
2936
+ },
2937
+ {
2938
+ "epoch": 2.322101529594325,
2939
+ "grad_norm": 0.09799373149871826,
2940
+ "learning_rate": 2.5654031814166524e-05,
2941
+ "loss": 0.6016,
2942
+ "step": 10475
2943
+ },
2944
+ {
2945
+ "epoch": 2.3276435380181777,
2946
+ "grad_norm": 0.11626887321472168,
2947
+ "learning_rate": 2.5255228389212803e-05,
2948
+ "loss": 0.6933,
2949
+ "step": 10500
2950
+ },
2951
+ {
2952
+ "epoch": 2.3331855464420306,
2953
+ "grad_norm": 0.09651979058980942,
2954
+ "learning_rate": 2.4859100608177456e-05,
2955
+ "loss": 0.595,
2956
+ "step": 10525
2957
+ },
2958
+ {
2959
+ "epoch": 2.3387275548658835,
2960
+ "grad_norm": 0.1168813407421112,
2961
+ "learning_rate": 2.44656626512749e-05,
2962
+ "loss": 0.6909,
2963
+ "step": 10550
2964
+ },
2965
+ {
2966
+ "epoch": 2.3442695632897363,
2967
+ "grad_norm": 0.09738708287477493,
2968
+ "learning_rate": 2.407492860243171e-05,
2969
+ "loss": 0.5848,
2970
+ "step": 10575
2971
+ },
2972
+ {
2973
+ "epoch": 2.3498115717135892,
2974
+ "grad_norm": 0.12800638377666473,
2975
+ "learning_rate": 2.368691244878254e-05,
2976
+ "loss": 0.6875,
2977
+ "step": 10600
2978
+ },
2979
+ {
2980
+ "epoch": 2.3553535801374417,
2981
+ "grad_norm": 0.09715036302804947,
2982
+ "learning_rate": 2.330162808016928e-05,
2983
+ "loss": 0.5575,
2984
+ "step": 10625
2985
+ },
2986
+ {
2987
+ "epoch": 2.3608955885612946,
2988
+ "grad_norm": 0.11949202418327332,
2989
+ "learning_rate": 2.2919089288643948e-05,
2990
+ "loss": 0.6873,
2991
+ "step": 10650
2992
+ },
2993
+ {
2994
+ "epoch": 2.3664375969851474,
2995
+ "grad_norm": 0.10298886895179749,
2996
+ "learning_rate": 2.253930976797489e-05,
2997
+ "loss": 0.5973,
2998
+ "step": 10675
2999
+ },
3000
+ {
3001
+ "epoch": 2.3719796054090003,
3002
+ "grad_norm": 0.1202768087387085,
3003
+ "learning_rate": 2.216230311315677e-05,
3004
+ "loss": 0.6889,
3005
+ "step": 10700
3006
+ },
3007
+ {
3008
+ "epoch": 2.377521613832853,
3009
+ "grad_norm": 0.10042477399110794,
3010
+ "learning_rate": 2.1788082819923637e-05,
3011
+ "loss": 0.5959,
3012
+ "step": 10725
3013
+ },
3014
+ {
3015
+ "epoch": 2.3830636222567056,
3016
+ "grad_norm": 0.11488287150859833,
3017
+ "learning_rate": 2.141666228426602e-05,
3018
+ "loss": 0.6842,
3019
+ "step": 10750
3020
+ },
3021
+ {
3022
+ "epoch": 2.3886056306805585,
3023
+ "grad_norm": 0.10014986246824265,
3024
+ "learning_rate": 2.104805480195128e-05,
3025
+ "loss": 0.602,
3026
+ "step": 10775
3027
+ },
3028
+ {
3029
+ "epoch": 2.3941476391044114,
3030
+ "grad_norm": 0.11838942021131516,
3031
+ "learning_rate": 2.0682273568047806e-05,
3032
+ "loss": 0.678,
3033
+ "step": 10800
3034
+ },
3035
+ {
3036
+ "epoch": 2.3996896475282643,
3037
+ "grad_norm": 0.09891260415315628,
3038
+ "learning_rate": 2.031933167645248e-05,
3039
+ "loss": 0.6204,
3040
+ "step": 10825
3041
+ },
3042
+ {
3043
+ "epoch": 2.405231655952117,
3044
+ "grad_norm": 0.11890105158090591,
3045
+ "learning_rate": 1.9959242119422062e-05,
3046
+ "loss": 0.6933,
3047
+ "step": 10850
3048
+ },
3049
+ {
3050
+ "epoch": 2.41077366437597,
3051
+ "grad_norm": 0.10138825327157974,
3052
+ "learning_rate": 1.9602017787108073e-05,
3053
+ "loss": 0.5633,
3054
+ "step": 10875
3055
+ },
3056
+ {
3057
+ "epoch": 2.4163156727998225,
3058
+ "grad_norm": 0.12140816450119019,
3059
+ "learning_rate": 1.9247671467095464e-05,
3060
+ "loss": 0.6962,
3061
+ "step": 10900
3062
+ },
3063
+ {
3064
+ "epoch": 2.4218576812236754,
3065
+ "grad_norm": 0.09580956399440765,
3066
+ "learning_rate": 1.8896215843944687e-05,
3067
+ "loss": 0.5756,
3068
+ "step": 10925
3069
+ },
3070
+ {
3071
+ "epoch": 2.4273996896475283,
3072
+ "grad_norm": 0.11642120033502579,
3073
+ "learning_rate": 1.8547663498737744e-05,
3074
+ "loss": 0.682,
3075
+ "step": 10950
3076
+ },
3077
+ {
3078
+ "epoch": 2.432941698071381,
3079
+ "grad_norm": 0.09648188948631287,
3080
+ "learning_rate": 1.8202026908627757e-05,
3081
+ "loss": 0.6105,
3082
+ "step": 10975
3083
+ },
3084
+ {
3085
+ "epoch": 2.438483706495234,
3086
+ "grad_norm": 0.11960633844137192,
3087
+ "learning_rate": 1.7859318446392438e-05,
3088
+ "loss": 0.6856,
3089
+ "step": 11000
3090
+ },
3091
+ {
3092
+ "epoch": 2.4440257149190865,
3093
+ "grad_norm": 0.10111569613218307,
3094
+ "learning_rate": 1.7519550379991022e-05,
3095
+ "loss": 0.5844,
3096
+ "step": 11025
3097
+ },
3098
+ {
3099
+ "epoch": 2.4495677233429394,
3100
+ "grad_norm": 0.11525440216064453,
3101
+ "learning_rate": 1.7182734872125194e-05,
3102
+ "loss": 0.6851,
3103
+ "step": 11050
3104
+ },
3105
+ {
3106
+ "epoch": 2.4551097317667923,
3107
+ "grad_norm": 0.10440149903297424,
3108
+ "learning_rate": 1.68488839798037e-05,
3109
+ "loss": 0.5659,
3110
+ "step": 11075
3111
+ },
3112
+ {
3113
+ "epoch": 2.460651740190645,
3114
+ "grad_norm": 0.11208430677652359,
3115
+ "learning_rate": 1.651800965391076e-05,
3116
+ "loss": 0.6895,
3117
+ "step": 11100
3118
+ },
3119
+ {
3120
+ "epoch": 2.466193748614498,
3121
+ "grad_norm": 0.09721311926841736,
3122
+ "learning_rate": 1.6190123738778195e-05,
3123
+ "loss": 0.5975,
3124
+ "step": 11125
3125
+ },
3126
+ {
3127
+ "epoch": 2.471735757038351,
3128
+ "grad_norm": 0.12069053202867508,
3129
+ "learning_rate": 1.5865237971761483e-05,
3130
+ "loss": 0.6999,
3131
+ "step": 11150
3132
+ },
3133
+ {
3134
+ "epoch": 2.4772777654622034,
3135
+ "grad_norm": 0.09901247173547745,
3136
+ "learning_rate": 1.5543363982819593e-05,
3137
+ "loss": 0.5735,
3138
+ "step": 11175
3139
+ },
3140
+ {
3141
+ "epoch": 2.4828197738860562,
3142
+ "grad_norm": 0.12278547137975693,
3143
+ "learning_rate": 1.522451329409863e-05,
3144
+ "loss": 0.6817,
3145
+ "step": 11200
3146
+ },
3147
+ {
3148
+ "epoch": 2.488361782309909,
3149
+ "grad_norm": 0.09708017110824585,
3150
+ "learning_rate": 1.4908697319519471e-05,
3151
+ "loss": 0.587,
3152
+ "step": 11225
3153
+ },
3154
+ {
3155
+ "epoch": 2.493903790733762,
3156
+ "grad_norm": 0.11575840413570404,
3157
+ "learning_rate": 1.4595927364369067e-05,
3158
+ "loss": 0.6873,
3159
+ "step": 11250
3160
+ },
3161
+ {
3162
+ "epoch": 2.499445799157615,
3163
+ "grad_norm": 0.09546317160129547,
3164
+ "learning_rate": 1.428621462489579e-05,
3165
+ "loss": 0.5687,
3166
+ "step": 11275
3167
+ },
3168
+ {
3169
+ "epoch": 2.5049878075814673,
3170
+ "grad_norm": 0.1179046705365181,
3171
+ "learning_rate": 1.3979570187908642e-05,
3172
+ "loss": 0.6869,
3173
+ "step": 11300
3174
+ },
3175
+ {
3176
+ "epoch": 2.51052981600532,
3177
+ "grad_norm": 0.09939169883728027,
3178
+ "learning_rate": 1.3676005030380434e-05,
3179
+ "loss": 0.5748,
3180
+ "step": 11325
3181
+ },
3182
+ {
3183
+ "epoch": 2.516071824429173,
3184
+ "grad_norm": 0.11648693680763245,
3185
+ "learning_rate": 1.3375530019054727e-05,
3186
+ "loss": 0.6955,
3187
+ "step": 11350
3188
+ },
3189
+ {
3190
+ "epoch": 2.521613832853026,
3191
+ "grad_norm": 0.09718350321054459,
3192
+ "learning_rate": 1.3078155910056933e-05,
3193
+ "loss": 0.6082,
3194
+ "step": 11375
3195
+ },
3196
+ {
3197
+ "epoch": 2.527155841276879,
3198
+ "grad_norm": 0.1167871356010437,
3199
+ "learning_rate": 1.2783893348509201e-05,
3200
+ "loss": 0.6818,
3201
+ "step": 11400
3202
+ },
3203
+ {
3204
+ "epoch": 2.5326978497007318,
3205
+ "grad_norm": 0.09479407966136932,
3206
+ "learning_rate": 1.2492752868149493e-05,
3207
+ "loss": 0.5897,
3208
+ "step": 11425
3209
+ },
3210
+ {
3211
