Hugging Face's logo

RayDu0010
/
tulu10k_klora

Safetensors
Model card Files
xet
 Community
RayDu0010 commited on
Commit
fe086ab
·
verified ·
1 Parent(s): 2c9a6ad

Upload folder using huggingface_hub

Browse files
Files changed (14) hide show
  1. instruct/140_128_e3_3e-5/adapter_config.json +39 -0
  2. instruct/140_128_e3_3e-5/adapter_model.safetensors +3 -0
  3. instruct/140_128_e3_3e-5/added_tokens.json +9 -0
  4. instruct/140_128_e3_3e-5/all_results.json +9 -0
  5. instruct/140_128_e3_3e-5/chat_template.jinja +62 -0
  6. instruct/140_128_e3_3e-5/config.json +32 -0
  7. instruct/140_128_e3_3e-5/merges.txt +0 -0
  8. instruct/140_128_e3_3e-5/special_tokens_map.json +33 -0
  9. instruct/140_128_e3_3e-5/tokenizer.json +0 -0
  10. instruct/140_128_e3_3e-5/tokenizer_config.json +234 -0
  11. instruct/140_128_e3_3e-5/train_results.json +9 -0
  12. instruct/140_128_e3_3e-5/trainer_state.json +3697 -0
  13. instruct/140_128_e3_3e-5/training_args.bin +3 -0
  14. instruct/140_128_e3_3e-5/vocab.json +0 -0
instruct/140_128_e3_3e-5/adapter_config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "ibm-granite/granite-3.3-8b-instruct",
5
+ "bias": "none",
6
+ "corda_config": null,
7
+ "eva_config": null,
8
+ "exclude_modules": null,
9
+ "fan_in_fan_out": false,
10
+ "inference_mode": true,
11
+ "init_lora_weights": true,
12
+ "layer_replication": null,
13
+ "layers_pattern": null,
14
+ "layers_to_transform": null,
15
+ "loftq_config": {},
16
+ "lora_alpha": 256,
17
+ "lora_bias": false,
18
+ "lora_dropout": 0.05,
19
+ "megatron_config": null,
20
+ "megatron_core": "megatron.core",
21
+ "modules_to_save": null,
22
+ "peft_type": "LORA",
23
+ "r": 128,
24
+ "rank_pattern": {},
25
+ "revision": null,
26
+ "target_modules": [
27
+ "q_proj",
28
+ "gate_proj",
29
+ "up_proj",
30
+ "v_proj",
31
+ "down_proj",
32
+ "k_proj",
33
+ "o_proj"
34
+ ],
35
+ "task_type": "CAUSAL_LM",
36
+ "trainable_token_indices": null,
37
+ "use_dora": false,
38
+ "use_rslora": false
39
+ }
instruct/140_128_e3_3e-5/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89fd84baa0d4c7952022b6bfe875d4667abfb2a3d20f26e1dbc0b07223497dba
3
+ size 791751704
instruct/140_128_e3_3e-5/added_tokens.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "<|end_of_cite|>": 49156,
3
+ "<|end_of_plugin|>": 49158,
4
+ "<|end_of_role|>": 49153,
5
+ "<|start_of_cite|>": 49155,
6
+ "<|start_of_plugin|>": 49157,
7
+ "<|start_of_role|>": 49152,
8
+ "<|tool_call|>": 49154
9
+ }
instruct/140_128_e3_3e-5/all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 3.4112320082133647e+18,
4
+ "train_loss": 0.3677417017376743,
5
+ "train_runtime": 1618.6418,
6
+ "train_samples": 27818,
7
+ "train_samples_per_second": 51.558,
8
+ "train_steps_per_second": 1.612
9
+ }
instruct/140_128_e3_3e-5/chat_template.jinja ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {# Alias tools -> available_tools #}
2
+ {%- if tools and not available_tools -%}
3
+ {%- set available_tools = tools -%}
4
+ {%- endif -%}
5
+ {%- if messages[0]['role'] == 'system' %}
6
+ {%- set system_message = messages[0]['content'] %}
7
+ {%- set loop_messages = messages[1:] %}
8
+ {%- else %}
9
+ {%- set system_message = "Knowledge Cutoff Date: April 2024.
10
+ Today's Date: " + strftime_now('%B %d, %Y') + ".
11
+ You are Granite, developed by IBM." %}
12
+ {%- if available_tools and documents %}
13
+ {%- set system_message = system_message + " You are a helpful assistant with access to the following tools. When a tool is required to answer the user's query, respond only with <|tool_call|> followed by a JSON list of tools used. If a tool does not exist in the provided list of tools, notify the user that you do not have the ability to fulfill the request.
14
+ Write the response to the user's input by strictly aligning with the facts in the provided documents. If the information needed to answer the question is not available in the documents, inform the user that the question cannot be answered based on the available data." %}
15
+ {%- elif available_tools %}
16
+ {%- set system_message = system_message + " You are a helpful assistant with access to the following tools. When a tool is required to answer the user's query, respond only with <|tool_call|> followed by a JSON list of tools used. If a tool does not exist in the provided list of tools, notify the user that you do not have the ability to fulfill the request." %}
17
+ {%- elif documents %}
18
+ {%- set system_message = system_message + " Write the response to the user's input by strictly aligning with the facts in the provided documents. If the information needed to answer the question is not available in the documents, inform the user that the question cannot be answered based on the available data." %}
19
+ {%- elif thinking %}
20
+ {%- set system_message = system_message + " You are a helpful AI assistant.
21
+ Respond to every user query in a comprehensive and detailed way. You can write down your thoughts and reasoning process before responding. In the thought process, engage in a comprehensive cycle of analysis, summarization, exploration, reassessment, reflection, backtracing, and iteration to develop well-considered thinking process. In the response section, based on various attempts, explorations, and reflections from the thoughts section, systematically present the final solution that you deem correct. The response should summarize the thought process. Write your thoughts between <think></think> and write your response between <response></response> for each user query." %}
22
+ {%- else %}
23
+ {%- set system_message = system_message + " You are a helpful AI assistant." %}
24
+ {%- endif %}
25
+ {%- if 'citations' in controls and documents %}
26
+ {%- set system_message = system_message + '
27
+ Use the symbols <|start_of_cite|> and <|end_of_cite|> to indicate when a fact comes from a document in the search result, e.g <|start_of_cite|> {document_id: 1}my fact <|end_of_cite|> for a fact from document 1. Afterwards, list all the citations with their corresponding documents in an ordered list.' %}
28
+ {%- endif %}
29
+ {%- if 'hallucinations' in controls and documents %}
30
+ {%- set system_message = system_message + '
31
+ Finally, after the response is written, include a numbered list of sentences from the response with a corresponding risk value that are hallucinated and not based in the documents.' %}
32
+ {%- endif %}
33
+ {%- set loop_messages = messages %}
34
+ {%- endif %}
35
+ {{- '<|start_of_role|>system<|end_of_role|>' + system_message + '<|end_of_text|>
36
+ ' }}
37
+ {%- if available_tools %}
38
+ {{- '<|start_of_role|>available_tools<|end_of_role|>' }}
39
+ {{- available_tools | tojson(indent=4) }}
40
+ {{- '<|end_of_text|>
41
+ ' }}
42
+ {%- endif %}
43
+ {%- if documents %}
44
+ {%- for document in documents %}
45
+ {{- '<|start_of_role|>document {"document_id": "' + document['doc_id'] | string + '"}<|end_of_role|>
46
+ ' }}
47
+ {{- document['text'] }}
48
+ {{- '<|end_of_text|>
49
+ ' }}
50
+ {%- endfor %}
51
+ {%- endif %}
52
+ {%- for message in loop_messages %}
53
+ {{- '<|start_of_role|>' + message['role'] + '<|end_of_role|>' + message['content'] + '<|end_of_text|>
54
+ ' }}
55
+ {%- if loop.last and add_generation_prompt %}
56
+ {{- '<|start_of_role|>assistant' }}
57
+ {%- if controls %}
58
+ {{- ' ' + controls | tojson()}}
59
+ {%- endif %}
60
+ {{- '<|end_of_role|>' }}
61
+ {%- endif %}
62
+ {%- endfor %}
instruct/140_128_e3_3e-5/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "GraniteForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attention_multiplier": 0.0078125,
8
+ "bos_token_id": 0,
9
+ "embedding_multiplier": 12.0,
10
+ "eos_token_id": 0,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 4096,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 12800,
15
+ "logits_scaling": 16.0,
16
+ "max_position_embeddings": 131072,
17
+ "mlp_bias": false,
18
+ "model_type": "granite",
19
+ "num_attention_heads": 32,
20
+ "num_hidden_layers": 40,
21
+ "num_key_value_heads": 8,
22
+ "pad_token_id": 0,
23
+ "residual_multiplier": 0.22,
24
+ "rms_norm_eps": 1e-05,
25
+ "rope_scaling": null,
26
+ "rope_theta": 10000000.0,
27
+ "tie_word_embeddings": true,
28
+ "torch_dtype": "bfloat16",
29
+ "transformers_version": "4.52.4",
30
+ "use_cache": true,
31
+ "vocab_size": 49159
32
+ }
instruct/140_128_e3_3e-5/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
instruct/140_128_e3_3e-5/special_tokens_map.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|start_of_role|>",
4
+ "<|end_of_role|>",
5
+ "<|tool_call|>",
6
+ "<|start_of_cite|>",
7
+ "<|end_of_cite|>",
8
+ "<|start_of_plugin|>",
9
+ "<|end_of_plugin|>"
10
+ ],
11
+ "bos_token": {
12
+ "content": "<|end_of_text|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "eos_token": {
19
+ "content": "<|end_of_text|>",
20
+ "lstrip": false,
21
+ "normalized": false,
22
+ "rstrip": false,
23
+ "single_word": false
24
+ },
25
+ "pad_token": "<|end_of_plugin|>",
26
+ "unk_token": {
27
+ "content": "<|end_of_text|>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
instruct/140_128_e3_3e-5/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
instruct/140_128_e3_3e-5/tokenizer_config.json ADDED
@@ -0,0 +1,234 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<|end_of_text|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<fim_prefix>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "<fim_middle>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "3": {
30
+ "content": "<fim_suffix>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "4": {
38
+ "content": "<fim_pad>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "5": {
46
+ "content": "<filename>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "6": {
54
+ "content": "<gh_stars>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "7": {
62
+ "content": "<issue_start>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "8": {
70
+ "content": "<issue_comment>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "9": {
78
+ "content": "<issue_closed>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "10": {
86
+ "content": "<jupyter_start>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "11": {
94
+ "content": "<jupyter_text>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "12": {
102
+ "content": "<jupyter_code>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "13": {
110
+ "content": "<jupyter_output>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "14": {
118
+ "content": "<empty_output>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": true
124
+ },
125
+ "15": {
126
+ "content": "<commit_before>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": true
132
+ },
133
+ "16": {
134
+ "content": "<commit_msg>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": true
140
+ },
141
+ "17": {
142
+ "content": "<commit_after>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": true
148
+ },
149
+ "18": {
150
+ "content": "<reponame>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": true
156
+ },
157
+ "49152": {
158
+ "content": "<|start_of_role|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": true
164
+ },
165
+ "49153": {
166
+ "content": "<|end_of_role|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": true
172
+ },
173
+ "49154": {
174
+ "content": "<|tool_call|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": true
180
+ },
181
+ "49155": {
182
+ "content": "<|start_of_cite|>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": true
188
+ },
189
+ "49156": {
190
+ "content": "<|end_of_cite|>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": true
196
+ },
197
+ "49157": {
198
+ "content": "<|start_of_plugin|>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": true
204
+ },
205
+ "49158": {
206
+ "content": "<|end_of_plugin|>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": true
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|start_of_role|>",
216
+ "<|end_of_role|>",
217
+ "<|tool_call|>",
218
+ "<|start_of_cite|>",
219
+ "<|end_of_cite|>",
220
+ "<|start_of_plugin|>",
221
+ "<|end_of_plugin|>"
222
+ ],
223
+ "bos_token": "<|end_of_text|>",
224
+ "clean_up_tokenization_spaces": true,
225
+ "eos_token": "<|end_of_text|>",
226
+ "errors": "replace",
227
+ "extra_special_tokens": {},
228
+ "model_max_length": 8192,
229
+ "pad_token": "<|end_of_plugin|>",
230
+ "padding_side": "left",
231
+ "tokenizer_class": "GPT2Tokenizer",
232
+ "unk_token": "<|end_of_text|>",
233
+ "vocab_size": 49152
234
+ }
instruct/140_128_e3_3e-5/train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 3.4112320082133647e+18,
4
+ "train_loss": 0.3677417017376743,
5
+ "train_runtime": 1618.6418,
6
+ "train_samples": 27818,
7
+ "train_samples_per_second": 51.558,
8
+ "train_steps_per_second": 1.612
9
+ }
instruct/140_128_e3_3e-5/trainer_state.json ADDED
@@ -0,0 +1,3697 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2610,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.005750431282346176,
14
+ "grad_norm": 2.3784658908843994,
15
+ "learning_rate": 9.16030534351145e-07,
16
+ "loss": 1.6109,
17
+ "step": 5
18
+ },
19
+ {
20
+ "epoch": 0.011500862564692352,
21
+ "grad_norm": 2.491219997406006,
22
+ "learning_rate": 2.0610687022900764e-06,
23
+ "loss": 1.573,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 0.017251293847038527,
28
+ "grad_norm": 1.646517276763916,
29
+ "learning_rate": 3.2061068702290075e-06,
30
+ "loss": 1.5339,
31
+ "step": 15
32
+ },
33
+ {
34
+ "epoch": 0.023001725129384705,
35
+ "grad_norm": 1.1218372583389282,
36
+ "learning_rate": 4.3511450381679385e-06,
37
+ "loss": 1.4703,
38
+ "step": 20
39
+ },
40
+ {
41
