Hugging Face's logo

RayDu0010
/
tulu10k_klora

Safetensors
Model card Files
xet
 Community
RayDu0010 commited on
Commit
aef707a
·
verified ·
1 Parent(s): 720a007

Upload folder using huggingface_hub

Browse files
Files changed (14) hide show
  1. instruct/102_128_e3_3e-5/adapter_config.json +39 -0
  2. instruct/102_128_e3_3e-5/adapter_model.safetensors +3 -0
  3. instruct/102_128_e3_3e-5/added_tokens.json +9 -0
  4. instruct/102_128_e3_3e-5/all_results.json +9 -0
  5. instruct/102_128_e3_3e-5/chat_template.jinja +62 -0
  6. instruct/102_128_e3_3e-5/config.json +32 -0
  7. instruct/102_128_e3_3e-5/merges.txt +0 -0
  8. instruct/102_128_e3_3e-5/special_tokens_map.json +33 -0
  9. instruct/102_128_e3_3e-5/tokenizer.json +0 -0
  10. instruct/102_128_e3_3e-5/tokenizer_config.json +234 -0
  11. instruct/102_128_e3_3e-5/train_results.json +9 -0
  12. instruct/102_128_e3_3e-5/trainer_state.json +3081 -0
  13. instruct/102_128_e3_3e-5/training_args.bin +3 -0
  14. instruct/102_128_e3_3e-5/vocab.json +0 -0
instruct/102_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
+ "down_proj",
28
+ "gate_proj",
29
+ "q_proj",
30
+ "up_proj",
31
+ "k_proj",
32
+ "o_proj",
33
+ "v_proj"
34
+ ],
35
+ "task_type": "CAUSAL_LM",
36
+ "trainable_token_indices": null,
37
+ "use_dora": false,
38
+ "use_rslora": false
39
+ }
instruct/102_128_e3_3e-5/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60a99796e2afa80a67936b322262b8dbdbc4d7c77e5b4cc324b0c448996e6452
3
+ size 791751704
instruct/102_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/102_128_e3_3e-5/all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 2.72283124759108e+18,
4
+ "train_loss": 0.35041305541470785,
5
+ "train_runtime": 1331.4434,
6
+ "train_samples": 23155,
7
+ "train_samples_per_second": 52.173,
8
+ "train_steps_per_second": 1.631
9
+ }
instruct/102_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/102_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/102_128_e3_3e-5/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
instruct/102_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/102_128_e3_3e-5/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
instruct/102_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/102_128_e3_3e-5/train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 2.72283124759108e+18,
4
+ "train_loss": 0.35041305541470785,
5
+ "train_runtime": 1331.4434,
6
+ "train_samples": 23155,
7
+ "train_samples_per_second": 52.173,
8
+ "train_steps_per_second": 1.631
9
+ }
instruct/102_128_e3_3e-5/trainer_state.json ADDED
@@ -0,0 +1,3081 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 2172,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.006906077348066298,
14
+ "grad_norm": 2.224489450454712,
15
+ "learning_rate": 1.1009174311926605e-06,
16
+ "loss": 1.5644,
17
+ "step": 5
18
+ },
19
+ {
20
+ "epoch": 0.013812154696132596,
21
+ "grad_norm": 2.1004650592803955,
22
+ "learning_rate": 2.4770642201834866e-06,
23
+ "loss": 1.5104,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 0.020718232044198894,
28
+ "grad_norm": 1.5441694259643555,
29
+ "learning_rate": 3.853211009174312e-06,
30
+ "loss": 1.5151,
31
+ "step": 15
32
+ },
33
+ {
34
+ "epoch": 0.027624309392265192,
35
+ "grad_norm": 1.125309944152832,
36
+ "learning_rate": 5.229357798165138e-06,
37
+ "loss": 1.5109,
38
+ "step": 20
39
+ },
40
+ {
41
+ "epoch": 0.034530386740331494,
42
+ "grad_norm": 0.5893253684043884,
43
+ "learning_rate": 6.605504587155964e-06,
44
+ "loss": 1.4704,
45
+ "step": 25
46
+ },
47
+ {
48
+ "epoch": 0.04143646408839779,
49
+ "grad_norm": 0.5172114372253418,
50
+ "learning_rate": 7.981651376146789e-06,
51
+ "loss": 1.3797,
52
+ "step": 30
53
+ },
54
+ {
55
+ "epoch": 0.04834254143646409,
56
+ "grad_norm": 0.46956631541252136,