+ "epoch": 2.538239858124584,
3212
+ "grad_norm": 0.11183246970176697,
3213
+ "learning_rate": 1.2204744890954312e-05,
3214
+ "loss": 0.677,
3215
+ "step": 11450
3216
+ },
3217
+ {
3218
+ "epoch": 2.543781866548437,
3219
+ "grad_norm": 0.0982903316617012,
3220
+ "learning_rate": 1.1919879726765704e-05,
3221
+ "loss": 0.6022,
3222
+ "step": 11475
3223
+ },
3224
+ {
3225
+ "epoch": 2.54932387497229,
3226
+ "grad_norm": 0.1157740131020546,
3227
+ "learning_rate": 1.1638167572922331e-05,
3228
+ "loss": 0.6883,
3229
+ "step": 11500
3230
+ },
3231
+ {
3232
+ "epoch": 2.554865883396143,
3233
+ "grad_norm": 0.0996859073638916,
3234
+ "learning_rate": 1.135961851389421e-05,
3235
+ "loss": 0.5939,
3236
+ "step": 11525
3237
+ },
3238
+ {
3239
+ "epoch": 2.5604078918199957,
3240
+ "grad_norm": 0.11946888267993927,
3241
+ "learning_rate": 1.1084242520921884e-05,
3242
+ "loss": 0.6865,
3243
+ "step": 11550
3244
+ },
3245
+ {
3246
+ "epoch": 2.565949900243848,
3247
+ "grad_norm": 0.10295979678630829,
3248
+ "learning_rate": 1.0812049451659367e-05,
3249
+ "loss": 0.5887,
3250
+ "step": 11575
3251
+ },
3252
+ {
3253
+ "epoch": 2.571491908667701,
3254
+ "grad_norm": 0.11419833451509476,
3255
+ "learning_rate": 1.0543049049821429e-05,
3256
+ "loss": 0.6946,
3257
+ "step": 11600
3258
+ },
3259
+ {
3260
+ "epoch": 2.577033917091554,
3261
+ "grad_norm": 0.09568798542022705,
3262
+ "learning_rate": 1.0277250944834594e-05,
3263
+ "loss": 0.5311,
3264
+ "step": 11625
3265
+ },
3266
+ {
3267
+ "epoch": 2.582575925515407,
3268
+ "grad_norm": 0.11659660935401917,
3269
+ "learning_rate": 1.0014664651492588e-05,
3270
+ "loss": 0.6942,
3271
+ "step": 11650
3272
+ },
3273
+ {
3274
+ "epoch": 2.5881179339392597,
3275
+ "grad_norm": 0.09711961448192596,
3276
+ "learning_rate": 9.755299569615661e-06,
3277
+ "loss": 0.58,
3278
+ "step": 11675
3279
+ },
3280
+ {
3281
+ "epoch": 2.5936599423631126,
3282
+ "grad_norm": 0.12282078713178635,
3283
+ "learning_rate": 9.49916498371416e-06,
3284
+ "loss": 0.6887,
3285
+ "step": 11700
3286
+ },
3287
+ {
3288
+ "epoch": 2.599201950786965,
3289
+ "grad_norm": 0.09802839159965515,
3290
+ "learning_rate": 9.246270062656104e-06,
3291
+ "loss": 0.5524,
3292
+ "step": 11725
3293
+ },
3294
+ {
3295
+ "epoch": 2.604743959210818,
3296
+ "grad_norm": 0.11624480038881302,
3297
+ "learning_rate": 8.996623859339026e-06,
3298
+ "loss": 0.6953,
3299
+ "step": 11750
3300
+ },
3301
+ {
3302
+ "epoch": 2.610285967634671,
3303
+ "grad_norm": 0.09748586267232895,
3304
+ "learning_rate": 8.75023531036584e-06,
3305
+ "loss": 0.5844,
3306
+ "step": 11775
3307
+ },
3308
+ {
3309
+ "epoch": 2.6158279760585237,
3310
+ "grad_norm": 0.12042013555765152,
3311
+ "learning_rate": 8.50711323572504e-06,
3312
+ "loss": 0.7066,
3313
+ "step": 11800
3314
+ },
3315
+ {
3316
+ "epoch": 2.6213699844823766,
3317
+ "grad_norm": 0.09437406063079834,
3318
+ "learning_rate": 8.267266338474833e-06,
3319
+ "loss": 0.5993,
3320
+ "step": 11825
3321
+ },
3322
+ {
3323
+ "epoch": 2.626911992906229,
3324
+ "grad_norm": 0.12176880985498428,
3325
+ "learning_rate": 8.030703204431711e-06,
3326
+ "loss": 0.6901,
3327
+ "step": 11850
3328
+ },
3329
+ {
3330
+ "epoch": 2.632454001330082,
3331
+ "grad_norm": 0.09464902430772781,
3332
+ "learning_rate": 7.797432301863029e-06,
3333
+ "loss": 0.5777,
3334
+ "step": 11875
3335
+ },
3336
+ {
3337
+ "epoch": 2.637996009753935,
3338
+ "grad_norm": 0.1198597401380539,
3339
+ "learning_rate": 7.567461981183954e-06,
3340
+ "loss": 0.6915,
3341
+ "step": 11900
3342
+ },
3343
+ {
3344
+ "epoch": 2.6435380181777877,
3345
+ "grad_norm": 0.09965650737285614,
3346
+ "learning_rate": 7.340800474658438e-06,
3347
+ "loss": 0.6047,
3348
+ "step": 11925
3349
+ },
3350
+ {
3351
+ "epoch": 2.6490800266016405,
3352
+ "grad_norm": 0.12205997854471207,
3353
+ "learning_rate": 7.117455896104586e-06,
3354
+ "loss": 0.6888,
3355
+ "step": 11950
3356
+ },
3357
+ {
3358
+ "epoch": 2.6546220350254934,
3359
+ "grad_norm": 0.10341254621744156,
3360
+ "learning_rate": 6.897436240604182e-06,
3361
+ "loss": 0.5954,
3362
+ "step": 11975
3363
+ },
3364
+ {
3365
+ "epoch": 2.6601640434493463,
3366
+ "grad_norm": 0.11666877567768097,
3367
+ "learning_rate": 6.680749384216556e-06,
3368
+ "loss": 0.6775,
3369
+ "step": 12000
3370
+ },
3371
+ {
3372
+ "epoch": 2.6657060518731988,
3373
+ "grad_norm": 0.09655077755451202,
3374
+ "learning_rate": 6.4674030836965435e-06,
3375
+ "loss": 0.5951,
3376
+ "step": 12025
3377
+ },
3378
+ {
3379
+ "epoch": 2.6712480602970516,
3380
+ "grad_norm": 0.11688444018363953,
3381
+ "learning_rate": 6.257404976216863e-06,
3382
+ "loss": 0.6762,
3383
+ "step": 12050
3384
+ },
3385
+ {
3386
+ "epoch": 2.6767900687209045,
3387
+ "grad_norm": 0.09534773975610733,
3388
+ "learning_rate": 6.050762579094727e-06,
3389
+ "loss": 0.6063,
3390
+ "step": 12075
3391
+ },
3392
+ {
3393
+ "epoch": 2.6823320771447574,
3394
+ "grad_norm": 0.11826001852750778,
3395
+ "learning_rate": 5.847483289522804e-06,
3396
+ "loss": 0.6893,
3397
+ "step": 12100
3398
+ },
3399
+ {
3400
+ "epoch": 2.68787408556861,
3401
+ "grad_norm": 0.09770546853542328,
3402
+ "learning_rate": 5.647574384304288e-06,
3403
+ "loss": 0.5977,
3404
+ "step": 12125
3405
+ },
3406
+ {
3407
+ "epoch": 2.6934160939924627,
3408
+ "grad_norm": 0.11583051085472107,
3409
+ "learning_rate": 5.451043019592506e-06,
3410
+ "loss": 0.6892,
3411
+ "step": 12150
3412
+ },
3413
+ {
3414
+ "epoch": 2.6989581024163156,
3415
+ "grad_norm": 0.09545658528804779,
3416
+ "learning_rate": 5.257896230634729e-06,
3417
+ "loss": 0.5911,
3418
+ "step": 12175
3419
+ },
3420
+ {
3421
+ "epoch": 2.7045001108401685,
3422
+ "grad_norm": 0.11835386604070663,
3423
+ "learning_rate": 5.06814093152036e-06,
3424
+ "loss": 0.6813,
3425
+ "step": 12200
3426
+ },
3427
+ {
3428
+ "epoch": 2.7100421192640214,
3429
+ "grad_norm": 0.09876852482557297,
3430
+ "learning_rate": 4.881783914933347e-06,
3431
+ "loss": 0.5886,
3432
+ "step": 12225
3433
+ },
3434
+ {
3435
+ "epoch": 2.7155841276878743,
3436
+ "grad_norm": 0.11607835441827774,
3437
+ "learning_rate": 4.698831851909113e-06,
3438
+ "loss": 0.6915,
3439
+ "step": 12250
3440
+ },
3441
+ {
3442
+ "epoch": 2.721126136111727,
3443
+ "grad_norm": 0.09856165945529938,
3444
+ "learning_rate": 4.519291291595673e-06,
3445
+ "loss": 0.588,
3446
+ "step": 12275
3447
+ },
3448
+ {
3449
+ "epoch": 2.7266681445355796,
3450
+ "grad_norm": 0.1222713515162468,
3451
+ "learning_rate": 4.343168661019304e-06,
3452
+ "loss": 0.7022,
3453
+ "step": 12300
3454
+ },
3455
+ {
3456
+ "epoch": 2.7322101529594325,
3457
+ "grad_norm": 0.09814934432506561,
3458
+ "learning_rate": 4.170470264854354e-06,
3459
+ "loss": 0.5736,
3460
+ "step": 12325
3461
+ },
3462
+ {
3463
+ "epoch": 2.7377521613832854,
3464
+ "grad_norm": 0.12004557251930237,
3465
+ "learning_rate": 4.001202285197614e-06,
3466
+ "loss": 0.6877,
3467
+ "step": 12350
3468
+ },
3469
+ {
3470
+ "epoch": 2.7432941698071382,
3471
+ "grad_norm": 0.09881303459405899,
3472
+ "learning_rate": 3.8353707813470255e-06,
3473
+ "loss": 0.5771,
3474
+ "step": 12375
3475
+ },
3476
+ {
3477
+ "epoch": 2.7488361782309907,
3478
+ "grad_norm": 0.11910282075405121,
3479
+ "learning_rate": 3.6729816895847646e-06,
3480
+ "loss": 0.6911,
3481
+ "step": 12400
3482
+ },
3483
+ {
3484
+ "epoch": 2.7543781866548436,
3485
+ "grad_norm": 0.09742295742034912,