+ "epoch": 0.02875215641173088,
42
+ "grad_norm": 0.6851752996444702,
43
+ "learning_rate": 5.49618320610687e-06,
44
+ "loss": 1.4597,
45
+ "step": 25
46
+ },
47
+ {
48
+ "epoch": 0.034502587694077054,
49
+ "grad_norm": 0.555807888507843,
50
+ "learning_rate": 6.641221374045802e-06,
51
+ "loss": 1.477,
52
+ "step": 30
53
+ },
54
+ {
55
+ "epoch": 0.04025301897642323,
56
+ "grad_norm": 0.5165654420852661,
57
+ "learning_rate": 7.786259541984733e-06,
58
+ "loss": 1.4053,
59
+ "step": 35
60
+ },
61
+ {
62
+ "epoch": 0.04600345025876941,
63
+ "grad_norm": 0.3987463116645813,
64
+ "learning_rate": 8.931297709923664e-06,
65
+ "loss": 1.4395,
66
+ "step": 40
67
+ },
68
+ {
69
+ "epoch": 0.051753881541115584,
70
+ "grad_norm": 0.47501805424690247,
71
+ "learning_rate": 1.0076335877862596e-05,
72
+ "loss": 1.3747,
73
+ "step": 45
74
+ },
75
+ {
76
+ "epoch": 0.05750431282346176,
77
+ "grad_norm": 0.3766283094882965,
78
+ "learning_rate": 1.1221374045801527e-05,
79
+ "loss": 1.3643,
80
+ "step": 50
81
+ },
82
+ {
83
+ "epoch": 0.06325474410580793,
84
+ "grad_norm": 0.44497984647750854,
85
+ "learning_rate": 1.2366412213740458e-05,
86
+ "loss": 1.3738,
87
+ "step": 55
88
+ },
89
+ {
90
+ "epoch": 0.06900517538815411,
91
+ "grad_norm": 0.37557175755500793,
92
+ "learning_rate": 1.351145038167939e-05,
93
+ "loss": 1.3185,
94
+ "step": 60
95
+ },
96
+ {
97
+ "epoch": 0.07475560667050028,
98
+ "grad_norm": 0.37465789914131165,
99
+ "learning_rate": 1.465648854961832e-05,
100
+ "loss": 1.3385,
101
+ "step": 65
102
+ },
103
+ {
104
+ "epoch": 0.08050603795284646,
105
+ "grad_norm": 0.40240317583084106,
106
+ "learning_rate": 1.580152671755725e-05,
107
+ "loss": 1.3676,
108
+ "step": 70
109
+ },
110
+ {
111
+ "epoch": 0.08625646923519265,
112
+ "grad_norm": 0.40319573879241943,
113
+ "learning_rate": 1.6946564885496184e-05,
114
+ "loss": 1.3501,
115
+ "step": 75
116
+ },
117
+ {
118
+ "epoch": 0.09200690051753882,
119
+ "grad_norm": 0.38340315222740173,
120
+ "learning_rate": 1.8091603053435114e-05,
121
+ "loss": 1.3496,
122
+ "step": 80
123
+ },
124
+ {
125
+ "epoch": 0.097757331799885,
126
+ "grad_norm": 0.45284807682037354,
127
+ "learning_rate": 1.9236641221374046e-05,
128
+ "loss": 1.3464,
129
+ "step": 85
130
+ },
131
+ {
132
+ "epoch": 0.10350776308223117,
133
+ "grad_norm": 0.42866745591163635,
134
+ "learning_rate": 2.038167938931298e-05,
135
+ "loss": 1.3252,
136
+ "step": 90
137
+ },
138
+ {
139
+ "epoch": 0.10925819436457734,
140
+ "grad_norm": 0.4262651205062866,
141
+ "learning_rate": 2.152671755725191e-05,
142
+ "loss": 1.3096,
143
+ "step": 95
144
+ },
145
+ {
146
+ "epoch": 0.11500862564692352,
147
+ "grad_norm": 0.4294212758541107,
148
+ "learning_rate": 2.267175572519084e-05,
149
+ "loss": 1.329,
150
+ "step": 100
151
+ },
152
+ {
153
+ "epoch": 0.12075905692926969,
154
+ "grad_norm": 0.4155022203922272,
155
+ "learning_rate": 2.381679389312977e-05,
156
+ "loss": 1.3037,
157
+ "step": 105
158
+ },
159
+ {
160
+ "epoch": 0.12650948821161587,
161
+ "grad_norm": 0.41887587308883667,
162
+ "learning_rate": 2.4961832061068703e-05,
163
+ "loss": 1.2645,
164
+ "step": 110
165
+ },
166
+ {
167
+ "epoch": 0.13225991949396204,
168
+ "grad_norm": 0.5171315670013428,
169
+ "learning_rate": 2.6106870229007633e-05,
170
+ "loss": 1.2457,
171
+ "step": 115
172
+ },
173
+ {
174
+ "epoch": 0.13801035077630822,
175
+ "grad_norm": 0.4775677025318146,
176
+ "learning_rate": 2.7251908396946566e-05,
177
+ "loss": 1.2657,
178
+ "step": 120
179
+ },
180
+ {
181
+ "epoch": 0.1437607820586544,
182
+ "grad_norm": 0.4620819389820099,
183
+ "learning_rate": 2.8396946564885498e-05,
184
+ "loss": 1.238,
185
+ "step": 125
186
+ },
187
+ {
188
+ "epoch": 0.14951121334100057,
189
+ "grad_norm": 0.5237811207771301,
190
+ "learning_rate": 2.9541984732824428e-05,
191
+ "loss": 1.2294,
192
+ "step": 130
193
+ },
194
+ {
195
+ "epoch": 0.15526164462334674,
196
+ "grad_norm": 0.5393637418746948,
197
+ "learning_rate": 2.9999891594843296e-05,
198
+ "loss": 1.2615,
199
+ "step": 135
200
+ },
201
+ {
202
+ "epoch": 0.16101207590569291,
203
+ "grad_norm": 0.5105319619178772,
204
+ "learning_rate": 2.9999229124559968e-05,
205
+ "loss": 1.269,
206
+ "step": 140
207
+ },
208
+ {
209
+ "epoch": 0.16676250718803912,
210
+ "grad_norm": 0.5121212005615234,
211
+ "learning_rate": 2.9997964435646305e-05,
212
+ "loss": 1.1527,
213
+ "step": 145
214
+ },
215
+ {
216
+ "epoch": 0.1725129384703853,
217
+ "grad_norm": 0.6401780247688293,
218
+ "learning_rate": 2.999609757887952e-05,
219
+ "loss": 1.2087,
220
+ "step": 150
221
+ },
222
+ {
223
+ "epoch": 0.17826336975273147,
224
+ "grad_norm": 0.6539484262466431,
225
+ "learning_rate": 2.9993628629213865e-05,
226
+ "loss": 1.1612,
227
+ "step": 155
228
+ },
229
+ {
230
+ "epoch": 0.18401380103507764,
231
+ "grad_norm": 0.5551204085350037,
232
+ "learning_rate": 2.99905576857776e-05,
233
+ "loss": 1.1467,
234
+ "step": 160
235
+ },
236
+ {
237
+ "epoch": 0.18976423231742381,
238
+ "grad_norm": 0.5668113827705383,
239
+ "learning_rate": 2.998688487186902e-05,
240
+ "loss": 1.1626,
241
+ "step": 165
242
+ },
243
+ {
244
+ "epoch": 0.19551466359977,
245
+ "grad_norm": 0.5834711790084839,
246
+ "learning_rate": 2.9982610334951495e-05,
247
+ "loss": 1.1903,
248
+ "step": 170
249
+ },
250
+ {
251
+ "epoch": 0.20126509488211616,
252
+ "grad_norm": 0.6087402701377869,
253
+ "learning_rate": 2.9977734246647567e-05,
254
+ "loss": 1.1442,
255
+ "step": 175
256
+ },
257
+ {
258
+ "epoch": 0.20701552616446234,
259
+ "grad_norm": 0.6042896509170532,
260
+ "learning_rate": 2.9972256802732047e-05,
261
+ "loss": 1.1532,
262
+ "step": 180
263
+ },
264
+ {
265
+ "epoch": 0.2127659574468085,
266
+ "grad_norm": 0.5594789981842041,
267
+ "learning_rate": 2.996617822312415e-05,
268
+ "loss": 1.1336,
269
+ "step": 185
270
+ },
271
+ {
272
+ "epoch": 0.2185163887291547,
273
+ "grad_norm": 0.5978757739067078,
274
+ "learning_rate": 2.9959498751878683e-05,
275
+ "loss": 1.1509,
276
+ "step": 190
277
+ },
278
+ {
279
+ "epoch": 0.22426682001150086,
280
+ "grad_norm": 0.612287163734436,
281
+ "learning_rate": 2.9952218657176225e-05,
282
+ "loss": 1.1118,
283
+ "step": 195
284
+ },
285
+ {
286
+ "epoch": 0.23001725129384704,
287
+ "grad_norm": 0.61568683385849,
288
+ "learning_rate": 2.9944338231312376e-05,
289
+ "loss": 1.0761,
290
+ "step": 200
291
+ },
292
+ {
293
+ "epoch": 0.2357676825761932,
294
+ "grad_norm": 0.7246344685554504,
295
+ "learning_rate": 2.9935857790686017e-05,
296
+ "loss": 1.0756,
297
+ "step": 205
298
+ },
299
+ {
300
+ "epoch": 0.24151811385853938,
301
+ "grad_norm": 0.6384976506233215,
302
+ "learning_rate": 2.99267776757866e-05,
303
+ "loss": 1.0855,
304
+ "step": 210
305
+ },
306
+ {
307
+ "epoch": 0.24726854514088556,
308
+ "grad_norm": 0.7112752795219421,
309
+ "learning_rate": 2.9917098251180488e-05,
310
+ "loss": 1.0955,
311
+ "step": 215
312
+ },
313
+ {
314
+ "epoch": 0.25301897642323173,
315
+ "grad_norm": 0.6914794445037842,
316
+ "learning_rate": 2.99068199054963e-05,
317
+ "loss": 1.0683,
318
+ "step": 220
319
+ },
320
+ {
321
+ "epoch": 0.25876940770557794,
322
+ "grad_norm": 0.6654996871948242,
323
+ "learning_rate": 2.9895943051409337e-05,
324
+ "loss": 1.024,
325
+ "step": 225
326
+ },
327
+ {
328
+ "epoch": 0.2645198389879241,
329
+ "grad_norm": 0.7781885266304016,
330
+ "learning_rate": 2.9884468125624985e-05,
331
+ "loss": 0.9651,
332
+ "step": 230
333
+ },
334
+ {
335
+ "epoch": 0.2702702702702703,
336
+ "grad_norm": 0.7528558969497681,
337
+ "learning_rate": 2.987239558886119e-05,
338
+ "loss": 1.0042,
339
+ "step": 235
340
+ },
341
+ {
342
+ "epoch": 0.27602070155261643,
343
+ "grad_norm": 0.6547510027885437,
344
+ "learning_rate": 2.985972592582997e-05,
345
+ "loss": 1.1114,
346
+ "step": 240
347
+ },
348
+ {
349
+ "epoch": 0.28177113283496263,
350
+ "grad_norm": 0.8212192058563232,
351
+ "learning_rate": 2.984645964521794e-05,
352
+ "loss": 1.0679,
353
+ "step": 245
354
+ },
355
+ {
356
+ "epoch": 0.2875215641173088,
357
+ "grad_norm": 0.7751755714416504,
358
+ "learning_rate": 2.9832597279665893e-05,
359
+ "loss": 1.011,
360
+ "step": 250
361
+ },
362
+ {
363
+ "epoch": 0.293271995399655,
364
+ "grad_norm": 0.7902916073799133,
365
+ "learning_rate": 2.9818139385747422e-05,
366
+ "loss": 0.9789,
367
+ "step": 255
368
+ },
369
+ {
370
+ "epoch": 0.29902242668200113,
371
+ "grad_norm": 0.9229731559753418,
372
+ "learning_rate": 2.9803086543946565e-05,
373
+ "loss": 0.9941,
374
+ "step": 260
375
+ },
376
+ {
377
+ "epoch": 0.30477285796434733,
378
+ "grad_norm": 0.8640813231468201,
379
+ "learning_rate": 2.9787439358634494e-05,
380
+ "loss": 0.9888,
381
+ "step": 265
382
+ },
383
+ {
384
+ "epoch": 0.3105232892466935,
385
+ "grad_norm": 0.7600303292274475,
386
+ "learning_rate": 2.977119845804527e-05,
387
+ "loss": 0.9897,
388
+ "step": 270
389
+ },
390
+ {
391
+ "epoch": 0.3162737205290397,
392
+ "grad_norm": 0.9156889915466309,
393
+ "learning_rate": 2.9754364494250595e-05,
394
+ "loss": 0.9431,
395
+ "step": 275
396
+ },
397
+ {
398
+ "epoch": 0.32202415181138583,
399
+ "grad_norm": 0.8935381770133972,
400
+ "learning_rate": 2.973693814313364e-05,
401
+ "loss": 0.9696,
402
+ "step": 280
403
+ },
404
+ {
405
+ "epoch": 0.32777458309373203,
406
+ "grad_norm": 0.881489098072052,
407
+ "learning_rate": 2.9718920104361926e-05,
408
+ "loss": 0.9329,
409
+ "step": 285
410
+ },
411
+ {
412
+ "epoch": 0.33352501437607823,
413
+ "grad_norm": 0.8638465404510498,
414
+ "learning_rate": 2.97003111013592e-05,
415
+ "loss": 0.9708,
416
+ "step": 290
417
+ },
418
+ {
419
+ "epoch": 0.3392754456584244,
420
+ "grad_norm": 0.8255776166915894,
421
+ "learning_rate": 2.9681111881276405e-05,
422
+ "loss": 0.9486,
423
+ "step": 295
424
+ },
425
+ {
426
+ "epoch": 0.3450258769407706,
427
+ "grad_norm": 0.9097798466682434,
428
+ "learning_rate": 2.9661323214961705e-05,
429
+ "loss": 0.9508,
430
+ "step": 300
431
+ },
432
+ {
433
+ "epoch": 0.35077630822311673,
434
+ "grad_norm": 0.8574341535568237,
435
+ "learning_rate": 2.9640945896929496e-05,
436
+ "loss": 0.971,
437
+ "step": 305
438
+ },
439
+ {
440
+ "epoch": 0.35652673950546293,
441
+ "grad_norm": 0.9147903323173523,
442
+ "learning_rate": 2.9619980745328522e-05,
443
+ "loss": 0.9663,
444
+ "step": 310
445
+ },
446
+ {
447
+ "epoch": 0.3622771707878091,
448
+ "grad_norm": 0.9136379957199097,
449
+ "learning_rate": 2.9598428601909037e-05,
450
+ "loss": 0.8944,
451
+ "step": 315
452
+ },
453
+ {
454
+ "epoch": 0.3680276020701553,
455
+ "grad_norm": 0.8256983160972595,
456
+ "learning_rate": 2.9576290331988996e-05,
457
+ "loss": 0.923,
458
+ "step": 320
459
+ },
460
+ {
461
+ "epoch": 0.3737780333525014,
462
+ "grad_norm": 0.9808917045593262,
463
+ "learning_rate": 2.9553566824419324e-05,
464
+ "loss": 0.9161,
465
+ "step": 325
466
+ },
467
+ {
468
+ "epoch": 0.37952846463484763,
469
+ "grad_norm": 0.8776918649673462,
470
+ "learning_rate": 2.9530258991548212e-05,
471
+ "loss": 0.8949,
472
+ "step": 330
473
+ },
474
+ {
475
+ "epoch": 0.3852788959171938,
476
+ "grad_norm": 0.8788167834281921,
477
+ "learning_rate": 2.9506367769184496e-05,
478
+ "loss": 0.9271,
479
+ "step": 335
480
+ },
481
+ {
482
+ "epoch": 0.39102932719954,
483
+ "grad_norm": 0.8725468516349792,
484
+ "learning_rate": 2.9481894116560082e-05,
485
+ "loss": 0.8883,
486
+ "step": 340
487
+ },
488
+ {
489
+ "epoch": 0.3967797584818861,
490
+ "grad_norm": 0.9964123368263245,
491
+ "learning_rate": 2.945683901629145e-05,
492
+ "loss": 0.8135,
493
+ "step": 345
494
+ },
495
+ {
496
+ "epoch": 0.4025301897642323,
497
+ "grad_norm": 0.8448891639709473,
498
+ "learning_rate": 2.9431203474340174e-05,
499
+ "loss": 0.8551,
500
+ "step": 350
501
+ },
502
+ {
503
+ "epoch": 0.4082806210465785,
504
+ "grad_norm": 1.0148142576217651,
505
+ "learning_rate": 2.9404988519972538e-05,
506
+ "loss": 0.8397,
507
+ "step": 355
508
+ },
509
+ {
510
+ "epoch": 0.4140310523289247,
511
+ "grad_norm": 1.0716159343719482,
512
+ "learning_rate": 2.937819520571824e-05,
513
+ "loss": 0.8284,
514
+ "step": 360
515
+ },
516
+ {
517
+ "epoch": 0.4197814836112708,
518
+ "grad_norm": 1.0697191953659058,
519
+ "learning_rate": 2.9350824607328087e-05,
520
+ "loss": 0.8275,
521
+ "step": 365
522
+ },
523
+ {
524
+ "epoch": 0.425531914893617,
525
+ "grad_norm": 0.9596944451332092,
526
+ "learning_rate": 2.9322877823730848e-05,
527
+ "loss": 0.8167,
528
+ "step": 370
529
+ },
530
+ {
531
+ "epoch": 0.43128234617596317,
532
+ "grad_norm": 1.0612306594848633,
533
+ "learning_rate": 2.9294355976989098e-05,
534
+ "loss": 0.8142,
535
+ "step": 375
536
+ },
537
+ {
538
+ "epoch": 0.4370327774583094,
539
+ "grad_norm": 0.8996721506118774,
540
+ "learning_rate": 2.926526021225419e-05,
541
+ "loss": 0.7927,
542
+ "step": 380
543
+ },
544
+ {
545
+ "epoch": 0.4427832087406556,
546
+ "grad_norm": 1.0055418014526367,
547
+ "learning_rate": 2.9235591697720274e-05,
548
+ "loss": 0.8216,
549
+ "step": 385
550
+ },
551
+ {
552
+ "epoch": 0.4485336400230017,
553
+ "grad_norm": 0.9926002025604248,
554
+ "learning_rate": 2.9205351624577365e-05,
555
+ "loss": 0.8032,
556
+ "step": 390
557
+ },
558
+ {
559
+ "epoch": 0.4542840713053479,
560
+ "grad_norm": 0.9696928262710571,
561
+ "learning_rate": 2.9174541206963555e-05,
562
+ "loss": 0.8316,
563
+ "step": 395
564
+ },
565
+ {
566
+ "epoch": 0.46003450258769407,
567
+ "grad_norm": 0.9355582594871521,
568
+ "learning_rate": 2.9143161681916264e-05,
569
+ "loss": 0.8133,
570
+ "step": 400
571
+ },
572
+ {
573
+ "epoch": 0.4657849338700403,
574
+ "grad_norm": 1.004063606262207,
575
+ "learning_rate": 2.9111214309322533e-05,
576
+ "loss": 0.8282,
577
+ "step": 405
578
+ },
579
+ {
580
+ "epoch": 0.4715353651523864,
581
+ "grad_norm": 1.0633341073989868,
582
+ "learning_rate": 2.907870037186849e-05,
583
+ "loss": 0.7532,
584
+ "step": 410
585
+ },
586
+ {
587
+ "epoch": 0.4772857964347326,
588
+ "grad_norm": 1.1237492561340332,
589
+ "learning_rate": 2.9045621174987807e-05,
590
+ "loss": 0.7559,
591
+ "step": 415
592
+ },
593
+ {
594
+ "epoch": 0.48303622771707877,
595
+ "grad_norm": 0.9814444780349731,
596
+ "learning_rate": 2.9011978046809317e-05,
597
+ "loss": 0.7502,
598
+ "step": 420
599
+ },
600
+ {
601
+ "epoch": 0.48878665899942497,
602
+ "grad_norm": 1.0093247890472412,
603
+ "learning_rate": 2.8977772338103683e-05,
604
+ "loss": 0.7011,
605
+ "step": 425
606
+ },
607
+ {
608
+ "epoch": 0.4945370902817711,
609
+ "grad_norm": 1.0123990774154663,
610
+ "learning_rate": 2.8943005422229144e-05,
611
+ "loss": 0.7696,
612
+ "step": 430
613
+ },
614
+ {
615
+ "epoch": 0.5002875215641173,
616
+ "grad_norm": 1.0406811237335205,
617
+ "learning_rate": 2.8907678695076406e-05,
618
+ "loss": 0.7176,
619
+ "step": 435
620
+ },
621
+ {
622
+ "epoch": 0.5060379528464635,
623
+ "grad_norm": 0.9639581441879272,
624
+ "learning_rate": 2.8871793575012576e-05,
625
+ "loss": 0.7632,
626
+ "step": 440
627
+ },
628
+ {
629
+ "epoch": 0.5117883841288097,
630
+ "grad_norm": 1.0470564365386963,
631
+ "learning_rate": 2.8835351502824224e-05,
632
+ "loss": 0.7436,
633
+ "step": 445
634
+ },
635
+ {
636
+ "epoch": 0.5175388154111559,
637
+ "grad_norm": 1.0286060571670532,
638
+ "learning_rate": 2.8798353941659528e-05,
639
+ "loss": 0.7367,
640
+ "step": 450
641
+ },
642
+ {
643
+ "epoch": 0.523289246693502,
644
+ "grad_norm": 0.9721962213516235,
645
+ "learning_rate": 2.8760802376969536e-05,
646
+ "loss": 0.7311,
647
+ "step": 455
648
+ },
649
+ {
650
+ "epoch": 0.5290396779758482,
651
+ "grad_norm": 1.024461269378662,
652
+ "learning_rate": 2.872269831644852e-05,
653
+ "loss": 0.7024,
654
+ "step": 460
655
+ },
656
+ {
657
+ "epoch": 0.5347901092581944,
658
+ "grad_norm": 1.0587124824523926,
659
+ "learning_rate": 2.8684043289973452e-05,
660
+ "loss": 0.7107,
661
+ "step": 465
662
+ },
663
+ {
664
+ "epoch": 0.5405405405405406,
665
+ "grad_norm": 0.9372551441192627,
666
+ "learning_rate": 2.8644838849542558e-05,
667
+ "loss": 0.6938,
668
+ "step": 470
669
+ },
670
+ {
671
+ "epoch": 0.5462909718228867,
672
+ "grad_norm": 1.0846896171569824,
673
+ "learning_rate": 2.860508656921303e-05,
674
+ "loss": 0.7824,
675
+ "step": 475
676
+ },
677
+ {
678
+ "epoch": 0.5520414031052329,
679
+ "grad_norm": 1.0523967742919922,
680
+ "learning_rate": 2.856478804503781e-05,
681
+ "loss": 0.6784,
682
+ "step": 480
683
+ },
684
+ {
685
+ "epoch": 0.5577918343875791,
686
+ "grad_norm": 1.0833064317703247,
687
+ "learning_rate": 2.8523944895001518e-05,
688
+ "loss": 0.7134,
689
+ "step": 485
690
+ },
691
+ {
692
+ "epoch": 0.5635422656699253,
693
+ "grad_norm": 1.0173583030700684,
694
+ "learning_rate": 2.8482558758955486e-05,
695
+ "loss": 0.6856,
696
+ "step": 490
697
+ },
698
+ {
699
+ "epoch": 0.5692926969522715,
700
+ "grad_norm": 1.0593191385269165,
701
+ "learning_rate": 2.8440631298551926e-05,
702
+ "loss": 0.6579,
703
+ "step": 495
704
+ },
705
+ {
706
+ "epoch": 0.5750431282346176,
707
+ "grad_norm": 1.0540302991867065,
708
+ "learning_rate": 2.83981641971772e-05,
709
+ "loss": 0.6722,
710
+ "step": 500
711
+ },
712
+ {
713
+ "epoch": 0.5807935595169638,
714
+ "grad_norm": 1.1252204179763794,
715
+ "learning_rate": 2.8355159159884235e-05,
716
+ "loss": 0.6558,
717
+ "step": 505
718
+ },
719
+ {
720
+ "epoch": 0.58654399079931,
721
+ "grad_norm": 1.0981403589248657,
722
+ "learning_rate": 2.8311617913324082e-05,
723
+ "loss": 0.6876,
724
+ "step": 510
725
+ },
726
+ {
727
+ "epoch": 0.5922944220816562,
728
+ "grad_norm": 1.1801058053970337,
729
+ "learning_rate": 2.8267542205676576e-05,
730
+ "loss": 0.717,
731
+ "step": 515
732
+ },
733
+ {
734
+ "epoch": 0.5980448533640023,
735
+ "grad_norm": 1.2502609491348267,
736
+ "learning_rate": 2.8222933806580145e-05,
737
+ "loss": 0.6395,
738
+ "step": 520
739
+ },
740
+ {
741
+ "epoch": 0.6037952846463485,
742
+ "grad_norm": 1.0796862840652466,
743
+ "learning_rate": 2.817779450706077e-05,
744
+ "loss": 0.7038,
745
+ "step": 525
746
+ },
747
+ {
748
+ "epoch": 0.6095457159286947,
749
+ "grad_norm": 1.0384612083435059,
750
+ "learning_rate": 2.813212611946006e-05,
751
+ "loss": 0.6366,
752
+ "step": 530
753
+ },
754
+ {
755
+ "epoch": 0.6152961472110409,
756
+ "grad_norm": 1.0874197483062744,
757
+ "learning_rate": 2.8085930477362504e-05,
758
+ "loss": 0.6432,
759
+ "step": 535
760
+ },
761
+ {
762
+ "epoch": 0.621046578493387,
763
+ "grad_norm": 1.0653762817382812,
764
+ "learning_rate": 2.8039209435521843e-05,
765
+ "loss": 0.6724,
766
+ "step": 540
767
+ },
768
+ {
769
+ "epoch": 0.6267970097757332,
770
+ "grad_norm": 1.1459101438522339,
771
+ "learning_rate": 2.7991964869786604e-05,
772
+ "loss": 0.6341,
773
+ "step": 545
774
+ },
775
+ {
776
+ "epoch": 0.6325474410580794,
777
+ "grad_norm": 1.1049786806106567,
778
+ "learning_rate": 2.7944198677024772e-05,
779
+ "loss": 0.5834,
780
+ "step": 550
781
+ },
782
+ {
783
+ "epoch": 0.6382978723404256,
784
+ "grad_norm": 1.0369532108306885,
785
+ "learning_rate": 2.7895912775047664e-05,
786
+ "loss": 0.7076,
787
+ "step": 555
788
+ },
789
+ {
790
+ "epoch": 0.6440483036227717,
791
+ "grad_norm": 1.1675684452056885,
792
+ "learning_rate": 2.7847109102532885e-05,
793
+ "loss": 0.6371,
794
+ "step": 560
795
+ },
796
+ {
797
+ "epoch": 0.6497987349051179,
798
+ "grad_norm": 1.0560969114303589,
799
+ "learning_rate": 2.779778961894653e-05,
800
+ "loss": 0.6627,
801
+ "step": 565
802
+ },
803
+ {
804
+ "epoch": 0.6555491661874641,
805
+ "grad_norm": 1.1536891460418701,
806
+ "learning_rate": 2.7747956304464487e-05,
807
+ "loss": 0.6515,
808
+ "step": 570
809
+ },
810
+ {
811
+ "epoch": 0.6612995974698103,
812
+ "grad_norm": 1.1131980419158936,
813
+ "learning_rate": 2.769761115989294e-05,
814
+ "loss": 0.6334,
815
+ "step": 575
816
+ },
817
+ {
818
+ "epoch": 0.6670500287521565,
819
+ "grad_norm": 1.0865933895111084,
820
+ "learning_rate": 2.7646756206588033e-05,
821
+ "loss": 0.5941,
822
+ "step": 580
823
+ },
824
+ {
825
+ "epoch": 0.6728004600345026,
826
+ "grad_norm": 1.0598678588867188,
827
+ "learning_rate": 2.7595393486374722e-05,
828
+ "loss": 0.6197,
829
+ "step": 585
830
+ },
831
+ {
832
+ "epoch": 0.6785508913168488,
833
+ "grad_norm": 1.0987083911895752,
834
+ "learning_rate": 2.7543525061464793e-05,
835
+ "loss": 0.623,
836
+ "step": 590
837
+ },
838
+ {
839
+ "epoch": 0.684301322599195,
840
+ "grad_norm": 1.0570906400680542,
841
+ "learning_rate": 2.7491153014374048e-05,
842
+ "loss": 0.5829,
843
+ "step": 595
844
+ },
845
+ {
846
+ "epoch": 0.6900517538815412,
847
+ "grad_norm": 1.22000253200531,
848
+ "learning_rate": 2.743827944783873e-05,
849
+ "loss": 0.5714,
850
+ "step": 600
851
+ },
852
+ {
853
+ "epoch": 0.6958021851638873,
854
+ "grad_norm": 1.021080493927002,
855
+ "learning_rate": 2.738490648473104e-05,
856
+ "loss": 0.5258,
857
+ "step": 605
858
+ },
859
+ {
860
+ "epoch": 0.7015526164462335,
861
+ "grad_norm": 1.1696360111236572,
862
+ "learning_rate": 2.7331036267973972e-05,
863
+ "loss": 0.6207,
864
+ "step": 610
865
+ },
866
+ {
867
+ "epoch": 0.7073030477285797,
868
+ "grad_norm": 1.077847957611084,
869
+ "learning_rate": 2.7276670960455213e-05,
870
+ "loss": 0.5939,
871
+ "step": 615
872
+ },
873
+ {
874
+ "epoch": 0.7130534790109259,
875
+ "grad_norm": 1.2236453294754028,
876
+ "learning_rate": 2.7221812744940348e-05,
877
+ "loss": 0.5737,
878
+ "step": 620
879
+ },
880
+ {
881
+ "epoch": 0.718803910293272,
882
+ "grad_norm": 1.211671233177185,
883
+ "learning_rate": 2.716646382398519e-05,
884
+ "loss": 0.5634,
885
+ "step": 625
886
+ },
887
+ {
888
+ "epoch": 0.7245543415756182,
889
+ "grad_norm": 1.200198769569397,
890
+ "learning_rate": 2.711062641984738e-05,
891
+ "loss": 0.5522,
892
+ "step": 630
893
+ },
894
+ {
895
+ "epoch": 0.7303047728579644,
896
+ "grad_norm": 1.137845516204834,
897
+ "learning_rate": 2.705430277439712e-05,
898
+ "loss": 0.5475,
899
+ "step": 635
900
+ },
901
+ {
902
+ "epoch": 0.7360552041403106,
903
+ "grad_norm": 1.1840213537216187,
904
+ "learning_rate": 2.6997495149027214e-05,
905
+ "loss": 0.5254,
906
+ "step": 640
907
+ },
908
+ {
909
+ "epoch": 0.7418056354226567,
910
+ "grad_norm": 1.0524778366088867,
911
+ "learning_rate": 2.6940205824562227e-05,
912
+ "loss": 0.5876,
913
+ "step": 645
914
+ },
915
+ {
916
+ "epoch": 0.7475560667050029,
917
+ "grad_norm": 1.1802597045898438,
918
+ "learning_rate": 2.688243710116694e-05,
919
+ "loss": 0.5452,
920
+ "step": 650
921
+ },
922
+ {
923
+ "epoch": 0.753306497987349,
924
+ "grad_norm": 1.145533561706543,
925
+ "learning_rate": 2.682419129825398e-05,
926
+ "loss": 0.5845,
927
+ "step": 655
928
+ },
929
+ {
930
+ "epoch": 0.7590569292696953,
931
+ "grad_norm": 1.0609757900238037,
932
+ "learning_rate": 2.676547075439071e-05,
933
+ "loss": 0.5739,
934
+ "step": 660
935
+ },
936
+ {
937
+ "epoch": 0.7648073605520413,
938
+ "grad_norm": 1.1946995258331299,
939
+ "learning_rate": 2.670627782720532e-05,
940
+ "loss": 0.5655,
941
+ "step": 665
942
+ },
943
+ {
944
+ "epoch": 0.7705577918343876,
945
+ "grad_norm": 1.0333503484725952,
946
+ "learning_rate": 2.6646614893292193e-05,
947
+ "loss": 0.5498,
948
+ "step": 670
949
+ },
950
+ {
951
+ "epoch": 0.7763082231167338,
952
+ "grad_norm": 1.2005255222320557,
953
+ "learning_rate": 2.6586484348116447e-05,
954
+ "loss": 0.525,
955
+ "step": 675
956
+ },
957
+ {
958
+ "epoch": 0.78205865439908,
959
+ "grad_norm": 1.1393511295318604,
960
+ "learning_rate": 2.65258886059178e-05,
961
+ "loss": 0.5223,
962
+ "step": 680
963
+ },
964
+ {
965
+ "epoch": 0.7878090856814262,
966
+ "grad_norm": 1.1966758966445923,
967
+ "learning_rate": 2.6464830099613605e-05,
968
+ "loss": 0.5351,
969
+ "step": 685
970
+ },
971
+ {
972
+ "epoch": 0.7935595169637722,
973
+ "grad_norm": 1.1263506412506104,
974
+ "learning_rate": 2.6403311280701184e-05,
975
+ "loss": 0.5264,
976
+ "step": 690
977
+ },
978
+ {
979
+ "epoch": 0.7993099482461185,
980
+ "grad_norm": 1.0516929626464844,
981
+ "learning_rate": 2.6341334619159398e-05,
982
+ "loss": 0.5102,
983
+ "step": 695
984
+ },
985
+ {
986
+ "epoch": 0.8050603795284647,
987
+ "grad_norm": 1.323657751083374,
988
+ "learning_rate": 2.627890260334948e-05,
989
+ "loss": 0.5004,
990
+ "step": 700
991
+ },
992
+ {
993
+ "epoch": 0.8108108108108109,
994
+ "grad_norm": 1.2010657787322998,
995
+ "learning_rate": 2.6216017739915115e-05,
996
+ "loss": 0.4761,
997
+ "step": 705
998
+ },
999
+ {
1000
+ "epoch": 0.816561242093157,
1001
+ "grad_norm": 1.2089436054229736,
1002
+ "learning_rate": 2.6152682553681814e-05,
1003
+ "loss": 0.5026,
1004
+ "step": 710
1005
+ },
1006
+ {
1007
+ "epoch": 0.8223116733755031,
1008
+ "grad_norm": 1.1619064807891846,
1009
+ "learning_rate": 2.6088899587555538e-05,
1010
+ "loss": 0.4623,
1011
+ "step": 715
1012
+ },
1013
+ {
1014
+ "epoch": 0.8280621046578494,
1015
+ "grad_norm": 1.0324089527130127,
1016
+ "learning_rate": 2.6024671402420595e-05,
1017
+ "loss": 0.4944,
1018
+ "step": 720
1019
+ },
1020
+ {
1021
+ "epoch": 0.8338125359401956,
1022
+ "grad_norm": 1.1487537622451782,
1023
+ "learning_rate": 2.5960000577036823e-05,
1024
+ "loss": 0.5168,
1025
+ "step": 725
1026
+ },
1027
+ {
1028
+ "epoch": 0.8395629672225416,
1029
+ "grad_norm": 1.1711839437484741,
1030
+ "learning_rate": 2.5894889707936058e-05,
1031
+ "loss": 0.4625,
1032
+ "step": 730
1033
+ },
1034
+ {
1035
+ "epoch": 0.8453133985048878,
1036
+ "grad_norm": 1.18077552318573,
1037
+ "learning_rate": 2.5829341409317866e-05,
1038
+ "loss": 0.471,
1039
+ "step": 735
1040
+ },
1041
+ {
1042
+ "epoch": 0.851063829787234,
1043
+ "grad_norm": 1.0369230508804321,
1044
+ "learning_rate": 2.5763358312944616e-05,
1045
+ "loss": 0.4837,
1046
+ "step": 740
1047
+ },
1048
+ {
1049
+ "epoch": 0.8568142610695803,
1050
+ "grad_norm": 1.2262332439422607,
1051
+ "learning_rate": 2.569694306803578e-05,
1052
+ "loss": 0.4912,
1053
+ "step": 745
1054
+ },
1055
+ {
1056
+ "epoch": 0.8625646923519263,
1057
+ "grad_norm": 1.271316409111023,
1058
+ "learning_rate": 2.5630098341161583e-05,
1059
+ "loss": 0.495,
1060
+ "step": 750
1061
+ },
1062
+ {
1063
+ "epoch": 0.8683151236342725,
1064
+ "grad_norm": 1.1962330341339111,
1065
+ "learning_rate": 2.5562826816135943e-05,
1066
+ "loss": 0.4667,
1067
+ "step": 755
1068
+ },
1069
+ {
1070
+ "epoch": 0.8740655549166187,
1071
+ "grad_norm": 1.0178544521331787,
1072
+ "learning_rate": 2.5495131193908716e-05,
1073
+ "loss": 0.4506,
1074
+ "step": 760
1075
+ },
1076
+ {
1077
+ "epoch": 0.879815986198965,
1078
+ "grad_norm": 1.1843115091323853,
1079
+ "learning_rate": 2.5427014192457233e-05,
1080
+ "loss": 0.4397,
1081
+ "step": 765
1082
+ },
1083
+ {
1084
+ "epoch": 0.8855664174813112,
1085
+ "grad_norm": 1.156090259552002,
1086
+ "learning_rate": 2.535847854667721e-05,
1087
+ "loss": 0.43,
1088
+ "step": 770
1089
+ },
1090
+ {
1091
+ "epoch": 0.8913168487636572,
1092
+ "grad_norm": 1.1655112504959106,
1093
+ "learning_rate": 2.5289527008272903e-05,
1094
+ "loss": 0.447,
1095
+ "step": 775
1096
+ },
1097
+ {
1098
+ "epoch": 0.8970672800460034,
1099
+ "grad_norm": 1.2115625143051147,
1100
+ "learning_rate": 2.5220162345646657e-05,
1101
+ "loss": 0.4237,
1102
+ "step": 780
1103
+ },
1104
+ {
1105
+ "epoch": 0.9028177113283496,
1106
+ "grad_norm": 1.2119576930999756,
1107
+ "learning_rate": 2.5150387343787736e-05,
1108
+ "loss": 0.4308,
1109
+ "step": 785
1110
+ },
1111
+ {
1112
+ "epoch": 0.9085681426106959,
1113
+ "grad_norm": 1.113875150680542,
1114
+ "learning_rate": 2.5080204804160515e-05,
1115