3486
+ "learning_rate": 3.514040822964715e-06,
3487
+ "loss": 0.6007,
3488
+ "step": 12425
3489
+ },
3490
+ {
3491
+ "epoch": 2.7599201950786965,
3492
+ "grad_norm": 0.12169786542654037,
3493
+ "learning_rate": 3.3585538711044197e-06,
3494
+ "loss": 0.6907,
3495
+ "step": 12450
3496
+ },
3497
+ {
3498
+ "epoch": 2.7654622035025493,
3499
+ "grad_norm": 0.09972374886274338,
3500
+ "learning_rate": 3.206526399981358e-06,
3501
+ "loss": 0.5764,
3502
+ "step": 12475
3503
+ },
3504
+ {
3505
+ "epoch": 2.7710042119264022,
3506
+ "grad_norm": 0.11850964277982712,
3507
+ "learning_rate": 3.057963851733803e-06,
3508
+ "loss": 0.6742,
3509
+ "step": 12500
3510
+ },
3511
+ {
3512
+ "epoch": 2.776546220350255,
3513
+ "grad_norm": 0.10190987586975098,
3514
+ "learning_rate": 2.912871544465834e-06,
3515
+ "loss": 0.5769,
3516
+ "step": 12525
3517
+ },
3518
+ {
3519
+ "epoch": 2.782088228774108,
3520
+ "grad_norm": 0.11625930666923523,
3521
+ "learning_rate": 2.771254672057144e-06,
3522
+ "loss": 0.6827,
3523
+ "step": 12550
3524
+ },
3525
+ {
3526
+ "epoch": 2.7876302371979604,
3527
+ "grad_norm": 0.10201703011989594,
3528
+ "learning_rate": 2.6331183039769892e-06,
3529
+ "loss": 0.5792,
3530
+ "step": 12575
3531
+ },
3532
+ {
3533
+ "epoch": 2.7931722456218133,
3534
+ "grad_norm": 0.1190221905708313,
3535
+ "learning_rate": 2.4984673851028095e-06,
3536
+ "loss": 0.6815,
3537
+ "step": 12600
3538
+ },
3539
+ {
3540
+ "epoch": 2.798714254045666,
3541
+ "grad_norm": 0.1018979623913765,
3542
+ "learning_rate": 2.367306735543151e-06,
3543
+ "loss": 0.5642,
3544
+ "step": 12625
3545
+ },
3546
+ {
3547
+ "epoch": 2.804256262469519,
3548
+ "grad_norm": 0.11812682449817657,
3549
+ "learning_rate": 2.2396410504651246e-06,
3550
+ "loss": 0.684,
3551
+ "step": 12650
3552
+ },
3553
+ {
3554
+ "epoch": 2.8097982708933715,
3555
+ "grad_norm": 0.09736626595258713,
3556
+ "learning_rate": 2.115474899926351e-06,
3557
+ "loss": 0.5792,
3558
+ "step": 12675
3559
+ },
3560
+ {
3561
+ "epoch": 2.8153402793172244,
3562
+ "grad_norm": 0.11478708684444427,
3563
+ "learning_rate": 1.994812728711404e-06,
3564
+ "loss": 0.6799,
3565
+ "step": 12700
3566
+ },
3567
+ {
3568
+ "epoch": 2.8208822877410773,
3569
+ "grad_norm": 0.09593895822763443,
3570
+ "learning_rate": 1.8822775987230811e-06,
3571
+ "loss": 0.583,
3572
+ "step": 12725
3573
+ },
3574
+ {
3575
+ "epoch": 2.82642429616493,
3576
+ "grad_norm": 0.12742695212364197,
3577
+ "learning_rate": 1.7684956399847686e-06,
3578
+ "loss": 0.6926,
3579
+ "step": 12750
3580
+ },
3581
+ {
3582
+ "epoch": 2.831966304588783,
3583
+ "grad_norm": 0.09630496799945831,
3584
+ "learning_rate": 1.6582300814115714e-06,
3585
+ "loss": 0.5821,
3586
+ "step": 12775
3587
+ },
3588
+ {
3589
+ "epoch": 2.837508313012636,
3590
+ "grad_norm": 0.11476560682058334,
3591
+ "learning_rate": 1.5514848701875473e-06,
3592
+ "loss": 0.6813,
3593
+ "step": 12800
3594
+ },
3595
+ {
3596
+ "epoch": 2.843050321436489,
3597
+ "grad_norm": 0.09966927021741867,
3598
+ "learning_rate": 1.4482638274786354e-06,
3599
+ "loss": 0.581,
3600
+ "step": 12825
3601
+ },
3602
+ {
3603
+ "epoch": 2.8485923298603413,
3604
+ "grad_norm": 0.12046293914318085,
3605
+ "learning_rate": 1.348570648295866e-06,
3606
+ "loss": 0.6785,
3607
+ "step": 12850
3608
+ },
3609
+ {
3610
+ "epoch": 2.854134338284194,
3611
+ "grad_norm": 0.09309112280607224,
3612
+ "learning_rate": 1.252408901363078e-06,
3613
+ "loss": 0.5868,
3614
+ "step": 12875
3615
+ },
3616
+ {
3617
+ "epoch": 2.859676346708047,
3618
+ "grad_norm": 0.13479410111904144,
3619
+ "learning_rate": 1.1597820289891758e-06,
3620
+ "loss": 0.6804,
3621
+ "step": 12900
3622
+ },
3623
+ {
3624
+ "epoch": 2.8652183551319,
3625
+ "grad_norm": 0.10119752585887909,
3626
+ "learning_rate": 1.0706933469449288e-06,
3627
+ "loss": 0.6251,
3628
+ "step": 12925
3629
+ },
3630
+ {
3631
+ "epoch": 2.8707603635557524,
3632
+ "grad_norm": 0.12458885461091995,
3633
+ "learning_rate": 9.851460443442317e-07,
3634
+ "loss": 0.6895,
3635
+ "step": 12950
3636
+ },
3637
+ {
3638
+ "epoch": 2.8763023719796053,
3639
+ "grad_norm": 0.09733375906944275,
3640
+ "learning_rate": 9.031431835299753e-07,
3641
+ "loss": 0.5619,
3642
+ "step": 12975
3643
+ },
3644
+ {
3645
+ "epoch": 2.881844380403458,
3646
+ "grad_norm": 0.12155549973249435,
3647
+ "learning_rate": 8.246876999644104e-07,
3648
+ "loss": 0.6861,
3649
+ "step": 13000
3650
+ }
3651
+ ],
3652
+ "logging_steps": 25,
3653
+ "max_steps": 13533,
3654
+ "num_input_tokens_seen": 0,
3655
+ "num_train_epochs": 3,
3656
+ "save_steps": 200,
3657
+ "stateful_callbacks": {
3658
+ "TrainerControl": {
3659
+ "args": {
3660
+ "should_epoch_stop": false,
3661
+ "should_evaluate": false,
3662
+ "should_log": false,
3663
+ "should_save": true,
3664
+ "should_training_stop": false
3665
+ },
3666
+ "attributes": {}
3667
+ }
3668
+ },
3669
+ "total_flos": 6.239618339309722e+18,
3670
+ "train_batch_size": 2,
3671
+ "trial_name": null,
3672
+ "trial_params": null
3673
+ }
checkpoint-13000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a13672a0401c8d0a53efc112f53ebd65f36fa9003e87b1710879c58d881a1e1
3
+ size 5051
checkpoint-5000/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: meta-llama/Llama-2-13b-chat-hf
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.12.0
checkpoint-5000/adapter_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf",
5
+ "bias": "lora_only",
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": 0.001,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 64,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "q_proj",
24
+ "v_proj"
25
+ ],
26
+ "task_type": "CAUSAL_LM",
27
+ "use_dora": false,
28
+ "use_rslora": false
29
+ }
checkpoint-5000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d180b331707bf9e081c1c962dde68c522775724e3d268142d0146a7effc62426
3
+ size 209736952
checkpoint-5000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e09ec34632c136b2763ae88c46570b2ac61d74bbf38e56b10bf22089ab85e48b
3
+ size 419529285
checkpoint-5000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1bb037a5c61b129a7d4af16a4c1484215dc381b39d1964c770ac2c0e53d0f462
3
+ size 14575
checkpoint-5000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99bb8a615f620c4d1745253ce4c9bf1a0da431ee38b91bdc2f185c8450df5bcc
3
+ size 627
checkpoint-5000/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-5000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5000/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-5000/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<<SYS>>\\n' + system_message + '\\n<</SYS>>\\n\\n' + message['content'] %}{% else %}{% set content = message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ bos_token + '[INST] ' + content.strip() + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ ' ' + content.strip() + ' ' + eos_token }}{% endif %}{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": false,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
checkpoint-5000/trainer_state.json ADDED
@@ -0,0 +1,1433 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.108401684770561,
5
+ "eval_steps": 500,
6
+ "global_step": 5000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.005542008423852805,
13
+ "grad_norm": 0.07243233174085617,
14
+ "learning_rate": 1.2315270935960592e-05,
15
+ "loss": 1.4594,
16