+ "loss": 0.4585,
1116
+ "step": 790
1117
+ },
1118
+ {
1119
+ "epoch": 0.9143185738930419,
1120
+ "grad_norm": 1.1130765676498413,
1121
+ "learning_rate": 2.500961754459201e-05,
1122
+ "loss": 0.4495,
1123
+ "step": 795
1124
+ },
1125
+ {
1126
+ "epoch": 0.9200690051753881,
1127
+ "grad_norm": 1.1530660390853882,
1128
+ "learning_rate": 2.49386283991587e-05,
1129
+ "loss": 0.433,
1130
+ "step": 800
1131
+ },
1132
+ {
1133
+ "epoch": 0.9258194364577343,
1134
+ "grad_norm": 1.182665228843689,
1135
+ "learning_rate": 2.4867240218072807e-05,
1136
+ "loss": 0.3933,
1137
+ "step": 805
1138
+ },
1139
+ {
1140
+ "epoch": 0.9315698677400805,
1141
+ "grad_norm": 1.167338252067566,
1142
+ "learning_rate": 2.4795455867567808e-05,
1143
+ "loss": 0.4505,
1144
+ "step": 810
1145
+ },
1146
+ {
1147
+ "epoch": 0.9373202990224266,
1148
+ "grad_norm": 1.4360686540603638,
1149
+ "learning_rate": 2.4723278229783363e-05,
1150
+ "loss": 0.4121,
1151
+ "step": 815
1152
+ },
1153
+ {
1154
+ "epoch": 0.9430707303047728,
1155
+ "grad_norm": 1.1436094045639038,
1156
+ "learning_rate": 2.4650710202649612e-05,
1157
+ "loss": 0.4451,
1158
+ "step": 820
1159
+ },
1160
+ {
1161
+ "epoch": 0.948821161587119,
1162
+ "grad_norm": 1.248976707458496,
1163
+ "learning_rate": 2.4577754699770826e-05,
1164
+ "loss": 0.4232,
1165
+ "step": 825
1166
+ },
1167
+ {
1168
+ "epoch": 0.9545715928694652,
1169
+ "grad_norm": 1.1697114706039429,
1170
+ "learning_rate": 2.4504414650308404e-05,
1171
+ "loss": 0.407,
1172
+ "step": 830
1173
+ },
1174
+ {
1175
+ "epoch": 0.9603220241518113,
1176
+ "grad_norm": 1.2601184844970703,
1177
+ "learning_rate": 2.4430692998863276e-05,
1178
+ "loss": 0.4144,
1179
+ "step": 835
1180
+ },
1181
+ {
1182
+ "epoch": 0.9660724554341575,
1183
+ "grad_norm": 1.260187029838562,
1184
+ "learning_rate": 2.4356592705357693e-05,
1185
+ "loss": 0.3683,
1186
+ "step": 840
1187
+ },
1188
+ {
1189
+ "epoch": 0.9718228867165037,
1190
+ "grad_norm": 1.1422863006591797,
1191
+ "learning_rate": 2.4282116744916373e-05,
1192
+ "loss": 0.3994,
1193
+ "step": 845
1194
+ },
1195
+ {
1196
+ "epoch": 0.9775733179988499,
1197
+ "grad_norm": 1.1310383081436157,
1198
+ "learning_rate": 2.4207268107747043e-05,
1199
+ "loss": 0.4197,
1200
+ "step": 850
1201
+ },
1202
+ {
1203
+ "epoch": 0.983323749281196,
1204
+ "grad_norm": 1.162474274635315,
1205
+ "learning_rate": 2.4132049799020395e-05,
1206
+ "loss": 0.3824,
1207
+ "step": 855
1208
+ },
1209
+ {
1210
+ "epoch": 0.9890741805635422,
1211
+ "grad_norm": 1.2372602224349976,
1212
+ "learning_rate": 2.405646483874942e-05,
1213
+ "loss": 0.4174,
1214
+ "step": 860
1215
+ },
1216
+ {
1217
+ "epoch": 0.9948246118458884,
1218
+ "grad_norm": 1.191551923751831,
1219
+ "learning_rate": 2.3980516261668168e-05,
1220
+ "loss": 0.404,
1221
+ "step": 865
1222
+ },
1223
+ {
1224
+ "epoch": 1.0,
1225
+ "grad_norm": 1.7543858289718628,
1226
+ "learning_rate": 2.3904207117109887e-05,
1227
+ "loss": 0.4104,
1228
+ "step": 870
1229
+ },
1230
+ {
1231
+ "epoch": 1.0057504312823462,
1232
+ "grad_norm": 1.182187795639038,
1233
+ "learning_rate": 2.3827540468884593e-05,
1234
+ "loss": 0.3283,
1235
+ "step": 875
1236
+ },
1237
+ {
1238
+ "epoch": 1.0115008625646924,
1239
+ "grad_norm": 1.314508318901062,
1240
+ "learning_rate": 2.3750519395156073e-05,
1241
+ "loss": 0.3592,
1242
+ "step": 880
1243
+ },
1244
+ {
1245
+ "epoch": 1.0172512938470386,
1246
+ "grad_norm": 1.0611480474472046,
1247
+ "learning_rate": 2.3673146988318288e-05,
1248
+ "loss": 0.3343,
1249
+ "step": 885
1250
+ },
1251
+ {
1252
+ "epoch": 1.0230017251293848,
1253
+ "grad_norm": 1.1292250156402588,
1254
+ "learning_rate": 2.359542635487121e-05,
1255
+ "loss": 0.3325,
1256
+ "step": 890
1257
+ },
1258
+ {
1259
+ "epoch": 1.0287521564117308,
1260
+ "grad_norm": 1.1909079551696777,
1261
+ "learning_rate": 2.35173606152961e-05,
1262
+ "loss": 0.3272,
1263
+ "step": 895
1264
+ },
1265
+ {
1266
+ "epoch": 1.034502587694077,
1267
+ "grad_norm": 1.2812501192092896,
1268
+ "learning_rate": 2.3438952903930223e-05,
1269
+ "loss": 0.3227,
1270
+ "step": 900
1271
+ },
1272
+ {
1273
+ "epoch": 1.0402530189764232,
1274
+ "grad_norm": 1.1142762899398804,
1275
+ "learning_rate": 2.3360206368840995e-05,
1276
+ "loss": 0.3121,
1277
+ "step": 905
1278
+ },
1279
+ {
1280
+ "epoch": 1.0460034502587694,
1281
+ "grad_norm": 1.2537274360656738,
1282
+ "learning_rate": 2.3281124171699605e-05,
1283
+ "loss": 0.3038,
1284
+ "step": 910
1285
+ },
1286
+ {
1287
+ "epoch": 1.0517538815411156,
1288
+ "grad_norm": 1.1135908365249634,
1289
+ "learning_rate": 2.320170948765406e-05,
1290
+ "loss": 0.3161,
1291
+ "step": 915
1292
+ },
1293
+ {
1294
+ "epoch": 1.0575043128234618,
1295
+ "grad_norm": 1.091538667678833,
1296
+ "learning_rate": 2.31219655052017e-05,
1297
+ "loss": 0.3339,
1298
+ "step": 920
1299
+ },
1300
+ {
1301
+ "epoch": 1.063254744105808,
1302
+ "grad_norm": 1.1415507793426514,
1303
+ "learning_rate": 2.3041895426061194e-05,
1304
+ "loss": 0.3029,
1305
+ "step": 925
1306
+ },
1307
+ {
1308
+ "epoch": 1.0690051753881542,
1309
+ "grad_norm": 1.200616478919983,
1310
+ "learning_rate": 2.296150246504399e-05,
1311
+ "loss": 0.3457,
1312
+ "step": 930
1313
+ },
1314
+ {
1315
+ "epoch": 1.0747556066705002,
1316
+ "grad_norm": 1.153767466545105,
1317
+ "learning_rate": 2.2880789849925222e-05,
1318
+ "loss": 0.315,
1319
+ "step": 935
1320
+ },
1321
+ {
1322
+ "epoch": 1.0805060379528464,
1323
+ "grad_norm": 1.3046640157699585,
1324
+ "learning_rate": 2.2799760821314132e-05,
1325
+ "loss": 0.3153,
1326
+ "step": 940
1327
+ },
1328
+ {
1329
+ "epoch": 1.0862564692351926,
1330
+ "grad_norm": 1.1953332424163818,
1331
+ "learning_rate": 2.271841863252395e-05,
1332
+ "loss": 0.3074,
1333
+ "step": 945
1334
+ },
1335
+ {
1336
+ "epoch": 1.0920069005175388,
1337
+ "grad_norm": 1.1311490535736084,
1338
+ "learning_rate": 2.2636766549441294e-05,
1339
+ "loss": 0.3365,
1340
+ "step": 950
1341
+ },
1342
+ {
1343
+ "epoch": 1.097757331799885,
1344
+ "grad_norm": 1.219886302947998,
1345
+ "learning_rate": 2.2554807850395013e-05,
1346
+ "loss": 0.2837,
1347
+ "step": 955
1348
+ },
1349
+ {
1350
+ "epoch": 1.1035077630822312,
1351
+ "grad_norm": 1.0347853899002075,
1352
+ "learning_rate": 2.247254582602458e-05,
1353
+ "loss": 0.3086,
1354
+ "step": 960
1355
+ },
1356
+ {
1357
+ "epoch": 1.1092581943645774,
1358
+ "grad_norm": 1.1474758386611938,
1359
+ "learning_rate": 2.2389983779147987e-05,
1360
+ "loss": 0.3104,
1361
+ "step": 965
1362
+ },
1363
+ {
1364
+ "epoch": 1.1150086256469236,
1365
+ "grad_norm": 1.1311414241790771,
1366
+ "learning_rate": 2.2307125024629103e-05,
1367
+ "loss": 0.3399,
1368
+ "step": 970
1369
+ },
1370
+ {
1371
+ "epoch": 1.1207590569292698,
1372
+ "grad_norm": 1.2091407775878906,
1373
+ "learning_rate": 2.2223972889244603e-05,
1374
+ "loss": 0.2971,
1375
+ "step": 975
1376
+ },
1377
+ {
1378
+ "epoch": 1.1265094882116158,
1379
+ "grad_norm": 1.3387929201126099,
1380
+ "learning_rate": 2.2140530711550408e-05,
1381
+ "loss": 0.2929,
1382
+ "step": 980
1383
+ },
1384
+ {
1385
+ "epoch": 1.132259919493962,
1386
+ "grad_norm": 1.1228324174880981,
1387
+ "learning_rate": 2.2056801841747616e-05,
1388
+ "loss": 0.3135,
1389
+ "step": 985
1390
+ },
1391
+ {
1392
+ "epoch": 1.1380103507763082,
1393
+ "grad_norm": 1.383875846862793,
1394
+ "learning_rate": 2.197278964154801e-05,
1395
+ "loss": 0.2799,
1396
+ "step": 990
1397
+ },
1398
+ {
1399
+ "epoch": 1.1437607820586544,
1400
+ "grad_norm": 1.197532057762146,
1401
+ "learning_rate": 2.1888497484039093e-05,
1402
+ "loss": 0.3059,
1403
+ "step": 995
1404
+ },
1405
+ {
1406
+ "epoch": 1.1495112133410006,
1407
+ "grad_norm": 1.2810454368591309,
1408
+ "learning_rate": 2.180392875354861e-05,
1409
+ "loss": 0.2988,
1410
+ "step": 1000
1411
+ },
1412
+ {
1413
+ "epoch": 1.1552616446233468,
1414
+ "grad_norm": 1.2195801734924316,
1415
+ "learning_rate": 2.171908684550875e-05,
1416
+ "loss": 0.2984,
1417
+ "step": 1005
1418
+ },
1419
+ {
1420
+ "epoch": 1.161012075905693,
1421
+ "grad_norm": 1.2179973125457764,
1422
+ "learning_rate": 2.1633975166319745e-05,
1423
+ "loss": 0.2901,
1424
+ "step": 1010
1425
+ },
1426
+ {
1427
+ "epoch": 1.1667625071880392,
1428
+ "grad_norm": 1.134080410003662,
1429
+ "learning_rate": 2.1548597133213127e-05,
1430
+ "loss": 0.2984,
1431
+ "step": 1015
1432
+ },
1433
+ {
1434
+ "epoch": 1.1725129384703852,
1435
+ "grad_norm": 1.1779953241348267,
1436
+ "learning_rate": 2.146295617411456e-05,
1437
+ "loss": 0.298,
1438
+ "step": 1020
1439
+ },
1440
+ {
1441
+ "epoch": 1.1782633697527314,
1442
+ "grad_norm": 1.224240779876709,
1443
+ "learning_rate": 2.137705572750614e-05,
1444
+ "loss": 0.274,
1445
+ "step": 1025
1446
+ },
1447
+ {
1448
+ "epoch": 1.1840138010350776,
1449
+ "grad_norm": 1.2111856937408447,
1450
+ "learning_rate": 2.12908992422884e-05,
1451
+ "loss": 0.2929,
1452
+ "step": 1030
1453
+ },
1454
+ {
1455
+ "epoch": 1.1897642323174238,
1456
+ "grad_norm": 1.1539922952651978,
1457
+ "learning_rate": 2.120449017764183e-05,
1458
+ "loss": 0.2716,
1459
+ "step": 1035
1460
+ },
1461
+ {
1462
+ "epoch": 1.19551466359977,
1463
+ "grad_norm": 1.1918421983718872,
1464
+ "learning_rate": 2.1117832002887953e-05,
1465
+ "loss": 0.2905,
1466
+ "step": 1040
1467
+ },
1468
+ {
1469
+ "epoch": 1.2012650948821162,
1470
+ "grad_norm": 1.2142211198806763,
1471
+ "learning_rate": 2.1030928197350056e-05,
1472
+ "loss": 0.2818,
1473
+ "step": 1045
1474
+ },
1475
+ {
1476
+ "epoch": 1.2070155261644624,
1477
+ "grad_norm": 1.1928277015686035,
1478
+ "learning_rate": 2.0943782250213517e-05,
1479
+ "loss": 0.2784,
1480
+ "step": 1050
1481
+ },
1482
+ {
1483
+ "epoch": 1.2127659574468086,
1484
+ "grad_norm": 1.086199164390564,
1485
+ "learning_rate": 2.085639766038567e-05,
1486
+ "loss": 0.265,
1487
+ "step": 1055
1488
+ },
1489
+ {
1490
+ "epoch": 1.2185163887291548,
1491
+ "grad_norm": 1.226786494255066,
1492
+ "learning_rate": 2.0768777936355364e-05,
1493
+ "loss": 0.274,
1494
+ "step": 1060
1495
+ },
1496
+ {
1497
+ "epoch": 1.2242668200115008,
1498
+ "grad_norm": 1.0753157138824463,
1499
+ "learning_rate": 2.0680926596052062e-05,
1500
+ "loss": 0.2837,
1501
+ "step": 1065
1502
+ },
1503
+ {
1504
+ "epoch": 1.230017251293847,
1505
+ "grad_norm": 1.0061887502670288,
1506
+ "learning_rate": 2.059284716670463e-05,
1507
+ "loss": 0.259,
1508
+ "step": 1070
1509
+ },
1510
+ {
1511
+ "epoch": 1.2357676825761932,
1512
+ "grad_norm": 1.1340038776397705,
1513
+ "learning_rate": 2.0504543184699692e-05,
1514
+ "loss": 0.2639,
1515
+ "step": 1075
1516
+ },
1517
+ {
1518
+ "epoch": 1.2415181138585394,
1519
+ "grad_norm": 1.056527853012085,
1520
+ "learning_rate": 2.0416018195439657e-05,
1521
+ "loss": 0.2796,
1522
+ "step": 1080
1523
+ },
1524
+ {
1525
+ "epoch": 1.2472685451408856,
1526
+ "grad_norm": 1.130855679512024,
1527
+ "learning_rate": 2.032727575320038e-05,
1528
+ "loss": 0.2706,
1529
+ "step": 1085
1530
+ },
1531
+ {
1532
+ "epoch": 1.2530189764232318,
1533
+ "grad_norm": 1.0675405263900757,
1534
+ "learning_rate": 2.0238319420988427e-05,
1535
+ "loss": 0.2554,
1536
+ "step": 1090
1537
+ },
1538
+ {
1539
+ "epoch": 1.258769407705578,
1540
+ "grad_norm": 1.3446725606918335,
1541
+ "learning_rate": 2.0149152770398054e-05,
1542
+ "loss": 0.2431,
1543
+ "step": 1095
1544
+ },
1545
+ {
1546
+ "epoch": 1.264519838987924,
1547
+ "grad_norm": 1.1526360511779785,
1548
+ "learning_rate": 2.00597793814678e-05,
1549
+ "loss": 0.2464,
1550
+ "step": 1100
1551
+ },
1552
+ {
1553
+ "epoch": 1.2702702702702702,
1554
+ "grad_norm": 1.4780713319778442,
1555
+ "learning_rate": 1.997020284253672e-05,
1556
+ "loss": 0.2552,
1557
+ "step": 1105
1558
+ },
1559
+ {
1560
+ "epoch": 1.2760207015526164,
1561
+ "grad_norm": 1.2046481370925903,
1562
+ "learning_rate": 1.9880426750100358e-05,
1563
+ "loss": 0.264,
1564
+ "step": 1110
1565
+ },
1566
+ {
1567
+ "epoch": 1.2817711328349626,
1568
+ "grad_norm": 1.2370834350585938,
1569
+ "learning_rate": 1.9790454708666318e-05,
1570
+ "loss": 0.273,
1571
+ "step": 1115
1572
+ },
1573
+ {
1574
+ "epoch": 1.2875215641173088,
1575
+ "grad_norm": 1.1575658321380615,
1576
+ "learning_rate": 1.970029033060955e-05,
1577
+ "loss": 0.2614,
1578
+ "step": 1120
1579
+ },
1580
+ {
1581
+ "epoch": 1.293271995399655,
1582
+ "grad_norm": 1.5018378496170044,
1583
+ "learning_rate": 1.9609937236027308e-05,
1584
+ "loss": 0.2548,
1585
+ "step": 1125
1586
+ },
1587
+ {
1588
+ "epoch": 1.2990224266820012,
1589
+ "grad_norm": 1.022945761680603,
1590
+ "learning_rate": 1.9519399052593823e-05,
1591
+ "loss": 0.2101,
1592
+ "step": 1130
1593
+ },
1594
+ {
1595
+ "epoch": 1.3047728579643474,
1596
+ "grad_norm": 1.1765093803405762,
1597
+ "learning_rate": 1.9428679415414626e-05,
1598
+ "loss": 0.2383,
1599
+ "step": 1135
1600
+ },
1601
+ {
1602
+ "epoch": 1.3105232892466936,
1603
+ "grad_norm": 1.3009532690048218,
1604
+ "learning_rate": 1.9337781966880612e-05,
1605
+ "loss": 0.2269,
1606
+ "step": 1140
1607
+ },
1608
+ {
1609
+ "epoch": 1.3162737205290398,
1610
+ "grad_norm": 1.0406996011734009,
1611
+ "learning_rate": 1.924671035652181e-05,
1612
+ "loss": 0.2271,
1613
+ "step": 1145
1614
+ },
1615
+ {
1616
+ "epoch": 1.3220241518113858,
1617
+ "grad_norm": 1.118814468383789,
1618
+ "learning_rate": 1.9155468240860828e-05,
1619
+ "loss": 0.2107,
1620
+ "step": 1150
1621
+ },
1622
+ {
1623
+ "epoch": 1.327774583093732,
1624
+ "grad_norm": 1.1692357063293457,
1625
+ "learning_rate": 1.906405928326606e-05,
1626
+ "loss": 0.2368,
1627
+ "step": 1155
1628
+ },
1629
+ {
1630
+ "epoch": 1.3335250143760782,
1631
+ "grad_norm": 1.0028239488601685,
1632
+ "learning_rate": 1.8972487153804594e-05,
1633
+ "loss": 0.24,
1634
+ "step": 1160