+ "step": 25
17
+ },
18
+ {
19
+ "epoch": 0.01108401684770561,
20
+ "grad_norm": 0.40484485030174255,
21
+ "learning_rate": 2.4630541871921184e-05,
22
+ "loss": 2.2032,
23
+ "step": 50
24
+ },
25
+ {
26
+ "epoch": 0.016626025271558414,
27
+ "grad_norm": 0.06850667297840118,
28
+ "learning_rate": 3.694581280788178e-05,
29
+ "loss": 1.2931,
30
+ "step": 75
31
+ },
32
+ {
33
+ "epoch": 0.02216803369541122,
34
+ "grad_norm": 0.4395073354244232,
35
+ "learning_rate": 4.926108374384237e-05,
36
+ "loss": 1.5698,
37
+ "step": 100
38
+ },
39
+ {
40
+ "epoch": 0.027710042119264023,
41
+ "grad_norm": 0.077068030834198,
42
+ "learning_rate": 6.157635467980296e-05,
43
+ "loss": 1.0537,
44
+ "step": 125
45
+ },
46
+ {
47
+ "epoch": 0.03325205054311683,
48
+ "grad_norm": 0.3282291293144226,
49
+ "learning_rate": 7.389162561576355e-05,
50
+ "loss": 0.9749,
51
+ "step": 150
52
+ },
53
+ {
54
+ "epoch": 0.03879405896696963,
55
+ "grad_norm": 0.0593000203371048,
56
+ "learning_rate": 8.620689655172413e-05,
57
+ "loss": 0.9349,
58
+ "step": 175
59
+ },
60
+ {
61
+ "epoch": 0.04433606739082244,
62
+ "grad_norm": 0.25612473487854004,
63
+ "learning_rate": 9.852216748768474e-05,
64
+ "loss": 0.8974,
65
+ "step": 200
66
+ },
67
+ {
68
+ "epoch": 0.04987807581467524,
69
+ "grad_norm": 0.0757347121834755,
70
+ "learning_rate": 0.00011083743842364534,
71
+ "loss": 0.9081,
72
+ "step": 225
73
+ },
74
+ {
75
+ "epoch": 0.055420084238528046,
76
+ "grad_norm": 0.14145499467849731,
77
+ "learning_rate": 0.00012315270935960593,
78
+ "loss": 0.8607,
79
+ "step": 250
80
+ },
81
+ {
82
+ "epoch": 0.06096209266238085,
83
+ "grad_norm": 0.07710155844688416,
84
+ "learning_rate": 0.00013546798029556652,
85
+ "loss": 0.8973,
86
+ "step": 275
87
+ },
88
+ {
89
+ "epoch": 0.06650410108623366,
90
+ "grad_norm": 0.14791467785835266,
91
+ "learning_rate": 0.0001477832512315271,
92
+ "loss": 0.7924,
93
+ "step": 300
94
+ },
95
+ {
96
+ "epoch": 0.07204610951008646,
97
+ "grad_norm": 0.07742594182491302,
98
+ "learning_rate": 0.00016009852216748767,
99
+ "loss": 0.8698,
100
+ "step": 325
101
+ },
102
+ {
103
+ "epoch": 0.07758811793393926,
104
+ "grad_norm": 0.14303487539291382,
105
+ "learning_rate": 0.00017241379310344826,
106
+ "loss": 0.786,
107
+ "step": 350
108
+ },
109
+ {
110
+ "epoch": 0.08313012635779206,
111
+ "grad_norm": 0.0865108072757721,
112
+ "learning_rate": 0.00018472906403940888,
113
+ "loss": 0.8606,
114
+ "step": 375
115
+ },
116
+ {
117
+ "epoch": 0.08867213478164487,
118
+ "grad_norm": 0.7533164024353027,
119
+ "learning_rate": 0.00019704433497536947,
120
+ "loss": 0.807,
121
+ "step": 400
122
+ },
123
+ {
124
+ "epoch": 0.09421414320549767,
125
+ "grad_norm": 0.08325570821762085,
126
+ "learning_rate": 0.00019999896617927833,
127
+ "loss": 0.8635,
128
+ "step": 425
129
+ },
130
+ {
131
+ "epoch": 0.09975615162935048,
132
+ "grad_norm": 0.1043543666601181,
133
+ "learning_rate": 0.0001999944557842899,
134
+ "loss": 0.7825,
135
+ "step": 450
136
+ },
137
+ {
138
+ "epoch": 0.10529816005320328,
139
+ "grad_norm": 0.07949995994567871,
140
+ "learning_rate": 0.0001999863658806385,
141
+ "loss": 0.8379,
142
+ "step": 475
143
+ },
144
+ {
145
+ "epoch": 0.11084016847705609,
146
+ "grad_norm": 0.12020070850849152,
147
+ "learning_rate": 0.00019997469675791905,
148
+ "loss": 0.768,
149
+ "step": 500
150
+ },
151
+ {
152
+ "epoch": 0.11638217690090889,
153
+ "grad_norm": 0.0803595781326294,
154
+ "learning_rate": 0.00019995944883385196,
155
+ "loss": 0.8487,
156
+ "step": 525
157
+ },
158
+ {
159
+ "epoch": 0.1219241853247617,
160
+ "grad_norm": 0.11509452760219574,
161
+ "learning_rate": 0.0001999406226542682,
162
+ "loss": 0.7787,
163
+ "step": 550
164
+ },
165
+ {
166
+ "epoch": 0.1274661937486145,
167
+ "grad_norm": 0.07928384840488434,
168
+ "learning_rate": 0.00019991821889308987,
169
+ "loss": 0.8357,
170
+ "step": 575
171
+ },
172
+ {
173
+ "epoch": 0.1330082021724673,
174
+ "grad_norm": 0.09423446655273438,
175
+ "learning_rate": 0.00019989223835230606,
176
+ "loss": 0.7564,
177
+ "step": 600
178
+ },
179
+ {
180
+ "epoch": 0.1385502105963201,
181
+ "grad_norm": 0.0835939422249794,
182
+ "learning_rate": 0.000199862681961944,
183
+ "loss": 0.8568,
184
+ "step": 625
185
+ },
186
+ {
187
+ "epoch": 0.1440922190201729,
188
+ "grad_norm": 0.09292898327112198,
189
+ "learning_rate": 0.0001998295507800359,
190
+ "loss": 0.7612,
191
+ "step": 650
192
+ },
193
+ {
194
+ "epoch": 0.1496342274440257,
195
+ "grad_norm": 0.07704215496778488,
196
+ "learning_rate": 0.00019979284599258107,
197
+ "loss": 0.8263,
198
+ "step": 675
199
+ },
200
+ {
201
+ "epoch": 0.15517623586787851,
202
+ "grad_norm": 0.10980474948883057,
203
+ "learning_rate": 0.0001997525689135034,
204
+ "loss": 0.7677,
205
+ "step": 700
206
+ },
207
+ {
208
+ "epoch": 0.16071824429173132,
209
+ "grad_norm": 0.08016064018011093,
210
+ "learning_rate": 0.0001997087209846043,
211
+ "loss": 0.8344,
212
+ "step": 725
213
+ },
214
+ {
215
+ "epoch": 0.16626025271558412,
216
+ "grad_norm": 0.0950881615281105,
217
+ "learning_rate": 0.0001996613037755113,
218
+ "loss": 0.769,
219
+ "step": 750
220
+ },
221
+ {
222
+ "epoch": 0.17180226113943692,
223
+ "grad_norm": 0.07932984828948975,
224
+ "learning_rate": 0.00019961031898362152,
225
+ "loss": 0.8156,
226
+ "step": 775
227
+ },
228
+ {
229
+ "epoch": 0.17734426956328975,
230
+ "grad_norm": 0.09336528927087784,
231
+ "learning_rate": 0.00019955576843404128,
232
+ "loss": 0.7767,
233
+ "step": 800
234
+ },
235
+ {
236
+ "epoch": 0.18288627798714255,
237
+ "grad_norm": 0.08560346812009811,
238
+ "learning_rate": 0.00019949765407952042,
239
+ "loss": 0.8228,
240
+ "step": 825
241
+ },
242
+ {
243
+ "epoch": 0.18842828641099535,
244
+ "grad_norm": 0.08475169539451599,
245
+ "learning_rate": 0.00019943597800038267,
246
+ "loss": 0.7669,
247
+ "step": 850
248
+ },
249
+ {
250
+ "epoch": 0.19397029483484815,
251
+ "grad_norm": 0.09038034081459045,
252
+ "learning_rate": 0.00019937074240445105,
253
+ "loss": 0.8182,
254
+ "step": 875
255
+ },
256
+ {
257
+ "epoch": 0.19951230325870095,
258
+ "grad_norm": 0.09195873886346817,
259
+ "learning_rate": 0.0001993019496269688,
260
+ "loss": 0.7598,
261
+ "step": 900
262
+ },
263
+ {
264
+ "epoch": 0.20505431168255375,
265
+ "grad_norm": 0.08655796200037003,
266
+ "learning_rate": 0.0001992296021305159,
267
+ "loss": 0.8167,
268
+ "step": 925
269
+ },
270
+ {
271
+ "epoch": 0.21059632010640655,
272
+ "grad_norm": 0.08353498578071594,
273
+ "learning_rate": 0.00019915370250492084,
274
+ "loss": 0.7486,
275
+ "step": 950
276
+ },
277
+ {
278
+ "epoch": 0.21613832853025935,
279
+ "grad_norm": 0.09225723147392273,
280
+ "learning_rate": 0.0001990742534671679,
281
+ "loss": 0.8138,
282
+ "step": 975
283
+ },
284
+ {
285
+ "epoch": 0.22168033695411218,
286
+ "grad_norm": 0.12104763090610504,
287
+ "learning_rate": 0.00019899125786129997,
288
+ "loss": 0.7153,
289
+ "step": 1000
290
+ },
291
+ {
292
+ "epoch": 0.22722234537796498,
293
+ "grad_norm": 0.0815986767411232,
294
+ "learning_rate": 0.00019890471865831669,
295
+ "loss": 0.7983,
296
+ "step": 1025
297
+ },
298
+ {
299
+ "epoch": 0.23276435380181779,
300
+ "grad_norm": 0.08845670521259308,
301
+ "learning_rate": 0.00019881463895606805,
302
+ "loss": 0.7187,
303
+ "step": 1050
304
+ },
305
+ {
306
+ "epoch": 0.2383063622256706,
307