1635
+ },
1636
+ {
1637
+ "epoch": 1.3392754456584244,
1638
+ "grad_norm": 1.1365278959274292,
1639
+ "learning_rate": 1.8880755529094883e-05,
1640
+ "loss": 0.2287,
1641
+ "step": 1165
1642
+ },
1643
+ {
1644
+ "epoch": 1.3450258769407706,
1645
+ "grad_norm": 1.106742024421692,
1646
+ "learning_rate": 1.8788868092159087e-05,
1647
+ "loss": 0.2108,
1648
+ "step": 1170
1649
+ },
1650
+ {
1651
+ "epoch": 1.3507763082231168,
1652
+ "grad_norm": 1.2229759693145752,
1653
+ "learning_rate": 1.869682853227523e-05,
1654
+ "loss": 0.2323,
1655
+ "step": 1175
1656
+ },
1657
+ {
1658
+ "epoch": 1.356526739505463,
1659
+ "grad_norm": 1.1377099752426147,
1660
+ "learning_rate": 1.860464054482908e-05,
1661
+ "loss": 0.2334,
1662
+ "step": 1180
1663
+ },
1664
+ {
1665
+ "epoch": 1.362277170787809,
1666
+ "grad_norm": 1.429868221282959,
1667
+ "learning_rate": 1.8512307831165748e-05,
1668
+ "loss": 0.2092,
1669
+ "step": 1185
1670
+ },
1671
+ {
1672
+ "epoch": 1.3680276020701552,
1673
+ "grad_norm": 1.0516825914382935,
1674
+ "learning_rate": 1.8419834098441118e-05,
1675
+ "loss": 0.2112,
1676
+ "step": 1190
1677
+ },
1678
+ {
1679
+ "epoch": 1.3737780333525014,
1680
+ "grad_norm": 1.2210971117019653,
1681
+ "learning_rate": 1.8327223059472967e-05,
1682
+ "loss": 0.2316,
1683
+ "step": 1195
1684
+ },
1685
+ {
1686
+ "epoch": 1.3795284646348476,
1687
+ "grad_norm": 1.2062695026397705,
1688
+ "learning_rate": 1.823447843259192e-05,
1689
+ "loss": 0.2437,
1690
+ "step": 1200
1691
+ },
1692
+ {
1693
+ "epoch": 1.3852788959171938,
1694
+ "grad_norm": 1.1227179765701294,
1695
+ "learning_rate": 1.8141603941492172e-05,
1696
+ "loss": 0.208,
1697
+ "step": 1205
1698
+ },
1699
+ {
1700
+ "epoch": 1.39102932719954,
1701
+ "grad_norm": 1.1238170862197876,
1702
+ "learning_rate": 1.8048603315081925e-05,
1703
+ "loss": 0.2167,
1704
+ "step": 1210
1705
+ },
1706
+ {
1707
+ "epoch": 1.3967797584818862,
1708
+ "grad_norm": 1.0642367601394653,
1709
+ "learning_rate": 1.795548028733374e-05,
1710
+ "loss": 0.2182,
1711
+ "step": 1215
1712
+ },
1713
+ {
1714
+ "epoch": 1.4025301897642324,
1715
+ "grad_norm": 1.0099825859069824,
1716
+ "learning_rate": 1.786223859713459e-05,
1717
+ "loss": 0.2037,
1718
+ "step": 1220
1719
+ },
1720
+ {
1721
+ "epoch": 1.4082806210465786,
1722
+ "grad_norm": 1.1034737825393677,
1723
+ "learning_rate": 1.7768881988135717e-05,
1724
+ "loss": 0.2042,
1725
+ "step": 1225
1726
+ },
1727
+ {
1728
+ "epoch": 1.4140310523289248,
1729
+ "grad_norm": 1.078342080116272,
1730
+ "learning_rate": 1.767541420860237e-05,
1731
+ "loss": 0.2286,
1732
+ "step": 1230
1733
+ },
1734
+ {
1735
+ "epoch": 1.4197814836112708,
1736
+ "grad_norm": 1.2092900276184082,
1737
+ "learning_rate": 1.758183901126329e-05,
1738
+ "loss": 0.2003,
1739
+ "step": 1235
1740
+ },
1741
+ {
1742
+ "epoch": 1.425531914893617,
1743
+ "grad_norm": 1.201078176498413,
1744
+ "learning_rate": 1.748816015316003e-05,
1745
+ "loss": 0.2254,
1746
+ "step": 1240
1747
+ },
1748
+ {
1749
+ "epoch": 1.4312823461759632,
1750
+ "grad_norm": 1.1587811708450317,
1751
+ "learning_rate": 1.7394381395496128e-05,
1752
+ "loss": 0.2219,
1753
+ "step": 1245
1754
+ },
1755
+ {
1756
+ "epoch": 1.4370327774583094,
1757
+ "grad_norm": 1.0706764459609985,
1758
+ "learning_rate": 1.7300506503486082e-05,
1759
+ "loss": 0.2253,
1760
+ "step": 1250
1761
+ },
1762
+ {
1763
+ "epoch": 1.4427832087406556,
1764
+ "grad_norm": 0.9957166910171509,
1765
+ "learning_rate": 1.720653924620419e-05,
1766
+ "loss": 0.2326,
1767
+ "step": 1255
1768
+ },
1769
+ {
1770
+ "epoch": 1.4485336400230018,
1771
+ "grad_norm": 1.0204159021377563,
1772
+ "learning_rate": 1.7112483396433204e-05,
1773
+ "loss": 0.1955,
1774
+ "step": 1260
1775
+ },
1776
+ {
1777
+ "epoch": 1.454284071305348,
1778
+ "grad_norm": 1.2959260940551758,
1779
+ "learning_rate": 1.7018342730512863e-05,
1780
+ "loss": 0.201,
1781
+ "step": 1265
1782
+ },
1783
+ {
1784
+ "epoch": 1.460034502587694,
1785
+ "grad_norm": 1.0318083763122559,
1786
+ "learning_rate": 1.6924121028188278e-05,
1787
+ "loss": 0.2014,
1788
+ "step": 1270
1789
+ },
1790
+ {
1791
+ "epoch": 1.4657849338700402,
1792
+ "grad_norm": 1.123473882675171,
1793
+ "learning_rate": 1.6829822072458156e-05,
1794
+ "loss": 0.2114,
1795
+ "step": 1275
1796
+ },
1797
+ {
1798
+ "epoch": 1.4715353651523864,
1799
+ "grad_norm": 1.2916359901428223,
1800
+ "learning_rate": 1.6735449649422945e-05,
1801
+ "loss": 0.1988,
1802
+ "step": 1280
1803
+ },
1804
+ {
1805
+ "epoch": 1.4772857964347326,
1806
+ "grad_norm": 1.1253008842468262,
1807
+ "learning_rate": 1.6641007548132796e-05,
1808
+ "loss": 0.2008,
1809
+ "step": 1285
1810
+ },
1811
+ {
1812
+ "epoch": 1.4830362277170788,
1813
+ "grad_norm": 1.2874746322631836,
1814
+ "learning_rate": 1.6546499560435418e-05,
1815
+ "loss": 0.2112,
1816
+ "step": 1290
1817
+ },
1818
+ {
1819
+ "epoch": 1.488786658999425,
1820
+ "grad_norm": 1.0030890703201294,
1821
+ "learning_rate": 1.6451929480823897e-05,
1822
+ "loss": 0.1919,
1823
+ "step": 1295
1824
+ },
1825
+ {
1826
+ "epoch": 1.4945370902817712,
1827
+ "grad_norm": 1.1379762887954712,
1828
+ "learning_rate": 1.6357301106284288e-05,
1829
+ "loss": 0.1976,
1830
+ "step": 1300
1831
+ },
1832
+ {
1833
+ "epoch": 1.5002875215641174,
1834
+ "grad_norm": 1.4394164085388184,
1835
+ "learning_rate": 1.6262618236143172e-05,
1836
+ "loss": 0.1952,
1837
+ "step": 1305
1838
+ },
1839
+ {
1840
+ "epoch": 1.5060379528464636,
1841
+ "grad_norm": 1.0223239660263062,
1842
+ "learning_rate": 1.6167884671915148e-05,
1843
+ "loss": 0.1803,
1844
+ "step": 1310
1845
+ },
1846
+ {
1847
+ "epoch": 1.5117883841288098,
1848
+ "grad_norm": 1.181185007095337,
1849
+ "learning_rate": 1.6073104217150172e-05,
1850
+ "loss": 0.1896,
1851
+ "step": 1315
1852
+ },
1853
+ {
1854
+ "epoch": 1.517538815411156,
1855
+ "grad_norm": 1.0288665294647217,
1856
+ "learning_rate": 1.597828067728085e-05,
1857
+ "loss": 0.1867,
1858
+ "step": 1320
1859
+ },
1860
+ {
1861
+ "epoch": 1.523289246693502,
1862
+ "grad_norm": 1.2485613822937012,
1863
+ "learning_rate": 1.5883417859469672e-05,
1864
+ "loss": 0.1808,
1865
+ "step": 1325
1866
+ },
1867
+ {
1868
+ "epoch": 1.5290396779758482,
1869
+ "grad_norm": 1.1305049657821655,
1870
+ "learning_rate": 1.578851957245611e-05,
1871
+ "loss": 0.1812,
1872
+ "step": 1330
1873
+ },
1874
+ {
1875
+ "epoch": 1.5347901092581944,
1876
+ "grad_norm": 1.053968906402588,
1877
+ "learning_rate": 1.5693589626403743e-05,
1878
+ "loss": 0.2006,
1879
+ "step": 1335
1880
+ },
1881
+ {
1882
+ "epoch": 1.5405405405405406,
1883
+ "grad_norm": 1.1373350620269775,
1884
+ "learning_rate": 1.559863183274725e-05,
1885
+ "loss": 0.1831,
1886
+ "step": 1340
1887
+ },
1888
+ {
1889
+ "epoch": 1.5462909718228866,
1890
+ "grad_norm": 1.0823122262954712,
1891
+ "learning_rate": 1.5503650004039394e-05,
1892
+ "loss": 0.1784,
1893
+ "step": 1345
1894
+ },
1895
+ {
1896
+ "epoch": 1.5520414031052328,
1897
+ "grad_norm": 1.0690486431121826,
1898
+ "learning_rate": 1.540864795379795e-05,
1899
+ "loss": 0.1952,
1900
+ "step": 1350
1901
+ },
1902
+ {
1903
+ "epoch": 1.557791834387579,
1904
+ "grad_norm": 1.1026623249053955,
1905
+ "learning_rate": 1.5313629496352574e-05,
1906
+ "loss": 0.1856,
1907
+ "step": 1355
1908
+ },
1909
+ {
1910
+ "epoch": 1.5635422656699252,
1911
+ "grad_norm": 1.0244990587234497,
1912
+ "learning_rate": 1.5218598446691678e-05,
1913
+ "loss": 0.1775,
1914
+ "step": 1360
1915
+ },
1916
+ {
1917
+ "epoch": 1.5692926969522714,
1918
+ "grad_norm": 1.0843275785446167,
1919
+ "learning_rate": 1.512355862030926e-05,
1920
+ "loss": 0.1573,
1921
+ "step": 1365
1922
+ },
1923
+ {
1924
+ "epoch": 1.5750431282346176,
1925
+ "grad_norm": 1.1366690397262573,
1926
+ "learning_rate": 1.5028513833051685e-05,
1927
+ "loss": 0.1779,
1928
+ "step": 1370
1929
+ },
1930
+ {
1931
+ "epoch": 1.5807935595169638,
1932
+ "grad_norm": 1.0362428426742554,
1933
+ "learning_rate": 1.493346790096451e-05,
1934
+ "loss": 0.1819,
1935
+ "step": 1375
1936
+ },
1937
+ {
1938
+ "epoch": 1.58654399079931,
1939
+ "grad_norm": 1.0112504959106445,
1940
+ "learning_rate": 1.483842464013925e-05,
1941
+ "loss": 0.1719,
1942
+ "step": 1380
1943
+ },
1944
+ {
1945
+ "epoch": 1.5922944220816562,
1946
+ "grad_norm": 1.048193097114563,
1947
+ "learning_rate": 1.4743387866560178e-05,
1948
+ "loss": 0.1818,
1949
+ "step": 1385
1950
+ },
1951
+ {
1952
+ "epoch": 1.5980448533640024,
1953
+ "grad_norm": 0.9836743474006653,
1954
+ "learning_rate": 1.4648361395951094e-05,
1955
+ "loss": 0.1727,
1956
+ "step": 1390
1957
+ },
1958
+ {
1959
+ "epoch": 1.6037952846463486,
1960
+ "grad_norm": 0.9633564352989197,
1961
+ "learning_rate": 1.4553349043622138e-05,
1962
+ "loss": 0.1728,
1963
+ "step": 1395
1964
+ },
1965
+ {
1966
+ "epoch": 1.6095457159286948,
1967
+ "grad_norm": 1.1259740591049194,
1968
+ "learning_rate": 1.4458354624316609e-05,
1969
+ "loss": 0.1524,
1970
+ "step": 1400
1971
+ },
1972
+ {
1973
+ "epoch": 1.615296147211041,
1974
+ "grad_norm": 0.9631626009941101,
1975
+ "learning_rate": 1.4363381952057779e-05,
1976
+ "loss": 0.1744,
1977
+ "step": 1405
1978
+ },
1979
+ {
1980
+ "epoch": 1.621046578493387,
1981
+ "grad_norm": 1.1584100723266602,
1982
+ "learning_rate": 1.426843483999579e-05,
1983
+ "loss": 0.1611,
1984
+ "step": 1410
1985
+ },
1986
+ {
1987
+ "epoch": 1.6267970097757332,
1988
+ "grad_norm": 1.1624690294265747,
1989
+ "learning_rate": 1.4173517100254534e-05,
1990
+ "loss": 0.1711,
1991
+ "step": 1415
1992
+ },
1993
+ {
1994
+ "epoch": 1.6325474410580794,
1995
+ "grad_norm": 1.2575827836990356,
1996
+ "learning_rate": 1.4078632543778618e-05,
1997
+ "loss": 0.1569,
1998
+ "step": 1420
1999
+ },
2000
+ {
2001
+ "epoch": 1.6382978723404256,
2002
+ "grad_norm": 1.2725316286087036,
2003
+ "learning_rate": 1.3983784980180332e-05,
2004
+ "loss": 0.1782,
2005
+ "step": 1425
2006
+ },
2007
+ {
2008
+ "epoch": 1.6440483036227715,
2009
+ "grad_norm": 0.9941499829292297,
2010
+ "learning_rate": 1.3888978217586705e-05,
2011
+ "loss": 0.1598,
2012
+ "step": 1430
2013
+ },
2014
+ {
2015
+ "epoch": 1.6497987349051177,
2016
+ "grad_norm": 1.23379385471344,
2017
+ "learning_rate": 1.3794216062486603e-05,
2018
+ "loss": 0.1618,
2019
+ "step": 1435
2020
+ },
2021
+ {
2022
+ "epoch": 1.655549166187464,
2023
+ "grad_norm": 1.1839122772216797,
2024
+ "learning_rate": 1.3699502319577906e-05,
2025
+ "loss": 0.1638,
2026
+ "step": 1440
2027
+ },
2028
+ {
2029
+ "epoch": 1.6612995974698102,
2030
+ "grad_norm": 1.0749571323394775,
2031
+ "learning_rate": 1.3604840791614748e-05,
2032
+ "loss": 0.1592,
2033
+ "step": 1445
2034
+ },
2035
+ {
2036
+ "epoch": 1.6670500287521564,
2037
+ "grad_norm": 0.9488906264305115,
2038
+ "learning_rate": 1.3510235279254834e-05,
2039
+ "loss": 0.1656,
2040
+ "step": 1450
2041
+ },
2042
+ {
2043
+ "epoch": 1.6728004600345026,
2044
+ "grad_norm": 1.0232586860656738,
2045
+ "learning_rate": 1.3415689580906846e-05,
2046
+ "loss": 0.1404,
2047
+ "step": 1455
2048
+ },
2049
+ {
2050
+ "epoch": 1.6785508913168488,
2051
+ "grad_norm": 1.0860204696655273,
2052
+ "learning_rate": 1.3321207492577919e-05,
2053
+ "loss": 0.1618,
2054
+ "step": 1460
2055
+ },
2056
+ {
2057
+ "epoch": 1.684301322599195,
2058
+ "grad_norm": 1.0101890563964844,
2059
+ "learning_rate": 1.3226792807721269e-05,
2060
+ "loss": 0.1535,
2061
+ "step": 1465
2062
+ },
2063
+ {
2064
+ "epoch": 1.6900517538815412,
2065
+ "grad_norm": 0.9916139245033264,
2066
+ "learning_rate": 1.3132449317083854e-05,
2067
+ "loss": 0.1484,
2068
+ "step": 1470
2069
+ },
2070
+ {
2071
+ "epoch": 1.6958021851638874,
2072
+ "grad_norm": 1.3357138633728027,
2073
+ "learning_rate": 1.3038180808554187e-05,
2074
+ "loss": 0.1782,
2075
+ "step": 1475
2076
+ },
2077
+ {
2078
+ "epoch": 1.7015526164462336,
2079
+ "grad_norm": 1.2024328708648682,
2080
+ "learning_rate": 1.2943991067010256e-05,
2081
+ "loss": 0.1597,
2082
+ "step": 1480
2083
+ },
2084
+ {
2085
+ "epoch": 1.7073030477285798,
2086
+ "grad_norm": 1.2084858417510986,
2087
+ "learning_rate": 1.2849883874167547e-05,
2088
+ "loss": 0.1474,
2089
+ "step": 1485
2090
+ },
2091
+ {
2092
+ "epoch": 1.713053479010926,
2093
+ "grad_norm": 1.1819859743118286,
2094
+ "learning_rate": 1.2755863008427222e-05,
2095
+ "loss": 0.1546,
2096
+ "step": 1490
2097
+ },
2098
+ {
2099
+ "epoch": 1.718803910293272,
2100
+ "grad_norm": 0.956397294998169,
2101
+ "learning_rate": 1.2661932244724421e-05,
2102
+ "loss": 0.1595,
2103
+ "step": 1495
2104
+ },
2105
+ {
2106
+ "epoch": 1.7245543415756182,
2107
+ "grad_norm": 0.9791158437728882,
2108
+ "learning_rate": 1.256809535437668e-05,
2109
+ "loss": 0.1549,
2110
+ "step": 1500
2111
+ },
2112
+ {
2113
+ "epoch": 1.7303047728579644,
2114
+ "grad_norm": 1.029632329940796,
2115
+ "learning_rate": 1.2474356104932527e-05,
2116
+ "loss": 0.1477,
2117
+ "step": 1505
2118
+ },
2119
+ {
2120
+ "epoch": 1.7360552041403106,
2121
+ "grad_norm": 1.0042787790298462,
2122
+ "learning_rate": 1.2380718260020203e-05,
2123
+ "loss": 0.1451,
2124
+ "step": 1510
2125
+ },
2126
+ {
2127
+ "epoch": 1.7418056354226565,
2128
+ "grad_norm": 0.965639054775238,
2129
+ "learning_rate": 1.228718557919657e-05,
2130
+ "loss": 0.1599,
2131
+ "step": 1515
2132
+ },
2133
+ {
2134
+ "epoch": 1.7475560667050027,
2135
+ "grad_norm": 0.8879011273384094,
2136
+ "learning_rate": 1.2193761817796143e-05,
2137
+ "loss": 0.1492,
2138
+ "step": 1520
2139
+ },
2140
+ {
2141
+ "epoch": 1.753306497987349,
2142
+ "grad_norm": 1.0300003290176392,
2143
+ "learning_rate": 1.2100450726780344e-05,
2144
+ "loss": 0.1478,
2145
+ "step": 1525
2146
+ },
2147
+ {
2148
+ "epoch": 1.7590569292696951,
2149
+ "grad_norm": 1.1561692953109741,
2150
+ "learning_rate": 1.2007256052586864e-05,
2151
+ "loss": 0.1301,
2152
+ "step": 1530
2153
+ },
2154
+ {
2155
+ "epoch": 1.7648073605520413,
2156
+ "grad_norm": 0.9975626468658447,
2157
+ "learning_rate": 1.1914181536979271e-05,
2158
+ "loss": 0.1602,
2159
+ "step": 1535
2160
+ },
2161
+ {
2162
+ "epoch": 1.7705577918343876,
2163
+ "grad_norm": 1.0221582651138306,
2164
+ "learning_rate": 1.1821230916896763e-05,
2165
+ "loss": 0.152,
2166
+ "step": 1540
2167
+ },
2168
+ {
2169
+ "epoch": 1.7763082231167338,
2170
+ "grad_norm": 0.8500351309776306,
2171
+ "learning_rate": 1.1728407924304141e-05,
2172
+ "loss": 0.1343,
2173
+ "step": 1545
2174
+ },
2175
+ {
2176
+ "epoch": 1.78205865439908,
2177
+ "grad_norm": 1.1222327947616577,
2178
+ "learning_rate": 1.1635716286041966e-05,
2179
+ "loss": 0.1569,
2180
+ "step": 1550
2181
+ },
2182
+ {
2183
+ "epoch": 1.7878090856814262,
2184
+ "grad_norm": 1.1146150827407837,
2185
+ "learning_rate": 1.154315972367692e-05,
2186
+ "loss": 0.1367,
2187
+ "step": 1555
2188
+ },
2189
+ {
2190
+ "epoch": 1.7935595169637724,
2191
+ "grad_norm": 0.942156970500946,
2192
+ "learning_rate": 1.145074195335241e-05,
2193
+ "loss": 0.156,
2194
+ "step": 1560
2195
+ },
2196
+ {
2197
+ "epoch": 1.7993099482461186,
2198
+ "grad_norm": 1.0856579542160034,
2199
+ "learning_rate": 1.1358466685639313e-05,
2200
+ "loss": 0.1411,
2201
+ "step": 1565
2202
+ },
2203
+ {
2204
+ "epoch": 1.8050603795284648,
2205
+ "grad_norm": 1.0479731559753418,
2206
+ "learning_rate": 1.1266337625387065e-05,
2207
+ "loss": 0.1303,
2208
+ "step": 1570
2209
+ },
2210
+ {
2211
+ "epoch": 1.810810810810811,
2212
+ "grad_norm": 0.9391495585441589,
2213
+ "learning_rate": 1.1174358471574854e-05,
2214
+ "loss": 0.1379,
2215
+ "step": 1575
2216
+ },
2217
+ {
2218
+ "epoch": 1.816561242093157,
2219
+ "grad_norm": 1.1122937202453613,
2220
+ "learning_rate": 1.108253291716314e-05,
2221
+ "loss": 0.149,
2222
+ "step": 1580
2223
+ },
2224
+ {
2225
+ "epoch": 1.8223116733755031,
2226
+ "grad_norm": 0.9562496542930603,
2227
+ "learning_rate": 1.099086464894537e-05,
2228
+ "loss": 0.1579,
2229
+ "step": 1585
2230
+ },
2231
+ {
2232
+ "epoch": 1.8280621046578494,
2233
+ "grad_norm": 0.9799585342407227,
2234
+ "learning_rate": 1.0899357347399943e-05,
2235
+ "loss": 0.1611,
2236
+ "step": 1590
2237
+ },
2238
+ {
2239
+ "epoch": 1.8338125359401956,
2240
+ "grad_norm": 0.9481344223022461,
2241
+ "learning_rate": 1.0808014686542463e-05,
2242
+ "loss": 0.1159,
2243
+ "step": 1595
2244
+ },
2245
+ {
2246
+ "epoch": 1.8395629672225415,
2247
+ "grad_norm": 1.0899279117584229,
2248
+ "learning_rate": 1.0716840333778217e-05,
2249
+ "loss": 0.1317,
2250
+ "step": 1600
2251
+ },
2252
+ {
2253
+ "epoch": 1.8453133985048877,
2254
+ "grad_norm": 0.9299836158752441,
2255
+ "learning_rate": 1.0625837949754918e-05,
2256
+ "loss": 0.149,
2257
+ "step": 1605
2258
+ },
2259
+ {
2260
+ "epoch": 1.851063829787234,
2261
+ "grad_norm": 0.9929051399230957,
2262
+ "learning_rate": 1.0535011188215742e-05,
2263
+ "loss": 0.13,
2264
+ "step": 1610
2265
+ },
2266
+ {
2267
+ "epoch": 1.8568142610695801,
2268
+ "grad_norm": 1.0908143520355225,
2269
+ "learning_rate": 1.0444363695852631e-05,
2270
+ "loss": 0.1273,
2271
+ "step": 1615
2272
+ },
2273
+ {
2274
+ "epoch": 1.8625646923519263,
2275
+ "grad_norm": 0.9257168173789978,
2276
+ "learning_rate": 1.0353899112159868e-05,
2277
+ "loss": 0.1149,
2278
+ "step": 1620
2279
+ },
2280
+ {
2281
+ "epoch": 1.8683151236342725,
2282
+ "grad_norm": 0.903671383857727,
2283
+ "learning_rate": 1.0263621069287958e-05,
2284
+ "loss": 0.1232,
2285
+ "step": 1625
2286
+ },
2287
+ {
2288
+ "epoch": 1.8740655549166187,
2289
+ "grad_norm": 1.1721129417419434,
2290
+ "learning_rate": 1.01735331918978e-05,
2291
+ "loss": 0.1225,
2292
+ "step": 1630
2293
+ },
2294
+ {
2295
+ "epoch": 1.879815986198965,
2296
+ "grad_norm": 0.9155874848365784,
2297
+ "learning_rate": 1.0083639097015164e-05,
2298
+ "loss": 0.1207,
2299
+ "step": 1635
2300
+ },
2301
+ {
2302
+ "epoch": 1.8855664174813112,
2303
+ "grad_norm": 0.8894999027252197,
2304
+ "learning_rate": 9.993942393885444e-06,
2305
+ "loss": 0.1292,
2306
+ "step": 1640
2307
+ },
2308
+ {
2309
+ "epoch": 1.8913168487636574,
2310
+ "grad_norm": 0.8587804436683655,
2311
+ "learning_rate": 9.90444668382876e-06,
2312
+ "loss": 0.1264,
2313
+ "step": 1645
2314
+ },
2315
+ {
2316
+ "epoch": 1.8970672800460036,
2317
+ "grad_norm": 0.8552981019020081,
2318
+ "learning_rate": 9.815155560095374e-06,
2319
+ "loss": 0.12,
2320
+ "step": 1650
2321
+ },
2322
+ {
2323
+ "epoch": 1.9028177113283498,
2324
+ "grad_norm": 0.9915004968643188,
2325
+ "learning_rate": 9.72607260772141e-06,
2326
+ "loss": 0.1235,
2327
+ "step": 1655
2328
+ },
2329
+ {
2330
+ "epoch": 1.908568142610696,
2331
+ "grad_norm": 1.0675162076950073,
2332
+ "learning_rate": 9.637201403384914e-06,
2333
+ "loss": 0.1295,
2334
+ "step": 1660
2335
+ },
2336
+ {
2337
+ "epoch": 1.914318573893042,
2338
+ "grad_norm": 1.0216548442840576,
2339
+ "learning_rate": 9.54854551526227e-06,
2340
+ "loss": 0.1363,
2341
+ "step": 1665
2342
+ },
2343
+ {
2344
+ "epoch": 1.9200690051753881,
2345
+ "grad_norm": 1.1097060441970825,
2346
+ "learning_rate": 9.460108502884894e-06,
2347
+ "loss": 0.1228,
2348
+ "step": 1670
2349
+ },
2350
+ {
2351
+ "epoch": 1.9258194364577343,
2352
+ "grad_norm": 0.7692320346832275,
2353
+ "learning_rate": 9.371893916996366e-06,
2354
+ "loss": 0.1225,
2355
+ "step": 1675
2356
+ },
2357
+ {
2358
+ "epoch": 1.9315698677400805,
2359
+ "grad_norm": 1.0181418657302856,
2360
+ "learning_rate": 9.283905299409848e-06,
2361
+ "loss": 0.1291,
2362
+ "step": 1680
2363
+ },
2364
+ {
2365
+ "epoch": 1.9373202990224265,
2366
+ "grad_norm": 0.9889492392539978,
2367
+ "learning_rate": 9.19614618286588e-06,
2368
+ "loss": 0.1198,
2369
+ "step": 1685
2370
+ },
2371
+ {
2372
+ "epoch": 1.9430707303047727,
2373
+ "grad_norm": 1.0681936740875244,
2374
+ "learning_rate": 9.108620090890534e-06,
2375
+ "loss": 0.1251,
2376
+ "step": 1690
2377
+ },
2378
+ {
2379
+ "epoch": 1.948821161587119,
2380
+ "grad_norm": 0.8951586484909058,
2381
+ "learning_rate": 9.021330537653967e-06,
2382
+ "loss": 0.1097,
2383
+ "step": 1695
2384
+ },
2385
+ {
2386
+ "epoch": 1.9545715928694651,
2387
+ "grad_norm": 0.9336473941802979,
2388
+ "learning_rate": 8.934281027829293e-06,
2389
+ "loss": 0.117,
2390
+ "step": 1700
2391
+ },
2392
+ {
2393
+ "epoch": 1.9603220241518113,
2394
+ "grad_norm": 1.0660419464111328,
2395
+ "learning_rate": 8.8474750564519e-06,
2396
+ "loss": 0.1185,
2397
+ "step": 1705
2398
+ },
2399
+ {
2400
+ "epoch": 1.9660724554341575,
2401
+ "grad_norm": 0.9332837462425232,
2402
+ "learning_rate": 8.760916108779122e-06,
2403
+ "loss": 0.1277,
2404
+ "step": 1710
2405
+ },
2406
+ {
2407
+ "epoch": 1.9718228867165037,
2408
+ "grad_norm": 0.9789974689483643,
2409
+ "learning_rate": 8.674607660150278e-06,
2410
+ "loss": 0.1245,
2411
+ "step": 1715
2412
+ },
2413
+ {
2414
+ "epoch": 1.97757331799885,
2415
+ "grad_norm": 0.9933024644851685,
2416
+ "learning_rate": 8.588553175847192e-06,
2417
+ "loss": 0.1127,
2418
+ "step": 1720
2419
+ },
2420
+ {
2421
+ "epoch": 1.9833237492811961,
2422
+ "grad_norm": 0.8819986581802368,
2423
+ "learning_rate": 8.50275611095498e-06,
2424
+ "loss": 0.1178,
2425
+ "step": 1725
2426
+ },
2427
+ {
2428
+ "epoch": 1.9890741805635423,
2429
+ "grad_norm": 0.9804098010063171,
2430
+ "learning_rate": 8.417219910223408e-06,
2431
+ "loss": 0.1213,
2432
+ "step": 1730
2433
+ },
2434
+ {
2435
+ "epoch": 1.9948246118458886,
2436
+ "grad_norm": 0.9790911674499512,
2437
+ "learning_rate": 8.331948007928552e-06,
2438
+ "loss": 0.1158,
2439
+ "step": 1735
2440
+ },
2441
+ {
2442
+ "epoch": 2.0,
2443
+ "grad_norm": 1.2117398977279663,
2444
+ "learning_rate": 8.246943827734899e-06,
2445
+ "loss": 0.1225,
2446
+ "step": 1740
2447
+ },
2448
+ {
2449
+ "epoch": 2.005750431282346,
2450
+ "grad_norm": 0.7414922714233398,
2451
+ "learning_rate": 8.16221078255792e-06,
2452
+ "loss": 0.0935,
2453
+ "step": 1745
2454
+ },
2455
+ {
2456
+ "epoch": 2.0115008625646924,
2457
+ "grad_norm": 0.9437388181686401,
2458
+ "learning_rate": 8.077752274427006e-06,
2459
+ "loss": 0.087,
2460
+ "step": 1750
2461
+ },
2462
+ {
2463
+ "epoch": 2.0172512938470386,
2464
+ "grad_norm": 0.971380352973938,
2465
+ "learning_rate": 7.99357169434889e-06,
2466
+ "loss": 0.1035,
2467
+ "step": 1755
2468
+ },
2469
+ {
2470
+ "epoch": 2.023001725129385,
2471
+ "grad_norm": 1.07891845703125,
2472
+ "learning_rate": 7.909672422171519e-06,
2473
+ "loss": 0.0998,
2474
+ "step": 1760
2475
+ },
2476
+ {
2477
+ "epoch": 2.028752156411731,
2478
+ "grad_norm": 0.7205830216407776,
2479
+ "learning_rate": 7.826057826448334e-06,
2480
+ "loss": 0.0919,
2481
+ "step": 1765
2482
+ },
2483
+ {
2484
+ "epoch": 2.034502587694077,
2485
+ "grad_norm": 0.742916464805603,
2486
+ "learning_rate": 7.742731264303005e-06,
2487
+ "loss": 0.0881,
2488
+ "step": 1770
2489
+ },
2490
+ {
2491
+ "epoch": 2.0402530189764234,
2492
+ "grad_norm": 0.8644776344299316,
2493
+ "learning_rate": 7.659696081294683e-06,
2494
+ "loss": 0.0848,
2495
+ "step": 1775
2496
+ },
2497
+ {
2498
+ "epoch": 2.0460034502587696,
2499
+ "grad_norm": 0.8643521070480347,
2500
+ "learning_rate": 7.5769556112836456e-06,
2501
+ "loss": 0.0882,
2502
+ "step": 1780
2503
+ },
2504
+ {
2505
+ "epoch": 2.0517538815411154,
2506
+ "grad_norm": 0.9698953032493591,
2507
+ "learning_rate": 7.494513176297444e-06,
2508
+ "loss": 0.0906,
2509
+ "step": 1785
2510
+ },
2511
+ {
2512
+ "epoch": 2.0575043128234616,
2513
+ "grad_norm": 0.7276870012283325,
2514
+ "learning_rate": 7.412372086397545e-06,
2515
+ "loss": 0.0911,
2516
+ "step": 1790
2517
+ },
2518
+ {
2519
+ "epoch": 2.063254744105808,
2520
+ "grad_norm": 0.7837342023849487,
2521
+ "learning_rate": 7.330535639546418e-06,
2522
+ "loss": 0.0797,
2523
+ "step": 1795
2524
+ },
2525
+ {
2526
+ "epoch": 2.069005175388154,
2527
+ "grad_norm": 0.8142509460449219,
2528
+ "learning_rate": 7.2490071214751155e-06,
2529
+ "loss": 0.0845,
2530
+ "step": 1800
2531
+ },
2532
+ {
2533
+ "epoch": 2.0747556066705,
2534
+ "grad_norm": 0.9334195256233215,
2535
+ "learning_rate": 7.167789805551354e-06,
2536
+ "loss": 0.0926,
2537
+ "step": 1805
2538
+ },
2539
+ {
2540
+ "epoch": 2.0805060379528464,
2541
+ "grad_norm": 0.7895086407661438,
2542
+ "learning_rate": 7.086886952648108e-06,
2543
+ "loss": 0.0974,
2544
+ "step": 1810
2545
+ },
2546
+ {
2547
+ "epoch": 2.0862564692351926,
2548
+ "grad_norm": 0.6973230838775635,
2549
+ "learning_rate": 7.006301811012653e-06,
2550
+ "loss": 0.0846,
2551
+ "step": 1815
2552
+ },
2553
+ {
2554
+ "epoch": 2.092006900517539,
2555
+ "grad_norm": 0.7749288082122803,
2556
+ "learning_rate": 6.926037616136183e-06,
2557
+ "loss": 0.0941,
2558
+ "step": 1820
2559
+ },
2560
+ {
2561
+ "epoch": 2.097757331799885,
2562
+ "grad_norm": 0.8271239399909973,
2563
+ "learning_rate": 6.846097590623877e-06,
2564
+ "loss": 0.0811,
2565
+ "step": 1825
2566
+ },
2567
+ {
2568
+ "epoch": 2.103507763082231,
2569
+ "grad_norm": 0.8368834257125854,
2570
+ "learning_rate": 6.7664849440655205e-06,
2571
+ "loss": 0.0849,
2572
+ "step": 1830
2573
+ },
2574
+ {
2575
+ "epoch": 2.1092581943645774,
2576
+ "grad_norm": 0.8112236857414246,
2577
+ "learning_rate": 6.687202872906652e-06,
2578
+ "loss": 0.0787,
2579
+ "step": 1835
2580
+ },
2581
+ {
2582
+ "epoch": 2.1150086256469236,
2583
+ "grad_norm": 0.7273387908935547,
2584
+ "learning_rate": 6.608254560320222e-06,
2585
+ "loss": 0.0872,
2586
+ "step": 1840
2587
+ },
2588
+ {
2589
+ "epoch": 2.12075905692927,
2590
+ "grad_norm": 0.8497496247291565,
2591
+ "learning_rate": 6.529643176078763e-06,
2592
+ "loss": 0.0845,
2593
+ "step": 1845
2594
+ },
2595
+ {
2596
+ "epoch": 2.126509488211616,
2597
+ "grad_norm": 0.7193707227706909,
2598
+ "learning_rate": 6.451371876427168e-06,
2599
+ "loss": 0.0975,
2600
+ "step": 1850
2601
+ },
2602
+ {
2603
+ "epoch": 2.132259919493962,
2604
+ "grad_norm": 0.7538492679595947,
2605
+ "learning_rate": 6.3734438039559246e-06,
2606
+ "loss": 0.0856,
2607
+ "step": 1855
2608
+ },
2609
+ {
2610
+ "epoch": 2.1380103507763084,
2611
+ "grad_norm": 0.8019741177558899,
2612
+ "learning_rate": 6.29586208747496e-06,
2613
+ "loss": 0.1007,
2614
+ "step": 1860
2615
+ },
2616
+ {
2617
+ "epoch": 2.1437607820586546,
2618
+ "grad_norm": 0.7402260303497314,
2619
+ "learning_rate": 6.218629841888031e-06,
2620
+ "loss": 0.0814,
2621
+ "step": 1865
2622
+ },
2623
+ {
2624
+ "epoch": 2.1495112133410004,
2625
+ "grad_norm": 0.7180666923522949,
2626
+ "learning_rate": 6.141750168067645e-06,
2627
+ "loss": 0.0972,
2628
+ "step": 1870
2629
+ },
2630
+ {
2631
+ "epoch": 2.1552616446233466,
2632
+ "grad_norm": 0.7025020718574524,
2633
+ "learning_rate": 6.065226152730547e-06,
2634
+ "loss": 0.089,
2635
+ "step": 1875
2636
+ },
2637
+ {
2638
+ "epoch": 2.1610120759056928,
2639
+ "grad_norm": 0.7331565618515015,
2640
+ "learning_rate": 5.989060868313826e-06,
2641
+ "loss": 0.0952,
2642
+ "step": 1880
2643
+ },
2644
+ {
2645
+ "epoch": 2.166762507188039,
2646
+ "grad_norm": 0.7998526692390442,
2647
+ "learning_rate": 5.913257372851519e-06,
2648
+ "loss": 0.0927,
2649
+ "step": 1885
2650
+ },
2651
+ {
2652
+ "epoch": 2.172512938470385,
2653
+ "grad_norm": 0.826391339302063,
2654
+ "learning_rate": 5.837818709851845e-06,
2655
+ "loss": 0.083,
2656
+ "step": 1890
2657
+ },
2658
+ {
2659
+ "epoch": 2.1782633697527314,
2660
+ "grad_norm": 0.8038471341133118,
2661
+ "learning_rate": 5.7627479081750254e-06,
2662
+ "loss": 0.0885,
2663
+ "step": 1895
2664
+ },
2665
+ {
2666
+ "epoch": 2.1840138010350776,
2667
+ "grad_norm": 0.8005362749099731,
2668
+ "learning_rate": 5.688047981911642e-06,
2669
+ "loss": 0.0833,