+ "grad_norm": 0.0821809321641922,
308
+ "learning_rate": 0.00019872102197914359,
309
+ "loss": 0.804,
310
+ "step": 1075
311
+ },
312
+ {
313
+ "epoch": 0.2438483706495234,
314
+ "grad_norm": 0.08711609989404678,
315
+ "learning_rate": 0.00019862387107875688,
316
+ "loss": 0.7795,
317
+ "step": 1100
318
+ },
319
+ {
320
+ "epoch": 0.2493903790733762,
321
+ "grad_norm": 0.08517508953809738,
322
+ "learning_rate": 0.00019852318973262567,
323
+ "loss": 0.7937,
324
+ "step": 1125
325
+ },
326
+ {
327
+ "epoch": 0.254932387497229,
328
+ "grad_norm": 0.10830071568489075,
329
+ "learning_rate": 0.00019841898154484726,
330
+ "loss": 0.7458,
331
+ "step": 1150
332
+ },
333
+ {
334
+ "epoch": 0.2604743959210818,
335
+ "grad_norm": 0.08541836589574814,
336
+ "learning_rate": 0.0001983112502457696,
337
+ "loss": 0.8131,
338
+ "step": 1175
339
+ },
340
+ {
341
+ "epoch": 0.2660164043449346,
342
+ "grad_norm": 0.08794037252664566,
343
+ "learning_rate": 0.00019819999969185762,
344
+ "loss": 0.7577,
345
+ "step": 1200
346
+ },
347
+ {
348
+ "epoch": 0.2715584127687874,
349
+ "grad_norm": 0.08078176528215408,
350
+ "learning_rate": 0.00019808523386555542,
351
+ "loss": 0.812,
352
+ "step": 1225
353
+ },
354
+ {
355
+ "epoch": 0.2771004211926402,
356
+ "grad_norm": 0.09263130277395248,
357
+ "learning_rate": 0.0001979669568751434,
358
+ "loss": 0.7582,
359
+ "step": 1250
360
+ },
361
+ {
362
+ "epoch": 0.282642429616493,
363
+ "grad_norm": 0.08198932558298111,
364
+ "learning_rate": 0.00019784517295459147,
365
+ "loss": 0.7958,
366
+ "step": 1275
367
+ },
368
+ {
369
+ "epoch": 0.2881844380403458,
370
+ "grad_norm": 0.07858102023601532,
371
+ "learning_rate": 0.00019771988646340725,
372
+ "loss": 0.7744,
373
+ "step": 1300
374
+ },
375
+ {
376
+ "epoch": 0.2937264464641986,
377
+ "grad_norm": 0.0851408839225769,
378
+ "learning_rate": 0.00019759110188648026,
379
+ "loss": 0.7913,
380
+ "step": 1325
381
+ },
382
+ {
383
+ "epoch": 0.2992684548880514,
384
+ "grad_norm": 0.09252189099788666,
385
+ "learning_rate": 0.00019745882383392116,
386
+ "loss": 0.7675,
387
+ "step": 1350
388
+ },
389
+ {
390
+ "epoch": 0.30481046331190426,
391
+ "grad_norm": 0.08306555449962616,
392
+ "learning_rate": 0.0001973230570408968,
393
+ "loss": 0.8059,
394
+ "step": 1375
395
+ },
396
+ {
397
+ "epoch": 0.31035247173575703,
398
+ "grad_norm": 0.0797729641199112,
399
+ "learning_rate": 0.0001971838063674608,
400
+ "loss": 0.7424,
401
+ "step": 1400
402
+ },
403
+ {
404
+ "epoch": 0.31589448015960986,
405
+ "grad_norm": 0.08266165107488632,
406
+ "learning_rate": 0.0001970410767983794,
407
+ "loss": 0.7847,
408
+ "step": 1425
409
+ },
410
+ {
411
+ "epoch": 0.32143648858346263,
412
+ "grad_norm": 0.09364205598831177,
413
+ "learning_rate": 0.00019689487344295322,
414
+ "loss": 0.6924,
415
+ "step": 1450
416
+ },
417
+ {
418
+ "epoch": 0.32697849700731546,
419
+ "grad_norm": 0.08461842685937881,
420
+ "learning_rate": 0.00019674520153483414,
421
+ "loss": 0.8007,
422
+ "step": 1475
423
+ },
424
+ {
425
+ "epoch": 0.33252050543116823,
426
+ "grad_norm": 0.0840207040309906,
427
+ "learning_rate": 0.00019659206643183813,
428
+ "loss": 0.7139,
429
+ "step": 1500
430
+ },
431
+ {
432
+ "epoch": 0.33806251385502106,
433
+ "grad_norm": 0.08344192802906036,
434
+ "learning_rate": 0.00019643547361575343,
435
+ "loss": 0.7982,
436
+ "step": 1525
437
+ },
438
+ {
439
+ "epoch": 0.34360452227887384,
440
+ "grad_norm": 0.07934779673814774,
441
+ "learning_rate": 0.0001962754286921442,
442
+ "loss": 0.7164,
443
+ "step": 1550
444
+ },
445
+ {
446
+ "epoch": 0.34914653070272667,
447
+ "grad_norm": 0.08716201782226562,
448
+ "learning_rate": 0.00019611193739015,
449
+ "loss": 0.7846,
450
+ "step": 1575
451
+ },
452
+ {
453
+ "epoch": 0.3546885391265795,
454
+ "grad_norm": 0.08384064584970474,
455
+ "learning_rate": 0.0001959450055622806,
456
+ "loss": 0.7416,
457
+ "step": 1600
458
+ },
459
+ {
460
+ "epoch": 0.36023054755043227,
461
+ "grad_norm": 0.08661937713623047,
462
+ "learning_rate": 0.0001957746391842066,
463
+ "loss": 0.8075,
464
+ "step": 1625
465
+ },
466
+ {
467
+ "epoch": 0.3657725559742851,
468
+ "grad_norm": 0.09327207505702972,
469
+ "learning_rate": 0.00019560084435454536,
470
+ "loss": 0.7596,
471
+ "step": 1650
472
+ },
473
+ {
474
+ "epoch": 0.37131456439813787,
475
+ "grad_norm": 0.08391096442937851,
476
+ "learning_rate": 0.00019542362729464273,
477
+ "loss": 0.7794,
478
+ "step": 1675
479
+ },
480
+ {
481
+ "epoch": 0.3768565728219907,
482
+ "grad_norm": 0.07694080471992493,
483
+ "learning_rate": 0.00019524299434835052,
484
+ "loss": 0.7424,
485
+ "step": 1700
486
+ },
487
+ {
488
+ "epoch": 0.38239858124584347,
489
+ "grad_norm": 0.08567491173744202,
490
+ "learning_rate": 0.00019505895198179912,
491
+ "loss": 0.7996,
492
+ "step": 1725
493
+ },
494
+ {
495
+ "epoch": 0.3879405896696963,
496
+ "grad_norm": 0.08828684687614441,
497
+ "learning_rate": 0.0001948715067831663,
498
+ "loss": 0.7394,
499
+ "step": 1750
500
+ },
501
+ {
502
+ "epoch": 0.39348259809354913,
503
+ "grad_norm": 0.08347714692354202,
504
+ "learning_rate": 0.00019468066546244117,
505
+ "loss": 0.7734,
506
+ "step": 1775
507
+ },
508
+ {
509
+ "epoch": 0.3990246065174019,
510
+ "grad_norm": 0.07736373692750931,
511
+ "learning_rate": 0.00019448643485118412,
512
+ "loss": 0.7134,
513
+ "step": 1800
514
+ },
515
+ {
516
+ "epoch": 0.40456661494125473,
517
+ "grad_norm": 0.0840897262096405,
518
+ "learning_rate": 0.00019428882190228216,
519
+ "loss": 0.787,
520
+ "step": 1825
521
+ },
522
+ {
523
+ "epoch": 0.4101086233651075,
524
+ "grad_norm": 0.08665871620178223,
525
+ "learning_rate": 0.0001940878336897001,
526
+ "loss": 0.7151,
527
+ "step": 1850
528
+ },
529
+ {
530
+ "epoch": 0.41565063178896033,
531
+ "grad_norm": 0.08358912914991379,
532
+ "learning_rate": 0.0001938834774082274,
533
+ "loss": 0.7982,
534
+ "step": 1875
535
+ },
536
+ {
537
+ "epoch": 0.4211926402128131,
538
+ "grad_norm": 0.07928963005542755,
539
+ "learning_rate": 0.0001936757603732203,
540
+ "loss": 0.7195,
541
+ "step": 1900
542
+ },
543
+ {
544
+ "epoch": 0.42673464863666594,
545
+ "grad_norm": 0.08886470645666122,
546
+ "learning_rate": 0.00019346469002034042,
547
+ "loss": 0.7762,
548
+ "step": 1925
549
+ },
550
+ {
551
+ "epoch": 0.4322766570605187,
552
+ "grad_norm": 0.1071886494755745,
553
+ "learning_rate": 0.00019325027390528822,
554
+ "loss": 0.7453,
555
+ "step": 1950
556
+ },
557
+ {
558
+ "epoch": 0.43781866548437154,
559
+ "grad_norm": 0.08474262803792953,
560
+ "learning_rate": 0.00019303251970353261,
561
+ "loss": 0.7839,
562
+ "step": 1975
563
+ },
564
+ {
565
+ "epoch": 0.44336067390822437,
566
+ "grad_norm": 0.08803894370794296,
567
+ "learning_rate": 0.0001928114352100363,
568
+ "loss": 0.7171,
569
+ "step": 2000
570
+ },
571
+ {
572
+ "epoch": 0.44890268233207714,
573
+ "grad_norm": 0.08429575711488724,
574
+ "learning_rate": 0.00019258702833897665,
575
+ "loss": 0.7781,
576
+ "step": 2025
577
+ },
578
+ {
579
+ "epoch": 0.45444469075592997,
580
+ "grad_norm": 0.08510231226682663,
581
+ "learning_rate": 0.00019235930712346248,
582
+ "loss": 0.6949,
583
+ "step": 2050
584
+ },
585
+ {
586
+ "epoch": 0.45998669917978274,
587
+ "grad_norm": 0.08167176693677902,
588
+ "learning_rate": 0.00019212827971524634,
589
+ "loss": 0.7722,
590
+ "step": 2075
591