2670
+ "step": 1900
2671
+ },
2672
+ {
2673
+ "epoch": 2.189764232317424,
2674
+ "grad_norm": 0.7602734565734863,
2675
+ "learning_rate": 5.613721930261659e-06,
2676
+ "loss": 0.0907,
2677
+ "step": 1905
2678
+ },
2679
+ {
2680
+ "epoch": 2.19551466359977,
2681
+ "grad_norm": 0.7925649285316467,
2682
+ "learning_rate": 5.53977273741396e-06,
2683
+ "loss": 0.0773,
2684
+ "step": 1910
2685
+ },
2686
+ {
2687
+ "epoch": 2.201265094882116,
2688
+ "grad_norm": 0.6295461654663086,
2689
+ "learning_rate": 5.4662033724265855e-06,
2690
+ "loss": 0.08,
2691
+ "step": 1915
2692
+ },
2693
+ {
2694
+ "epoch": 2.2070155261644624,
2695
+ "grad_norm": 0.81055748462677,
2696
+ "learning_rate": 5.393016789107472e-06,
2697
+ "loss": 0.0864,
2698
+ "step": 1920
2699
+ },
2700
+ {
2701
+ "epoch": 2.2127659574468086,
2702
+ "grad_norm": 0.8425546288490295,
2703
+ "learning_rate": 5.3202159258959054e-06,
2704
+ "loss": 0.0814,
2705
+ "step": 1925
2706
+ },
2707
+ {
2708
+ "epoch": 2.218516388729155,
2709
+ "grad_norm": 0.6603655219078064,
2710
+ "learning_rate": 5.2478037057445015e-06,
2711
+ "loss": 0.0957,
2712
+ "step": 1930
2713
+ },
2714
+ {
2715
+ "epoch": 2.224266820011501,
2716
+ "grad_norm": 0.6792340874671936,
2717
+ "learning_rate": 5.175783036001885e-06,
2718
+ "loss": 0.0898,
2719
+ "step": 1935
2720
+ },
2721
+ {
2722
+ "epoch": 2.230017251293847,
2723
+ "grad_norm": 0.7001053094863892,
2724
+ "learning_rate": 5.104156808295932e-06,
2725
+ "loss": 0.0838,
2726
+ "step": 1940
2727
+ },
2728
+ {
2729
+ "epoch": 2.2357676825761934,
2730
+ "grad_norm": 0.7168072462081909,
2731
+ "learning_rate": 5.032927898417676e-06,
2732
+ "loss": 0.0884,
2733
+ "step": 1945
2734
+ },
2735
+ {
2736
+ "epoch": 2.2415181138585396,
2737
+ "grad_norm": 0.7977825999259949,
2738
+ "learning_rate": 4.962099166205864e-06,
2739
+ "loss": 0.0861,
2740
+ "step": 1950
2741
+ },
2742
+ {
2743
+ "epoch": 2.2472685451408854,
2744
+ "grad_norm": 0.6316221356391907,
2745
+ "learning_rate": 4.8916734554321215e-06,
2746
+ "loss": 0.0781,
2747
+ "step": 1955
2748
+ },
2749
+ {
2750
+ "epoch": 2.2530189764232316,
2751
+ "grad_norm": 0.71268230676651,
2752
+ "learning_rate": 4.821653593686765e-06,
2753
+ "loss": 0.0828,
2754
+ "step": 1960
2755
+ },
2756
+ {
2757
+ "epoch": 2.2587694077055778,
2758
+ "grad_norm": 0.6494830250740051,
2759
+ "learning_rate": 4.75204239226528e-06,
2760
+ "loss": 0.075,
2761
+ "step": 1965
2762
+ },
2763
+ {
2764
+ "epoch": 2.264519838987924,
2765
+ "grad_norm": 0.7507233023643494,
2766
+ "learning_rate": 4.68284264605547e-06,
2767
+ "loss": 0.077,
2768
+ "step": 1970
2769
+ },
2770
+ {
2771
+ "epoch": 2.27027027027027,
2772
+ "grad_norm": 0.7941622138023376,
2773
+ "learning_rate": 4.614057133425203e-06,
2774
+ "loss": 0.0827,
2775
+ "step": 1975
2776
+ },
2777
+ {
2778
+ "epoch": 2.2760207015526164,
2779
+ "grad_norm": 0.8505463004112244,
2780
+ "learning_rate": 4.545688616110894e-06,
2781
+ "loss": 0.0745,
2782
+ "step": 1980
2783
+ },
2784
+ {
2785
+ "epoch": 2.2817711328349626,
2786
+ "grad_norm": 0.7615712881088257,
2787
+ "learning_rate": 4.477739839106606e-06,
2788
+ "loss": 0.0882,
2789
+ "step": 1985
2790
+ },
2791
+ {
2792
+ "epoch": 2.287521564117309,
2793
+ "grad_norm": 0.6822249889373779,
2794
+ "learning_rate": 4.410213530553838e-06,
2795
+ "loss": 0.0824,
2796
+ "step": 1990
2797
+ },
2798
+ {
2799
+ "epoch": 2.293271995399655,
2800
+ "grad_norm": 0.8102391958236694,
2801
+ "learning_rate": 4.343112401631982e-06,
2802
+ "loss": 0.0919,
2803
+ "step": 1995
2804
+ },
2805
+ {
2806
+ "epoch": 2.299022426682001,
2807
+ "grad_norm": 0.621198296546936,
2808
+ "learning_rate": 4.2764391464495015e-06,
2809
+ "loss": 0.0722,
2810
+ "step": 2000
2811
+ },
2812
+ {
2813
+ "epoch": 2.3047728579643474,
2814
+ "grad_norm": 0.7153138518333435,
2815
+ "learning_rate": 4.210196441935719e-06,
2816
+ "loss": 0.0719,
2817
+ "step": 2005
2818
+ },
2819
+ {
2820
+ "epoch": 2.3105232892466936,
2821
+ "grad_norm": 0.6881626844406128,
2822
+ "learning_rate": 4.144386947733377e-06,
2823
+ "loss": 0.0818,
2824
+ "step": 2010
2825
+ },
2826
+ {
2827
+ "epoch": 2.31627372052904,
2828
+ "grad_norm": 0.7306118607521057,
2829
+ "learning_rate": 4.079013306091837e-06,
2830
+ "loss": 0.0843,
2831
+ "step": 2015
2832
+ },
2833
+ {
2834
+ "epoch": 2.322024151811386,
2835
+ "grad_norm": 0.7310144901275635,
2836
+ "learning_rate": 4.014078141760966e-06,
2837
+ "loss": 0.0797,
2838
+ "step": 2020
2839
+ },
2840
+ {
2841
+ "epoch": 2.327774583093732,
2842
+ "grad_norm": 0.6377416253089905,
2843
+ "learning_rate": 3.949584061885807e-06,
2844
+ "loss": 0.0771,
2845
+ "step": 2025
2846
+ },
2847
+ {
2848
+ "epoch": 2.3335250143760784,
2849
+ "grad_norm": 0.7563802003860474,
2850
+ "learning_rate": 3.885533655901871e-06,
2851
+ "loss": 0.0783,
2852
+ "step": 2030
2853
+ },
2854
+ {
2855
+ "epoch": 2.3392754456584246,
2856
+ "grad_norm": 0.609529435634613,
2857
+ "learning_rate": 3.821929495431159e-06,
2858
+ "loss": 0.0665,
2859
+ "step": 2035
2860
+ },
2861
+ {
2862
+ "epoch": 2.3450258769407704,
2863
+ "grad_norm": 0.6828349828720093,
2864
+ "learning_rate": 3.758774134178948e-06,
2865
+ "loss": 0.0877,
2866
+ "step": 2040
2867
+ },
2868
+ {
2869
+ "epoch": 2.3507763082231166,
2870
+ "grad_norm": 0.7285950779914856,
2871
+ "learning_rate": 3.69607010783122e-06,
2872
+ "loss": 0.0797,
2873
+ "step": 2045
2874
+ },
2875
+ {
2876
+ "epoch": 2.3565267395054628,
2877
+ "grad_norm": 0.5805037021636963,
2878
+ "learning_rate": 3.63381993395288e-06,
2879
+ "loss": 0.0826,
2880
+ "step": 2050
2881
+ },
2882
+ {
2883
+ "epoch": 2.362277170787809,
2884
+ "grad_norm": 0.7669761776924133,
2885
+ "learning_rate": 3.5720261118866726e-06,
2886
+ "loss": 0.079,
2887
+ "step": 2055
2888
+ },
2889
+ {
2890
+ "epoch": 2.368027602070155,
2891
+ "grad_norm": 0.615234375,
2892
+ "learning_rate": 3.5106911226528336e-06,
2893
+ "loss": 0.0769,
2894
+ "step": 2060
2895
+ },
2896
+ {
2897
+ "epoch": 2.3737780333525014,
2898
+ "grad_norm": 0.6917020678520203,
2899
+ "learning_rate": 3.4498174288494623e-06,
2900
+ "loss": 0.0734,
2901
+ "step": 2065
2902
+ },
2903
+ {
2904
+ "epoch": 2.3795284646348476,
2905
+ "grad_norm": 0.7300450801849365,
2906
+ "learning_rate": 3.3894074745536583e-06,
2907
+ "loss": 0.0831,
2908
+ "step": 2070
2909
+ },
2910
+ {
2911
+ "epoch": 2.3852788959171938,
2912
+ "grad_norm": 0.5797042846679688,
2913
+ "learning_rate": 3.329463685223411e-06,
2914
+ "loss": 0.0859,
2915
+ "step": 2075
2916
+ },
2917
+ {
2918
+ "epoch": 2.39102932719954,
2919
+ "grad_norm": 0.5983770489692688,
2920
+ "learning_rate": 3.2699884676001747e-06,
2921
+ "loss": 0.0743,
2922
+ "step": 2080
2923
+ },
2924
+ {
2925
+ "epoch": 2.396779758481886,
2926
+ "grad_norm": 0.9608801603317261,
2927
+ "learning_rate": 3.2109842096122837e-06,
2928
+ "loss": 0.0716,
2929
+ "step": 2085
2930
+ },
2931
+ {
2932
+ "epoch": 2.4025301897642324,
2933
+ "grad_norm": 0.6366323828697205,
2934
+ "learning_rate": 3.1524532802790506e-06,
2935
+ "loss": 0.0758,
2936
+ "step": 2090
2937
+ },
2938
+ {
2939
+ "epoch": 2.4082806210465786,
2940
+ "grad_norm": 0.8448190689086914,
2941
+ "learning_rate": 3.0943980296156476e-06,
2942
+ "loss": 0.0797,
2943
+ "step": 2095
2944
+ },
2945
+ {
2946
+ "epoch": 2.414031052328925,
2947
+ "grad_norm": 0.7333036661148071,
2948
+ "learning_rate": 3.0368207885387662e-06,
2949
+ "loss": 0.0806,
2950
+ "step": 2100
2951
+ },
2952
+ {
2953
+ "epoch": 2.419781483611271,
2954
+ "grad_norm": 0.7811900973320007,
2955
+ "learning_rate": 2.979723868773031e-06,
2956
+ "loss": 0.0739,
2957
+ "step": 2105
2958
+ },
2959
+ {
2960
+ "epoch": 2.425531914893617,
2961
+ "grad_norm": 0.7617394328117371,
2962
+ "learning_rate": 2.9231095627581705e-06,
2963
+ "loss": 0.0865,
2964
+ "step": 2110
2965
+ },
2966
+ {
2967
+ "epoch": 2.431282346175963,
2968
+ "grad_norm": 0.8095620274543762,
2969
+ "learning_rate": 2.8669801435569944e-06,
2970
+ "loss": 0.0753,
2971
+ "step": 2115
2972
+ },
2973
+ {
2974
+ "epoch": 2.4370327774583096,
2975
+ "grad_norm": 0.8697852492332458,
2976
+ "learning_rate": 2.8113378647641197e-06,
2977
+ "loss": 0.0785,
2978
+ "step": 2120
2979
+ },
2980
+ {
2981
+ "epoch": 2.4427832087406554,
2982
+ "grad_norm": 0.5686136484146118,
2983
+ "learning_rate": 2.756184960415473e-06,
2984
+ "loss": 0.0677,
2985
+ "step": 2125
2986
+ },
2987
+ {
2988
+ "epoch": 2.4485336400230016,
2989
+ "grad_norm": 0.656196653842926,
2990
+ "learning_rate": 2.7015236448986313e-06,
2991
+ "loss": 0.0744,
2992
+ "step": 2130
2993
+ },
2994
+ {
2995
+ "epoch": 2.4542840713053478,
2996
+ "grad_norm": 0.7253570556640625,
2997
+ "learning_rate": 2.6473561128638875e-06,
2998
+ "loss": 0.0751,
2999
+ "step": 2135
3000
+ },
3001
+ {
3002
+ "epoch": 2.460034502587694,
3003
+ "grad_norm": 0.8234943747520447,
3004
+ "learning_rate": 2.593684539136133e-06,
3005
+ "loss": 0.074,
3006
+ "step": 2140
3007
+ },
3008
+ {
3009
+ "epoch": 2.46578493387004,
3010
+ "grad_norm": 0.5695974826812744,
3011
+ "learning_rate": 2.54051107862756e-06,
3012
+ "loss": 0.0673,
3013
+ "step": 2145
3014
+ },
3015
+ {
3016
+ "epoch": 2.4715353651523864,
3017
+ "grad_norm": 0.7163220643997192,
3018
+ "learning_rate": 2.4878378662511208e-06,
3019
+ "loss": 0.0715,
3020
+ "step": 2150
3021
+ },
3022
+ {
3023
+ "epoch": 2.4772857964347326,
3024
+ "grad_norm": 0.7621355652809143,
3025
+ "learning_rate": 2.4356670168348154e-06,
3026
+ "loss": 0.081,
3027
+ "step": 2155
3028
+ },
3029
+ {
3030
+ "epoch": 2.4830362277170788,
3031
+ "grad_norm": 0.6446969509124756,
3032
+ "learning_rate": 2.3840006250367964e-06,
3033
+ "loss": 0.0782,
3034
+ "step": 2160
3035
+ },
3036
+ {
3037
+ "epoch": 2.488786658999425,
3038
+ "grad_norm": 0.6422896981239319,
3039
+ "learning_rate": 2.3328407652612515e-06,
3040
+ "loss": 0.0766,
3041
+ "step": 2165
3042
+ },
3043
+ {
3044
+ "epoch": 2.494537090281771,
3045
+ "grad_norm": 0.6008805632591248,
3046
+ "learning_rate": 2.2821894915751163e-06,
3047
+ "loss": 0.0639,
3048
+ "step": 2170
3049
+ },
3050
+ {
3051
+ "epoch": 2.5002875215641174,
3052
+ "grad_norm": 0.6368566155433655,
3053
+ "learning_rate": 2.2320488376256228e-06,
3054
+ "loss": 0.0738,
3055
+ "step": 2175
3056
+ },
3057
+ {
3058
+ "epoch": 2.5060379528464636,
3059
+ "grad_norm": 0.6397927403450012,
3060
+ "learning_rate": 2.182420816558619e-06,
3061
+ "loss": 0.0792,
3062
+ "step": 2180
3063
+ },
3064
+ {
3065
+ "epoch": 2.51178838412881,
3066
+ "grad_norm": 0.5549697875976562,
3067
+ "learning_rate": 2.1333074209377644e-06,
3068
+ "loss": 0.0637,
3069
+ "step": 2185
3070
+ },
3071
+ {
3072
+ "epoch": 2.517538815411156,
3073
+ "grad_norm": 0.6596667766571045,
3074
+ "learning_rate": 2.084710622664523e-06,
3075
+ "loss": 0.0804,
3076
+ "step": 2190
3077
+ },
3078
+ {
3079
+ "epoch": 2.523289246693502,
3080
+ "grad_norm": 0.6269316673278809,
3081
+ "learning_rate": 2.0366323728989927e-06,
3082
+ "loss": 0.0768,
3083
+ "step": 2195
3084
+ },
3085
+ {
3086
+ "epoch": 2.529039677975848,
3087
+ "grad_norm": 0.7077573537826538,
3088
+ "learning_rate": 1.989074601981553e-06,
3089
+ "loss": 0.0786,
3090
+ "step": 2200
3091
+ },
3092
+ {
3093
+ "epoch": 2.5347901092581946,
3094
+ "grad_norm": 0.6656134724617004,
3095
+ "learning_rate": 1.942039219355376e-06,
3096
+ "loss": 0.0666,
3097
+ "step": 2205
3098
+ },
3099
+ {
3100
+ "epoch": 2.5405405405405403,
3101
+ "grad_norm": 0.8383512496948242,
3102
+ "learning_rate": 1.8955281134897656e-06,
3103
+ "loss": 0.0735,
3104
+ "step": 2210
3105
+ },
3106
+ {
3107
+ "epoch": 2.5462909718228866,
3108
+ "grad_norm": 0.606957197189331,
3109
+ "learning_rate": 1.8495431518043177e-06,
3110
+ "loss": 0.0684,
3111
+ "step": 2215
3112
+ },
3113
+ {
3114
+ "epoch": 2.5520414031052328,
3115
+ "grad_norm": 0.4918835461139679,
3116
+ "learning_rate": 1.8040861805939624e-06,
3117
+ "loss": 0.0697,
3118
+ "step": 2220
3119
+ },
3120
+ {
3121
+ "epoch": 2.557791834387579,
3122
+ "grad_norm": 0.6413158178329468,
3123
+ "learning_rate": 1.759159024954829e-06,
3124
+ "loss": 0.0683,
3125
+ "step": 2225
3126
+ },
3127
+ {
3128
+ "epoch": 2.563542265669925,
3129
+ "grad_norm": 0.6759584546089172,
3130
+ "learning_rate": 1.714763488710951e-06,
3131
+ "loss": 0.0633,
3132
+ "step": 2230
3133
+ },
3134
+ {
3135
+ "epoch": 2.5692926969522714,
3136
+ "grad_norm": 0.7383154630661011,
3137
+ "learning_rate": 1.6709013543418722e-06,
3138
+ "loss": 0.076,
3139
+ "step": 2235
3140
+ },
3141
+ {
3142
+ "epoch": 2.5750431282346176,
3143
+ "grad_norm": 0.7430702447891235,
3144
+ "learning_rate": 1.6275743829110662e-06,
3145
+ "loss": 0.0675,
3146
+ "step": 2240
3147
+ },
3148
+ {
3149
+ "epoch": 2.5807935595169638,
3150
+ "grad_norm": 0.6013872027397156,
3151
+ "learning_rate": 1.5847843139952205e-06,
3152
+ "loss": 0.0663,
3153
+ "step": 2245
3154
+ },
3155
+ {
3156
+ "epoch": 2.58654399079931,
3157
+ "grad_norm": 0.5785378217697144,
3158
+ "learning_rate": 1.5425328656144116e-06,
3159
+ "loss": 0.0661,
3160
+ "step": 2250
3161
+ },
3162
+ {
3163
+ "epoch": 2.592294422081656,
3164
+ "grad_norm": 0.6916348934173584,
3165
+ "learning_rate": 1.5008217341631093e-06,
3166
+ "loss": 0.0799,
3167
+ "step": 2255
3168
+ },
3169
+ {
3170
+ "epoch": 2.5980448533640024,
3171
+ "grad_norm": 0.7627856731414795,
3172
+ "learning_rate": 1.459652594342073e-06,
3173
+ "loss": 0.075,
3174
+ "step": 2260
3175
+ },
3176
+ {
3177
+ "epoch": 2.6037952846463486,
3178
+ "grad_norm": 0.7530559301376343,
3179
+ "learning_rate": 1.4190270990911198e-06,
3180
+ "loss": 0.0767,
3181
+ "step": 2265
3182
+ },
3183
+ {
3184
+ "epoch": 2.6095457159286948,
3185
+ "grad_norm": 0.6937177181243896,
3186
+ "learning_rate": 1.3789468795227495e-06,
3187
+ "loss": 0.0705,
3188
+ "step": 2270
3189
+ },
3190
+ {
3191
+ "epoch": 2.615296147211041,
3192
+ "grad_norm": 0.7135265469551086,
3193
+ "learning_rate": 1.3394135448566563e-06,
3194
+ "loss": 0.0796,
3195
+ "step": 2275
3196
+ },
3197
+ {
3198
+ "epoch": 2.621046578493387,
3199
+ "grad_norm": 0.46663105487823486,
3200
+ "learning_rate": 1.300428682355127e-06,
3201
+ "loss": 0.0756,
3202
+ "step": 2280
3203
+ },
3204
+ {
3205
+ "epoch": 2.626797009775733,
3206
+ "grad_norm": 0.5759928822517395,
3207
+ "learning_rate": 1.2619938572592965e-06,
3208
+ "loss": 0.0632,
3209
+ "step": 2285
3210
+ },
3211
+ {
3212
+ "epoch": 2.6325474410580796,
3213
+ "grad_norm": 0.533166766166687,
3214
+ "learning_rate": 1.2241106127263202e-06,
3215
+ "loss": 0.0688,
3216
+ "step": 2290
3217
+ },
3218
+ {
3219
+ "epoch": 2.6382978723404253,
3220
+ "grad_norm": 0.6144905090332031,
3221
+ "learning_rate": 1.1867804697674102e-06,
3222
+ "loss": 0.0623,
3223
+ "step": 2295
3224
+ },
3225
+ {
3226
+ "epoch": 2.6440483036227715,
3227
+ "grad_norm": 0.624462902545929,
3228
+ "learning_rate": 1.1500049271867585e-06,
3229
+ "loss": 0.0709,
3230
+ "step": 2300
3231
+ },
3232
+ {
3233
+ "epoch": 2.6497987349051177,
3234
+ "grad_norm": 0.6312700510025024,
3235
+ "learning_rate": 1.1137854615213783e-06,
3236
+ "loss": 0.0814,
3237
+ "step": 2305
3238
+ },
3239
+ {
3240
+ "epoch": 2.655549166187464,
3241
+ "grad_norm": 0.5629951357841492,
3242
+ "learning_rate": 1.0781235269817996e-06,
3243
+ "loss": 0.0676,
3244
+ "step": 2310
3245
+ },
3246
+ {
3247
+ "epoch": 2.66129959746981,
3248
+ "grad_norm": 0.7106200456619263,
3249
+ "learning_rate": 1.0430205553937023e-06,
3250
+ "loss": 0.0734,
3251
+ "step": 2315
3252
+ },
3253
+ {
3254
+ "epoch": 2.6670500287521564,
3255
+ "grad_norm": 0.7540964484214783,
3256
+ "learning_rate": 1.0084779561404095e-06,
3257
+ "loss": 0.0684,
3258
+ "step": 2320
3259
+ },
3260
+ {
3261
+ "epoch": 2.6728004600345026,
3262
+ "grad_norm": 0.6287196278572083,
3263
+ "learning_rate": 9.744971161063222e-07,
3264
+ "loss": 0.0711,
3265
+ "step": 2325
3266
+ },
3267
+ {
3268
+ "epoch": 2.6785508913168488,
3269
+ "grad_norm": 0.6894131898880005,
3270
+ "learning_rate": 9.410793996212142e-07,
3271
+ "loss": 0.0738,
3272
+ "step": 2330
3273
+ },
3274
+ {
3275
+ "epoch": 2.684301322599195,
3276
+ "grad_norm": 0.5116211771965027,
3277
+ "learning_rate": 9.082261484054722e-07,
3278
+ "loss": 0.0661,
3279
+ "step": 2335
3280
+ },
3281
+ {
3282
+ "epoch": 2.690051753881541,
3283
+ "grad_norm": 0.7554183006286621,
3284
+ "learning_rate": 8.759386815162096e-07,
3285
+ "loss": 0.0737,
3286
+ "step": 2340
3287
+ },
3288
+ {
3289
+ "epoch": 2.6958021851638874,
3290
+ "grad_norm": 0.691075325012207,
3291
+ "learning_rate": 8.442182952943261e-07,
3292
+ "loss": 0.0719,
3293
+ "step": 2345
3294
+ },
3295
+ {
3296
+ "epoch": 2.7015526164462336,
3297
+ "grad_norm": 0.5890049934387207,
3298
+ "learning_rate": 8.130662633124375e-07,
3299
+ "loss": 0.0774,
3300
+ "step": 2350
3301
+ },
3302
+ {
3303
+ "epoch": 2.7073030477285798,
3304
+ "grad_norm": 0.631833016872406,
3305
+ "learning_rate": 7.824838363237629e-07,
3306
+ "loss": 0.0702,
3307
+ "step": 2355
3308
+ },
3309
+ {
3310
+ "epoch": 2.713053479010926,
3311
+ "grad_norm": 0.6396310329437256,
3312
+ "learning_rate": 7.524722422118874e-07,
3313
+ "loss": 0.077,
3314
+ "step": 2360
3315
+ },
3316
+ {
3317
+ "epoch": 2.718803910293272,
3318
+ "grad_norm": 0.6222050189971924,
3319
+ "learning_rate": 7.230326859414754e-07,
3320
+ "loss": 0.0717,
3321
+ "step": 2365
3322
+ },
3323
+ {
3324
+ "epoch": 2.724554341575618,
3325
+ "grad_norm": 0.5621128082275391,
3326
+ "learning_rate": 6.941663495098921e-07,
3327
+ "loss": 0.065,
3328
+ "step": 2370
3329
+ },
3330
+ {
3331
+ "epoch": 2.7303047728579646,
3332
+ "grad_norm": 0.676668643951416,
3333
+ "learning_rate": 6.658743918997357e-07,
3334
+ "loss": 0.0742,
3335
+ "step": 2375
3336
+ },
3337
+ {
3338
+ "epoch": 2.7360552041403103,
3339
+ "grad_norm": 0.5283545255661011,
3340
+ "learning_rate": 6.381579490323164e-07,
3341
+ "loss": 0.0634,
3342
+ "step": 2380
3343
+ },
3344
+ {
3345
+ "epoch": 2.7418056354226565,
3346
+ "grad_norm": 0.5941113829612732,
3347
+ "learning_rate": 6.1101813372204e-07,
3348
+ "loss": 0.0694,
3349
+ "step": 2385
3350
+ },
3351
+ {
3352
+ "epoch": 2.7475560667050027,
3353
+ "grad_norm": 0.5411125421524048,
3354
+ "learning_rate": 5.844560356317347e-07,
3355
+ "loss": 0.0717,
3356
+ "step": 2390
3357
+ },
3358
+ {
3359
+ "epoch": 2.753306497987349,
3360
+ "grad_norm": 0.517797589302063,
3361
+ "learning_rate": 5.584727212288932e-07,
3362
+ "loss": 0.067,
3363
+ "step": 2395
3364
+ },
3365
+ {
3366
+ "epoch": 2.759056929269695,
3367
+ "grad_norm": 0.6308820843696594,
3368
+ "learning_rate": 5.330692337428667e-07,
3369
+ "loss": 0.067,
3370
+ "step": 2400
3371
+ },
3372
+ {
3373
+ "epoch": 2.7648073605520413,
3374
+ "grad_norm": 0.6886422038078308,
3375
+ "learning_rate": 5.082465931229651e-07,
3376
+ "loss": 0.0727,
3377
+ "step": 2405
3378
+ },
3379
+ {
3380
+ "epoch": 2.7705577918343876,
3381
+ "grad_norm": 0.5804277062416077,
3382
+ "learning_rate": 4.840057959975169e-07,
3383
+ "loss": 0.0693,
3384
+ "step": 2410
3385
+ },
3386
+ {
3387
+ "epoch": 2.7763082231167338,
3388
+ "grad_norm": 0.6014822125434875,
3389
+ "learning_rate": 4.6034781563385363e-07,
3390
+ "loss": 0.0755,
3391
+ "step": 2415
3392
+ },
3393
+ {
3394
+ "epoch": 2.78205865439908,
3395
+ "grad_norm": 0.5807937979698181,
3396
+ "learning_rate": 4.3727360189921683e-07,
3397
+ "loss": 0.0677,
3398
+ "step": 2420
3399
+ },
3400
+ {
3401
+ "epoch": 2.787809085681426,
3402
+ "grad_norm": 0.5600463151931763,
3403
+ "learning_rate": 4.1478408122265166e-07,
3404
+ "loss": 0.0646,
3405
+ "step": 2425
3406
+ },
3407
+ {
3408
+ "epoch": 2.7935595169637724,
3409
+ "grad_norm": 0.667402446269989,
3410
+ "learning_rate": 3.9288015655778167e-07,
3411
+ "loss": 0.0624,
3412
+ "step": 2430
3413
+ },
3414
+ {
3415
+ "epoch": 2.7993099482461186,
3416
+ "grad_norm": 0.6037558913230896,
3417
+ "learning_rate": 3.715627073465694e-07,
3418
+ "loss": 0.0671,
3419
+ "step": 2435
3420
+ },
3421
+ {
3422
+ "epoch": 2.8050603795284648,
3423
+ "grad_norm": 0.5630262494087219,
3424
+ "learning_rate": 3.5083258948400807e-07,
3425
+ "loss": 0.0601,
3426
+ "step": 2440
3427
+ },
3428
+ {
3429
+ "epoch": 2.810810810810811,
3430
+ "grad_norm": 0.6534041166305542,
3431
+ "learning_rate": 3.306906352837491e-07,
3432
+ "loss": 0.07,
3433
+ "step": 2445
3434
+ },
3435
+ {
3436
+ "epoch": 2.816561242093157,
3437
+ "grad_norm": 0.49078068137168884,
3438
+ "learning_rate": 3.1113765344469026e-07,
3439
+ "loss": 0.0727,
3440
+ "step": 2450
3441
+ },
3442
+ {
3443
+ "epoch": 2.822311673375503,
3444
+ "grad_norm": 0.5533818602561951,
3445
+ "learning_rate": 2.921744290185069e-07,
3446
+ "loss": 0.0622,
3447
+ "step": 2455
3448
+ },
3449
+ {
3450
+ "epoch": 2.8280621046578496,
3451
+ "grad_norm": 0.6338596343994141,
3452
+ "learning_rate": 2.7380172337813036e-07,
3453
+ "loss": 0.0705,
3454
+ "step": 2460
3455
+ },
3456
+ {
3457
+ "epoch": 2.8338125359401953,
3458
+ "grad_norm": 0.7202207446098328,
3459
+ "learning_rate": 2.560202741871792e-07,
3460
+ "loss": 0.073,
3461
+ "step": 2465
3462
+ },
3463
+ {
3464
+ "epoch": 2.8395629672225415,
3465
+ "grad_norm": 0.5659433603286743,
3466
+ "learning_rate": 2.388307953703395e-07,
3467
+ "loss": 0.073,
3468
+ "step": 2470
3469
+ },
3470
+ {
3471
+ "epoch": 2.8453133985048877,
3472
+ "grad_norm": 0.7026423811912537,
3473
+ "learning_rate": 2.222339770847076e-07,
3474
+ "loss": 0.0756,
3475
+ "step": 2475
3476
+ },
3477
+ {
3478
+ "epoch": 2.851063829787234,
3479
+ "grad_norm": 0.5251172184944153,
3480
+ "learning_rate": 2.0623048569207436e-07,
3481
+ "loss": 0.0659,
3482
+ "step": 2480
3483
+ },
3484
+ {
3485
+ "epoch": 2.85681426106958,
3486
+ "grad_norm": 0.6493549346923828,
3487
+ "learning_rate": 1.9082096373216952e-07,
3488
+ "loss": 0.0624,
3489
+ "step": 2485
3490
+ },
3491
+ {
3492
+ "epoch": 2.8625646923519263,
3493
+ "grad_norm": 0.6146339774131775,
3494
+ "learning_rate": 1.760060298968724e-07,
3495
+ "loss": 0.0699,
3496
+ "step": 2490
3497
+ },
3498
+ {
3499
+ "epoch": 2.8683151236342725,
3500
+ "grad_norm": 0.5148099064826965,
3501
+ "learning_rate": 1.6178627900536037e-07,
3502
+ "loss": 0.0685,
3503
+ "step": 2495
3504
+ },
3505
+ {
3506
+ "epoch": 2.8740655549166187,
3507
+ "grad_norm": 0.594165027141571,
3508
+ "learning_rate": 1.4816228198023585e-07,
3509
+ "loss": 0.0709,
3510
+ "step": 2500
3511
+ },
3512
+ {
3513
+ "epoch": 2.879815986198965,
3514
+ "grad_norm": 0.5928375124931335,
3515
+ "learning_rate": 1.3513458582459836e-07,
3516
+ "loss": 0.0589,
3517
+ "step": 2505
3518
+ },
3519
+ {
3520
+ "epoch": 2.885566417481311,
3521
+ "grad_norm": 0.5928677916526794,
3522
+ "learning_rate": 1.227037136000836e-07,
3523
+ "loss": 0.0667,
3524
+ "step": 2510
3525
+ },
3526
+ {
3527
+ "epoch": 2.8913168487636574,
3528
+ "grad_norm": 0.6581249833106995,
3529
+ "learning_rate": 1.1087016440586683e-07,
3530
+ "loss": 0.0661,
3531
+ "step": 2515
3532
+ },
3533
+ {
3534
+ "epoch": 2.8970672800460036,
3535
+ "grad_norm": 0.6517856121063232,
3536
+ "learning_rate": 9.963441335861578e-08,
3537
+ "loss": 0.073,
3538
+ "step": 2520
3539
+ },
3540
+ {
3541
+ "epoch": 2.9028177113283498,
3542
+ "grad_norm": 0.5155625939369202,
3543
+ "learning_rate": 8.899691157341905e-08,
3544
+ "loss": 0.0693,
3545
+ "step": 2525
3546
+ },
3547
+ {
3548
+ "epoch": 2.908568142610696,
3549
+ "grad_norm": 0.49900123476982117,
3550
+ "learning_rate": 7.895808614568067e-08,
3551
+ "loss": 0.0666,
3552
+ "step": 2530
3553
+ },
3554
+ {
3555
+ "epoch": 2.914318573893042,
3556
+ "grad_norm": 0.5401294231414795,
3557
+ "learning_rate": 6.951834013395719e-08,
3558
+ "loss": 0.0645,
3559
+ "step": 2535
3560
+ },
3561
+ {
3562
+ "epoch": 2.920069005175388,
3563
+ "grad_norm": 0.6005806922912598,
3564
+ "learning_rate": 6.067805254378889e-08,
3565
+ "loss": 0.0676,
3566
+ "step": 2540
3567
+ },
3568
+ {
3569
+ "epoch": 2.9258194364577346,
3570
+ "grad_norm": 0.8095978498458862,
3571
+ "learning_rate": 5.243757831247697e-08,
3572
+ "loss": 0.0646,
3573
+ "step": 2545
3574
+ },
3575
+ {
3576
+ "epoch": 2.9315698677400803,
3577
+ "grad_norm": 0.6773854494094849,
3578
+ "learning_rate": 4.479724829483001e-08,
3579
+ "loss": 0.0668,
3580
+ "step": 2550
3581
+ },
3582
+ {
3583
+ "epoch": 2.9373202990224265,
3584
+ "grad_norm": 0.628518283367157,
3585
+ "learning_rate": 3.775736924988449e-08,
3586
+ "loss": 0.0735,
3587
+ "step": 2555
3588
+ },
3589
+ {
3590
+ "epoch": 2.9430707303047727,
3591
+ "grad_norm": 0.7223976850509644,
3592
+ "learning_rate": 3.131822382858807e-08,
3593
+ "loss": 0.0833,
3594
+ "step": 2560
3595
+ },
3596
+ {
3597
+ "epoch": 2.948821161587119,
3598
+ "grad_norm": 0.6249869465827942,
3599
+ "learning_rate": 2.548007056244861e-08,
3600
+ "loss": 0.0732,
3601
+ "step": 2565
3602
+ },
3603
+ {
3604
+ "epoch": 2.954571592869465,
3605
+ "grad_norm": 0.481516033411026,
3606
+ "learning_rate": 2.024314385315751e-08,
3607
+ "loss": 0.0706,
3608
+ "step": 2570
3609
+ },
3610
+ {
3611
+ "epoch": 2.9603220241518113,
3612
+ "grad_norm": 0.5232948064804077,
3613
+ "learning_rate": 1.5607653963177203e-08,
3614
+ "loss": 0.0636,
3615
+ "step": 2575
3616
+ },
3617
+ {
3618
+ "epoch": 2.9660724554341575,
3619
+ "grad_norm": 0.5923638939857483,
3620
+ "learning_rate": 1.1573787007294611e-08,
3621
+ "loss": 0.0726,
3622
+ "step": 2580
3623
+ },
3624
+ {
3625
+ "epoch": 2.9718228867165037,
3626
+ "grad_norm": 0.5257419943809509,
3627
+ "learning_rate": 8.141704945155425e-09,
3628
+ "loss": 0.0637,
3629
+ "step": 2585
3630
+ },
3631
+ {
3632
+ "epoch": 2.97757331799885,
3633
+ "grad_norm": 0.555911123752594,
3634
+ "learning_rate": 5.31154557476099e-09,
3635
+ "loss": 0.0722,
3636
+ "step": 2590
3637
+ },
3638
+ {
3639
+ "epoch": 2.983323749281196,
3640
+ "grad_norm": 0.5521672964096069,
3641
+ "learning_rate": 3.083422526931057e-09,
3642
+ "loss": 0.064,
3643
+ "step": 2595
3644
+ },
3645
+ {
3646
+ "epoch": 2.9890741805635423,
3647
+ "grad_norm": 0.5197116136550903,
3648
+ "learning_rate": 1.4574252607407745e-09,
3649
+ "loss": 0.0624,
3650
+ "step": 2600
3651
+ },
3652
+ {
3653
+ "epoch": 2.9948246118458886,
3654
+ "grad_norm": 0.5989261865615845,
3655
+ "learning_rate": 4.3361905993521966e-10,
3656
+ "loss": 0.0675,
3657
+ "step": 2605
3658
+ },
3659
+ {
3660
+ "epoch": 3.0,
3661
+ "grad_norm": 0.5791131854057312,
3662
+ "learning_rate": 1.2045030308160954e-11,
3663
+ "loss": 0.0626,
3664
+ "step": 2610
3665
+ },
3666
+ {
3667
+ "epoch": 3.0,
3668
+ "step": 2610,
3669
+ "total_flos": 3.4112320082133647e+18,
3670
+ "train_loss": 0.3677417017376743,
3671
+ "train_runtime": 1618.6418,
3672
+ "train_samples_per_second": 51.558,
3673
+ "train_steps_per_second": 1.612
3674
+ }
3675
+ ],
3676
+ "logging_steps": 5,
3677
+ "max_steps": 2610,
3678
+ "num_input_tokens_seen": 0,
3679
+ "num_train_epochs": 3,
3680
+ "save_steps": 20000,
3681
+ "stateful_callbacks": {
3682
+ "TrainerControl": {
3683
+ "args": {
3684
+ "should_epoch_stop": false,
3685
+ "should_evaluate": false,
3686
+ "should_log": false,
3687
+ "should_save": false,
3688
+ "should_training_stop": false
3689
+ },
3690
+ "attributes": {}
3691
+ }
3692
+ },
3693
+ "total_flos": 3.4112320082133647e+18,
3694
+ "train_batch_size": 2,
3695
+ "trial_name": null,
3696
+ "trial_params": null
3697
+ }
instruct/140_128_e3_3e-5/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38ba55f72a79656b0aefb09deca7739beebc43c5dbc456eb566e802ce5f64cc9
3
+ size 8273
instruct/140_128_e3_3e-5/vocab.json ADDED
The diff for this file is too large to render. See raw diff