+ },
592
+ {
593
+ "epoch": 0.46552870760363557,
594
+ "grad_norm": 0.06542418897151947,
595
+ "learning_rate": 0.00019189395438443278,
596
+ "loss": 0.7203,
597
+ "step": 2100
598
+ },
599
+ {
600
+ "epoch": 0.47107071602748835,
601
+ "grad_norm": 0.08293402194976807,
602
+ "learning_rate": 0.00019165633951918247,
603
+ "loss": 0.7735,
604
+ "step": 2125
605
+ },
606
+ {
607
+ "epoch": 0.4766127244513412,
608
+ "grad_norm": 0.0809284895658493,
609
+ "learning_rate": 0.00019141544362541162,
610
+ "loss": 0.7412,
611
+ "step": 2150
612
+ },
613
+ {
614
+ "epoch": 0.48215473287519395,
615
+ "grad_norm": 0.08212891221046448,
616
+ "learning_rate": 0.00019117127532648773,
617
+ "loss": 0.7629,
618
+ "step": 2175
619
+ },
620
+ {
621
+ "epoch": 0.4876967412990468,
622
+ "grad_norm": 0.08602219074964523,
623
+ "learning_rate": 0.0001909238433629208,
624
+ "loss": 0.6935,
625
+ "step": 2200
626
+ },
627
+ {
628
+ "epoch": 0.4932387497228996,
629
+ "grad_norm": 0.08529417216777802,
630
+ "learning_rate": 0.0001906731565920505,
631
+ "loss": 0.7915,
632
+ "step": 2225
633
+ },
634
+ {
635
+ "epoch": 0.4987807581467524,
636
+ "grad_norm": 0.08774964511394501,
637
+ "learning_rate": 0.00019041922398772897,
638
+ "loss": 0.7359,
639
+ "step": 2250
640
+ },
641
+ {
642
+ "epoch": 0.5043227665706052,
643
+ "grad_norm": 0.08649475872516632,
644
+ "learning_rate": 0.00019016205463999984,
645
+ "loss": 0.7696,
646
+ "step": 2275
647
+ },
648
+ {
649
+ "epoch": 0.509864774994458,
650
+ "grad_norm": 0.0878506749868393,
651
+ "learning_rate": 0.00018990165775477252,
652
+ "loss": 0.7365,
653
+ "step": 2300
654
+ },
655
+ {
656
+ "epoch": 0.5154067834183108,
657
+ "grad_norm": 0.09131711721420288,
658
+ "learning_rate": 0.0001896380426534929,
659
+ "loss": 0.7809,
660
+ "step": 2325
661
+ },
662
+ {
663
+ "epoch": 0.5209487918421636,
664
+ "grad_norm": 0.07379825413227081,
665
+ "learning_rate": 0.00018937121877280957,
666
+ "loss": 0.7029,
667
+ "step": 2350
668
+ },
669
+ {
670
+ "epoch": 0.5264908002660164,
671
+ "grad_norm": 0.08535836637020111,
672
+ "learning_rate": 0.00018910119566423598,
673
+ "loss": 0.7679,
674
+ "step": 2375
675
+ },
676
+ {
677
+ "epoch": 0.5320328086898692,
678
+ "grad_norm": 0.06719771772623062,
679
+ "learning_rate": 0.00018882798299380864,
680
+ "loss": 0.7121,
681
+ "step": 2400
682
+ },
683
+ {
684
+ "epoch": 0.537574817113722,
685
+ "grad_norm": 0.09019796550273895,
686
+ "learning_rate": 0.00018855159054174093,
687
+ "loss": 0.7754,
688
+ "step": 2425
689
+ },
690
+ {
691
+ "epoch": 0.5431168255375748,
692
+ "grad_norm": 0.08144286274909973,
693
+ "learning_rate": 0.0001882720282020732,
694
+ "loss": 0.7255,
695
+ "step": 2450
696
+ },
697
+ {
698
+ "epoch": 0.5486588339614277,
699
+ "grad_norm": 0.08412271738052368,
700
+ "learning_rate": 0.0001879893059823185,
701
+ "loss": 0.7722,
702
+ "step": 2475
703
+ },
704
+ {
705
+ "epoch": 0.5542008423852804,
706
+ "grad_norm": 0.09016039222478867,
707
+ "learning_rate": 0.0001877034340031042,
708
+ "loss": 0.7275,
709
+ "step": 2500
710
+ },
711
+ {
712
+ "epoch": 0.5597428508091332,
713
+ "grad_norm": 0.08850298821926117,
714
+ "learning_rate": 0.00018741442249781,
715
+ "loss": 0.7828,
716
+ "step": 2525
717
+ },
718
+ {
719
+ "epoch": 0.565284859232986,
720
+ "grad_norm": 0.06989564746618271,
721
+ "learning_rate": 0.00018712228181220128,
722
+ "loss": 0.7111,
723
+ "step": 2550
724
+ },
725
+ {
726
+ "epoch": 0.5708268676568389,
727
+ "grad_norm": 0.09214618802070618,
728
+ "learning_rate": 0.00018682702240405906,
729
+ "loss": 0.7752,
730
+ "step": 2575
731
+ },
732
+ {
733
+ "epoch": 0.5763688760806917,
734
+ "grad_norm": 0.07766986638307571,
735
+ "learning_rate": 0.0001865286548428054,
736
+ "loss": 0.7108,
737
+ "step": 2600
738
+ },
739
+ {
740
+ "epoch": 0.5819108845045444,
741
+ "grad_norm": 0.07919591665267944,
742
+ "learning_rate": 0.00018622718980912514,
743
+ "loss": 0.775,
744
+ "step": 2625
745
+ },
746
+ {
747
+ "epoch": 0.5874528929283972,
748
+ "grad_norm": 0.07524783164262772,
749
+ "learning_rate": 0.00018592263809458361,
750
+ "loss": 0.6941,
751
+ "step": 2650
752
+ },
753
+ {
754
+ "epoch": 0.5929949013522501,
755
+ "grad_norm": 0.08549198508262634,
756
+ "learning_rate": 0.00018561501060124024,
757
+ "loss": 0.7718,
758
+ "step": 2675
759
+ },
760
+ {
761
+ "epoch": 0.5985369097761029,
762
+ "grad_norm": 0.08182788640260696,
763
+ "learning_rate": 0.0001853043183412584,
764
+ "loss": 0.7072,
765
+ "step": 2700
766
+ },
767
+ {
768
+ "epoch": 0.6040789181999556,
769
+ "grad_norm": 0.084741972386837,
770
+ "learning_rate": 0.00018499057243651096,
771
+ "loss": 0.7478,
772
+ "step": 2725
773
+ },
774
+ {
775
+ "epoch": 0.6096209266238085,
776
+ "grad_norm": 0.06824459880590439,
777
+ "learning_rate": 0.0001846737841181825,
778
+ "loss": 0.7238,
779
+ "step": 2750
780
+ },
781
+ {
782
+ "epoch": 0.6151629350476613,
783
+ "grad_norm": 0.08315033465623856,
784
+ "learning_rate": 0.00018435396472636704,
785
+ "loss": 0.7597,
786
+ "step": 2775
787
+ },
788
+ {
789
+ "epoch": 0.6207049434715141,
790
+ "grad_norm": 0.07116558402776718,
791
+ "learning_rate": 0.00018403112570966216,
792
+ "loss": 0.7096,
793
+ "step": 2800
794
+ },
795
+ {
796
+ "epoch": 0.6262469518953668,
797
+ "grad_norm": 0.08500215411186218,
798
+ "learning_rate": 0.00018370527862475916,
799
+ "loss": 0.756,
800
+ "step": 2825
801
+ },
802
+ {
803
+ "epoch": 0.6317889603192197,
804
+ "grad_norm": 0.07979004830121994,
805
+ "learning_rate": 0.00018337643513602933,
806
+ "loss": 0.6886,
807
+ "step": 2850
808
+ },
809
+ {
810
+ "epoch": 0.6373309687430725,
811
+ "grad_norm": 0.08140358328819275,
812
+ "learning_rate": 0.00018304460701510652,
813
+ "loss": 0.7648,
814
+ "step": 2875
815
+ },
816
+ {
817
+ "epoch": 0.6428729771669253,
818
+ "grad_norm": 0.07779423147439957,
819
+ "learning_rate": 0.0001827098061404656,
820
+ "loss": 0.7222,
821
+ "step": 2900
822
+ },
823
+ {
824
+ "epoch": 0.6484149855907781,
825
+ "grad_norm": 0.08853591978549957,
826
+ "learning_rate": 0.0001823720444969974,
827
+ "loss": 0.7736,
828
+ "step": 2925
829
+ },
830
+ {
831
+ "epoch": 0.6539569940146309,
832
+ "grad_norm": 0.07350102066993713,
833
+ "learning_rate": 0.0001820313341755795,
834
+ "loss": 0.7256,
835
+ "step": 2950
836
+ },
837
+ {
838
+ "epoch": 0.6594990024384837,
839
+ "grad_norm": 0.08152145147323608,
840
+ "learning_rate": 0.0001816876873726436,
841
+ "loss": 0.7598,
842
+ "step": 2975
843
+ },
844
+ {
845
+ "epoch": 0.6650410108623365,
846
+ "grad_norm": 0.08045897632837296,
847
+ "learning_rate": 0.00018134111638973876,
848
+ "loss": 0.7275,
849
+ "step": 3000
850
+ },
851
+ {
852
+ "epoch": 0.6705830192861894,
853
+ "grad_norm": 0.08514434099197388,
854
+ "learning_rate": 0.00018099163363309123,
855
+ "loss": 0.7688,
856
+ "step": 3025
857
+ },
858
+ {
859
+ "epoch": 0.6761250277100421,
860
+ "grad_norm": 0.060850344598293304,
861
+ "learning_rate": 0.00018063925161316012,
862
+ "loss": 0.7019,
863
+ "step": 3050
864
+ },
865
+ {
866
+ "epoch": 0.6816670361338949,
867
+ "grad_norm": 0.08471492677927017,
868
+ "learning_rate": 0.00018028398294418977,
869
+ "loss": 0.7573,
870
+ "step": 3075
871
+ },
872
+ {
873
+ "epoch": 0.6872090445577477,
874
+ "grad_norm": 0.0642291009426117,
875
+ "learning_rate": 0.00017992584034375798,
876
+ "loss": 0.7108,
877
+ "step": 3100
878
+ },
879
+ {
880
+ "epoch": 0.6927510529816006,
881
+ "grad_norm": 0.09357668459415436,
882
+ "learning_rate": 0.000179564836632321,
883
+ "loss": 0.7478,
884
+ "step": 3125
885
+ },
886
+ {
887
+ "epoch": 0.6982930614054533,
888
+ "grad_norm": 0.07198700308799744,
889
+ "learning_rate": 0.00017920098473275445,
890
+ "loss": 0.6973,
891
+ "step": 3150
892
+ },
893
+ {
894
+ "epoch": 0.7038350698293061,
895
+ "grad_norm": 0.08420095592737198,
896
+ "learning_rate": 0.00017883429766989064,
897
+ "loss": 0.7487,
898
+ "step": 3175
899
+ },
900
+ {
901
+ "epoch": 0.709377078253159,
902
+ "grad_norm": 0.06639819592237473,
903
+ "learning_rate": 0.00017846478857005255,
904
+ "loss": 0.6741,
905
+ "step": 3200
906
+ },
907
+ {
908
+ "epoch": 0.7149190866770118,
909
+ "grad_norm": 0.08200914412736893,
910
+ "learning_rate": 0.00017809247066058378,
911
+ "loss": 0.7526,
912
+ "step": 3225
913
+ },
914
+ {
915
+ "epoch": 0.7204610951008645,
916
+ "grad_norm": 0.07311141490936279,
917
+ "learning_rate": 0.0001777173572693751,
918
+ "loss": 0.677,
919
+ "step": 3250
920
+ },
921
+ {
922
+ "epoch": 0.7260031035247173,
923
+ "grad_norm": 0.08722089231014252,
924
+ "learning_rate": 0.00017733946182438726,
925
+ "loss": 0.7585,
926
+ "step": 3275
927
+ },
928
+ {
929
+ "epoch": 0.7315451119485702,
930
+ "grad_norm": 0.06589449942111969,
931
+ "learning_rate": 0.00017695879785317048,
932
+ "loss": 0.708,
933
+ "step": 3300
934
+ },
935
+ {
936
+ "epoch": 0.737087120372423,
937
+ "grad_norm": 0.08262074738740921,
938
+ "learning_rate": 0.0001765753789823801,
939
+ "loss": 0.749,
940
+ "step": 3325
941
+ },
942
+ {
943
+ "epoch": 0.7426291287962757,
944
+ "grad_norm": 0.07514823973178864,
945
+ "learning_rate": 0.00017618921893728867,
946
+ "loss": 0.6918,
947
+ "step": 3350
948
+ },
949
+ {
950
+ "epoch": 0.7481711372201286,
951
+ "grad_norm": 0.08757175505161285,
952
+ "learning_rate": 0.00017580033154129503,
953
+ "loss": 0.7445,
954
+ "step": 3375
955
+ },
956
+ {
957
+ "epoch": 0.7537131456439814,
958
+ "grad_norm": 0.0716458335518837,
959
+ "learning_rate": 0.0001754087307154289,
960
+ "loss": 0.7122,
961
+ "step": 3400
962
+ },
963
+ {
964
+ "epoch": 0.7592551540678342,
965
+ "grad_norm": 0.08453212678432465,
966
+ "learning_rate": 0.00017501443047785296,
967
+ "loss": 0.7656,
968
+ "step": 3425
969
+ },
970
+ {
971
+ "epoch": 0.7647971624916869,
972
+ "grad_norm": 0.06761575490236282,
973
+ "learning_rate": 0.00017461744494336098,
974
+ "loss": 0.6673,
975
+ "step": 3450
976
+ },
977
+ {
978
+ "epoch": 0.7703391709155398,
979
+ "grad_norm": 0.08577297627925873,
980
+ "learning_rate": 0.0001742177883228724,
981
+ "loss": 0.7494,
982
+ "step": 3475
983
+ },
984
+ {
985
+ "epoch": 0.7758811793393926,
986
+ "grad_norm": 0.05691730976104736,
987
+ "learning_rate": 0.00017381547492292376,
988
+ "loss": 0.6972,
989
+ "step": 3500
990
+ },
991
+ {
992
+ "epoch": 0.7814231877632454,
993
+ "grad_norm": 0.09115194529294968,
994
+ "learning_rate": 0.00017341051914515656,
995
+ "loss": 0.7706,
996
+ "step": 3525
997
+ },
998
+ {
999
+ "epoch": 0.7869651961870983,
1000
+ "grad_norm": 0.07214304804801941,
1001
+ "learning_rate": 0.00017300293548580162,
1002
+ "loss": 0.6807,
1003
+ "step": 3550
1004
+ },
1005
+ {
1006
+ "epoch": 0.792507204610951,
1007
+ "grad_norm": 0.08448139578104019,
1008
+ "learning_rate": 0.00017259273853516028,
1009
+ "loss": 0.7661,
1010
+ "step": 3575
1011
+ },
1012
+ {
1013
+ "epoch": 0.7980492130348038,
1014
+ "grad_norm": 0.08282499015331268,
1015
+ "learning_rate": 0.00017217994297708195,
1016
+ "loss": 0.7391,
1017
+ "step": 3600
1018
+ },
1019
+ {
1020
+ "epoch": 0.8035912214586566,
1021
+ "grad_norm": 0.0804004818201065,
1022
+ "learning_rate": 0.00017176456358843875,
1023
+ "loss": 0.7402,
1024
+ "step": 3625
1025
+ },
1026
+ {
1027
+ "epoch": 0.8091332298825095,
1028
+ "grad_norm": 0.07265755534172058,
1029
+ "learning_rate": 0.00017134661523859622,
1030
+ "loss": 0.7019,
1031
+ "step": 3650
1032
+ },
1033
+ {
1034
+ "epoch": 0.8146752383063622,
1035
+ "grad_norm": 0.08803457766771317,
1036
+ "learning_rate": 0.00017092611288888125,
1037
+ "loss": 0.7572,
1038
+ "step": 3675
1039
+ },
1040
+ {
1041
+ "epoch": 0.820217246730215,
1042
+ "grad_norm": 0.0652441680431366,
1043
+ "learning_rate": 0.0001705030715920464,
1044
+ "loss": 0.706,
1045
+ "step": 3700
1046
+ },
1047
+ {
1048
+ "epoch": 0.8257592551540678,
1049
+ "grad_norm": 0.08185753971338272,
1050
+ "learning_rate": 0.0001700775064917312,
1051
+ "loss": 0.764,
1052
+ "step": 3725
1053
+ },
1054
+ {
1055
+ "epoch": 0.8313012635779207,
1056
+ "grad_norm": 0.0859500914812088,
1057
+ "learning_rate": 0.00016964943282191984,
1058
+ "loss": 0.6927,
1059
+ "step": 3750
1060
+ },
1061
+ {
1062
+ "epoch": 0.8368432720017734,
1063
+ "grad_norm": 0.09176376461982727,
1064
+ "learning_rate": 0.00016921886590639602,
1065
+ "loss": 0.7567,
1066
+ "step": 3775
1067
+ },
1068
+ {
1069
+ "epoch": 0.8423852804256262,
1070
+ "grad_norm": 0.0646485984325409,
1071
+ "learning_rate": 0.0001687858211581943,
1072
+ "loss": 0.6848,
1073
+ "step": 3800
1074
+ },
1075
+ {
1076
+ "epoch": 0.8479272888494791,
1077
+ "grad_norm": 0.08545655012130737,
1078
+ "learning_rate": 0.00016835031407904839,
1079
+ "loss": 0.7546,
1080
+ "step": 3825
1081
+ },
1082
+ {
1083
+ "epoch": 0.8534692972733319,
1084
+ "grad_norm": 0.06338818371295929,
1085
+ "learning_rate": 0.00016791236025883626,
1086
+ "loss": 0.6655,
1087
+ "step": 3850
1088
+ },
1089
+ {
1090
+ "epoch": 0.8590113056971846,
1091
+ "grad_norm": 0.08781229704618454,
1092
+ "learning_rate": 0.00016747197537502205,
1093
+ "loss": 0.7441,
1094
+ "step": 3875
1095
+ },
1096
+ {
1097
+ "epoch": 0.8645533141210374,
1098
+ "grad_norm": 0.06220358610153198,
1099
+ "learning_rate": 0.00016702917519209487,
1100
+ "loss": 0.6795,
1101
+ "step": 3900
1102
+ },
1103
+ {
1104
+ "epoch": 0.8700953225448903,
1105
+ "grad_norm": 0.08917712420225143,
1106
+ "learning_rate": 0.0001665839755610044,
1107
+ "loss": 0.7552,
1108
+ "step": 3925
1109
+ },
1110
+ {
1111
+ "epoch": 0.8756373309687431,
1112
+ "grad_norm": 0.06624036282300949,
1113
+ "learning_rate": 0.00016613639241859355,
1114
+ "loss": 0.6632,
1115
+ "step": 3950
1116
+ },
1117
+ {
1118
+ "epoch": 0.8811793393925959,
1119
+ "grad_norm": 0.08898719400167465,
1120
+ "learning_rate": 0.00016568644178702803,
1121
+ "loss": 0.757,
1122
+ "step": 3975
1123
+ },
1124
+ {
1125
+ "epoch": 0.8867213478164487,
1126
+ "grad_norm": 0.05095354840159416,
1127
+ "learning_rate": 0.0001652341397732227,
1128
+ "loss": 0.6992,
1129
+ "step": 4000
1130
+ },
1131
+ {
1132
+ "epoch": 0.8922633562403015,
1133
+ "grad_norm": 0.08842916786670685,
1134
+ "learning_rate": 0.0001647795025682649,
1135
+ "loss": 0.7504,
1136
+ "step": 4025
1137
+ },
1138
+ {
1139
+ "epoch": 0.8978053646641543,
1140
+ "grad_norm": 0.0758206844329834,
1141
+ "learning_rate": 0.00016432254644683516,
1142
+ "loss": 0.7081,
1143
+ "step": 4050
1144
+ },
1145
+ {
1146
+ "epoch": 0.903347373088007,
1147
+ "grad_norm": 0.0940496176481247,
1148
+ "learning_rate": 0.0001638632877666243,
1149
+ "loss": 0.746,
1150
+ "step": 4075
1151
+ },
1152
+ {
1153
+ "epoch": 0.9088893815118599,
1154
+ "grad_norm": 0.06626766920089722,
1155
+ "learning_rate": 0.00016340174296774804,
1156
+ "loss": 0.6647,
1157
+ "step": 4100
1158
+ },
1159
+ {
1160
+ "epoch": 0.9144313899357127,
1161
+ "grad_norm": 0.08919317275285721,
1162
+ "learning_rate": 0.00016293792857215844,
1163
+ "loss": 0.7516,
1164
+ "step": 4125
1165
+ },
1166
+ {
1167
+ "epoch": 0.9199733983595655,
1168
+ "grad_norm": 0.06990760564804077,
1169
+ "learning_rate": 0.00016247186118305252,
1170
+ "loss": 0.7011,
1171
+ "step": 4150
1172
+ },
1173
+ {
1174
+ "epoch": 0.9255154067834183,
1175
+ "grad_norm": 0.0870794802904129,
1176
+ "learning_rate": 0.00016200355748427782,
1177
+ "loss": 0.7529,
1178
+ "step": 4175
1179
+ },
1180
+ {
1181
+ "epoch": 0.9310574152072711,
1182
+ "grad_norm": 0.06882854551076889,
1183
+ "learning_rate": 0.00016153303423973526,
1184
+ "loss": 0.7005,
1185
+ "step": 4200
1186
+ },
1187
+ {
1188
+ "epoch": 0.9365994236311239,
1189
+ "grad_norm": 0.084992416203022,
1190
+ "learning_rate": 0.0001610603082927789,
1191
+ "loss": 0.7519,
1192
+ "step": 4225
1193
+ },
1194
+ {
1195
+ "epoch": 0.9421414320549767,
1196
+ "grad_norm": 0.0638299211859703,
1197
+ "learning_rate": 0.00016058539656561323,
1198
+ "loss": 0.716,
1199
+ "step": 4250
1200
+ },
1201
+ {
1202
+ "epoch": 0.9476834404788296,
1203
+ "grad_norm": 0.08899606764316559,
1204
+ "learning_rate": 0.00016010831605868715,
1205
+ "loss": 0.7257,
1206
+ "step": 4275
1207
+ },
1208
+ {
1209
+ "epoch": 0.9532254489026823,
1210
+ "grad_norm": 0.06550378352403641,
1211
+ "learning_rate": 0.00015962908385008565,
1212
+ "loss": 0.7174,
1213
+ "step": 4300
1214
+ },
1215
+ {
1216
+ "epoch": 0.9587674573265351,
1217
+ "grad_norm": 0.09001540392637253,
1218
+ "learning_rate": 0.00015914771709491828,
1219
+ "loss": 0.7271,
1220
+ "step": 4325
1221
+ },
1222
+ {
1223
+ "epoch": 0.9643094657503879,
1224
+ "grad_norm": 0.06641615182161331,
1225
+ "learning_rate": 0.000158664233024705,
1226
+ "loss": 0.69,
1227
+ "step": 4350
1228
+ },
1229
+ {
1230
+ "epoch": 0.9698514741742408,
1231
+ "grad_norm": 0.08917039632797241,
1232
+ "learning_rate": 0.0001581786489467596,
1233
+ "loss": 0.7483,
1234
+ "step": 4375
1235
+ },
1236
+ {
1237
+ "epoch": 0.9753934825980936,
1238
+ "grad_norm": 0.05995697155594826,
1239
+ "learning_rate": 0.00015769098224356992,
1240
+ "loss": 0.7033,
1241
+ "step": 4400
1242
+ },
1243
+ {
1244
+ "epoch": 0.9809354910219463,
1245
+ "grad_norm": 0.08998765051364899,
1246
+ "learning_rate": 0.00015720125037217572,
1247
+ "loss": 0.7462,
1248
+ "step": 4425
1249
+ },
1250
+ {
1251
+ "epoch": 0.9864774994457992,
1252
+ "grad_norm": 0.05868702754378319,
1253
+ "learning_rate": 0.00015670947086354376,
1254
+ "loss": 0.6654,
1255
+ "step": 4450
1256
+ },
1257
+ {
1258
+ "epoch": 0.992019507869652,
1259
+ "grad_norm": 0.0880926102399826,
1260
+ "learning_rate": 0.00015621566132194005,
1261
+ "loss": 0.752,
1262
+ "step": 4475
1263
+ },
1264
+ {
1265
+ "epoch": 0.9975615162935048,
1266
+ "grad_norm": 0.08538970351219177,
1267
+ "learning_rate": 0.00015571983942430005,
1268
+ "loss": 0.7338,
1269
+ "step": 4500
1270
+ },
1271
+ {
1272
+ "epoch": 1.0031035247173576,
1273
+ "grad_norm": 0.0827050730586052,
1274
+ "learning_rate": 0.0001552220229195956,
1275
+ "loss": 0.7174,
1276
+ "step": 4525
1277
+ },
1278
+ {
1279
+ "epoch": 1.0086455331412103,
1280
+ "grad_norm": 0.10867294669151306,
1281
+ "learning_rate": 0.00015472222962819955,
1282
+ "loss": 0.7637,
1283
+ "step": 4550
1284
+ },
1285
+ {
1286
+ "epoch": 1.0141875415650632,
1287
+ "grad_norm": 0.08738269656896591,
1288
+ "learning_rate": 0.00015422047744124802,
1289
+ "loss": 0.6247,
1290
+ "step": 4575
1291
+ },
1292
+ {
1293
+ "epoch": 1.019729549988916,
1294
+ "grad_norm": 0.12865987420082092,
1295
+ "learning_rate": 0.0001537167843199998,
1296
+ "loss": 0.7424,
1297
+ "step": 4600
1298
+ },
1299
+ {
1300
+ "epoch": 1.0252715584127687,
1301
+ "grad_norm": 0.08619695156812668,
1302
+ "learning_rate": 0.00015321116829519345,
1303
+ "loss": 0.6461,
1304
+ "step": 4625
1305
+ },
1306
+ {
1307
+ "epoch": 1.0308135668366216,
1308
+ "grad_norm": 0.11726492643356323,
1309
+ "learning_rate": 0.0001527036474664019,
1310
+ "loss": 0.7433,
1311
+ "step": 4650
1312
+ },
1313
+ {
1314
+ "epoch": 1.0363555752604743,
1315
+ "grad_norm": 0.08198727667331696,
1316
+ "learning_rate": 0.0001521942400013844,
1317
+ "loss": 0.6086,
1318
+ "step": 4675
1319
+ },
1320
+ {
1321
+ "epoch": 1.0418975836843272,
1322
+ "grad_norm": 0.11951526254415512,
1323
+ "learning_rate": 0.00015168296413543635,
1324
+ "loss": 0.7521,
1325
+ "step": 4700
1326
+ },
1327
+ {
1328
+ "epoch": 1.04743959210818,
1329
+ "grad_norm": 0.08714735507965088,
1330
+ "learning_rate": 0.0001511698381707363,
1331
+ "loss": 0.631,
1332
+ "step": 4725
1333
+ },
1334
+ {
1335
+ "epoch": 1.0529816005320327,
1336
+ "grad_norm": 0.13869455456733704,
1337
+ "learning_rate": 0.00015065488047569107,
1338
+ "loss": 0.7524,
1339
+ "step": 4750
1340
+ },
1341
+ {
1342
+ "epoch": 1.0585236089558856,
1343
+ "grad_norm": 0.08524268865585327,
1344
+ "learning_rate": 0.00015013810948427794,
1345
+ "loss": 0.6617,
1346
+ "step": 4775
1347
+ },
1348
+ {
1349
+ "epoch": 1.0640656173797385,
1350
+ "grad_norm": 0.11017199605703354,
1351
+ "learning_rate": 0.00014961954369538494,
1352
+ "loss": 0.7598,
1353
+ "step": 4800
1354
+ },
1355
+ {
1356
+ "epoch": 1.0696076258035911,
1357
+ "grad_norm": 0.0834374874830246,
1358
+ "learning_rate": 0.00014909920167214858,
1359
+ "loss": 0.627,
1360
+ "step": 4825
1361
+ },
1362
+ {
1363
+ "epoch": 1.075149634227444,
1364
+ "grad_norm": 0.1357167363166809,
1365
+ "learning_rate": 0.0001485771020412894,
1366
+ "loss": 0.7466,
1367
+ "step": 4850
1368
+ },
1369
+ {
1370
+ "epoch": 1.080691642651297,
1371
+ "grad_norm": 0.08910629153251648,
1372
+ "learning_rate": 0.00014805326349244503,
1373
+ "loss": 0.6238,
1374
+ "step": 4875
1375
+ },
1376
+ {
1377
+ "epoch": 1.0862336510751496,
1378
+ "grad_norm": 0.10706546157598495,
1379
+ "learning_rate": 0.00014752770477750144,
1380
+ "loss": 0.7533,
1381
+ "step": 4900
1382
+ },
1383
+ {
1384
+ "epoch": 1.0917756594990025,
1385
+ "grad_norm": 0.09201759845018387,
1386
+ "learning_rate": 0.00014700044470992136,
1387
+ "loss": 0.6521,
1388
+ "step": 4925
1389
+ },
1390
+ {
1391
+ "epoch": 1.0973176679228553,
1392
+ "grad_norm": 0.14048361778259277,
1393
+ "learning_rate": 0.00014647150216407106,
1394
+ "loss": 0.7412,
1395
+ "step": 4950
1396
+ },
1397
+ {
1398
+ "epoch": 1.102859676346708,
1399
+ "grad_norm": 0.08308299630880356,
1400
+ "learning_rate": 0.00014594089607454454,
1401
+ "loss": 0.6333,
1402
+ "step": 4975
1403
+ },
1404
+ {
1405
+ "epoch": 1.108401684770561,
1406
+ "grad_norm": 0.12057497352361679,
1407
+ "learning_rate": 0.00014540864543548582,
1408
+ "loss": 0.7538,
1409
+ "step": 5000
1410
+ }
1411
+ ],
1412
+ "logging_steps": 25,
1413
+ "max_steps": 13533,
1414
+ "num_input_tokens_seen": 0,
1415
+ "num_train_epochs": 3,
1416
+ "save_steps": 200,
1417
+ "stateful_callbacks": {
1418
+ "TrainerControl": {
1419
+ "args": {
1420
+ "should_epoch_stop": false,
1421
+ "should_evaluate": false,
1422
+ "should_log": false,
1423
+ "should_save": true,
1424
+ "should_training_stop": false
1425
+ },
1426
+ "attributes": {}
1427
+ }
1428
+ },
1429
+ "total_flos": 2.4028324505774285e+18,
1430
+ "train_batch_size": 2,
1431
+ "trial_name": null,
1432
+ "trial_params": null
1433
+ }