57
+ "learning_rate": 9.357798165137616e-06,
58
+ "loss": 1.3659,
59
+ "step": 35
60
+ },
61
+ {
62
+ "epoch": 0.055248618784530384,
63
+ "grad_norm": 5.122790813446045,
64
+ "learning_rate": 1.0733944954128442e-05,
65
+ "loss": 1.3903,
66
+ "step": 40
67
+ },
68
+ {
69
+ "epoch": 0.062154696132596686,
70
+ "grad_norm": 0.44037511944770813,
71
+ "learning_rate": 1.2110091743119267e-05,
72
+ "loss": 1.4508,
73
+ "step": 45
74
+ },
75
+ {
76
+ "epoch": 0.06906077348066299,
77
+ "grad_norm": 0.4232449233531952,
78
+ "learning_rate": 1.3486238532110092e-05,
79
+ "loss": 1.4045,
80
+ "step": 50
81
+ },
82
+ {
83
+ "epoch": 0.07596685082872928,
84
+ "grad_norm": 0.38281503319740295,
85
+ "learning_rate": 1.4862385321100918e-05,
86
+ "loss": 1.3533,
87
+ "step": 55
88
+ },
89
+ {
90
+ "epoch": 0.08287292817679558,
91
+ "grad_norm": 0.409792959690094,
92
+ "learning_rate": 1.6238532110091743e-05,
93
+ "loss": 1.3164,
94
+ "step": 60
95
+ },
96
+ {
97
+ "epoch": 0.08977900552486189,
98
+ "grad_norm": 0.4131454527378082,
99
+ "learning_rate": 1.761467889908257e-05,
100
+ "loss": 1.352,
101
+ "step": 65
102
+ },
103
+ {
104
+ "epoch": 0.09668508287292818,
105
+ "grad_norm": 0.5110914707183838,
106
+ "learning_rate": 1.8990825688073394e-05,
107
+ "loss": 1.3091,
108
+ "step": 70
109
+ },
110
+ {
111
+ "epoch": 0.10359116022099447,
112
+ "grad_norm": 0.4310168921947479,
113
+ "learning_rate": 2.0366972477064223e-05,
114
+ "loss": 1.3195,
115
+ "step": 75
116
+ },
117
+ {
118
+ "epoch": 0.11049723756906077,
119
+ "grad_norm": 0.5088695883750916,
120
+ "learning_rate": 2.1743119266055045e-05,
121
+ "loss": 1.2956,
122
+ "step": 80
123
+ },
124
+ {
125
+ "epoch": 0.11740331491712708,
126
+ "grad_norm": 0.44048404693603516,
127
+ "learning_rate": 2.3119266055045874e-05,
128
+ "loss": 1.2763,
129
+ "step": 85
130
+ },
131
+ {
132
+ "epoch": 0.12430939226519337,
133
+ "grad_norm": 0.49699866771698,
134
+ "learning_rate": 2.44954128440367e-05,
135
+ "loss": 1.2061,
136
+ "step": 90
137
+ },
138
+ {
139
+ "epoch": 0.13121546961325967,
140
+ "grad_norm": 0.5434209108352661,
141
+ "learning_rate": 2.5871559633027525e-05,
142
+ "loss": 1.2433,
143
+ "step": 95
144
+ },
145
+ {
146
+ "epoch": 0.13812154696132597,
147
+ "grad_norm": 0.4799324870109558,
148
+ "learning_rate": 2.724770642201835e-05,
149
+ "loss": 1.2981,
150
+ "step": 100
151
+ },
152
+ {
153
+ "epoch": 0.14502762430939226,
154
+ "grad_norm": 0.4849097728729248,
155
+ "learning_rate": 2.8623853211009175e-05,
156
+ "loss": 1.2277,
157
+ "step": 105
158
+ },
159
+ {
160
+ "epoch": 0.15193370165745856,
161
+ "grad_norm": 0.6652722358703613,
162
+ "learning_rate": 3e-05,
163
+ "loss": 1.2524,
164
+ "step": 110
165
+ },
166
+ {
167
+ "epoch": 0.15883977900552487,
168
+ "grad_norm": 0.5290014743804932,
169
+ "learning_rate": 2.9999565189059242e-05,
170
+ "loss": 1.2142,
171
+ "step": 115
172
+ },
173
+ {
174
+ "epoch": 0.16574585635359115,
175
+ "grad_norm": 0.5514317154884338,
176
+ "learning_rate": 2.9998260781445032e-05,
177
+ "loss": 1.1639,
178
+ "step": 120
179
+ },
180
+ {
181
+ "epoch": 0.17265193370165746,
182
+ "grad_norm": 0.5237447619438171,
183
+ "learning_rate": 2.9996086852780138e-05,
184
+ "loss": 1.2249,
185
+ "step": 125
186
+ },
187
+ {
188
+ "epoch": 0.17955801104972377,
189
+ "grad_norm": 0.5386995673179626,
190
+ "learning_rate": 2.9993043529097623e-05,
191
+ "loss": 1.1988,
192
+ "step": 130
193
+ },
194
+ {
195
+ "epoch": 0.18646408839779005,
196
+ "grad_norm": 0.6563480496406555,
197
+ "learning_rate": 2.998913098683354e-05,
198
+ "loss": 1.1562,
199
+ "step": 135
200
+ },
201
+ {
202
+ "epoch": 0.19337016574585636,
203
+ "grad_norm": 0.5946796536445618,
204
+ "learning_rate": 2.9984349452816713e-05,
205
+ "loss": 1.1905,
206
+ "step": 140
207
+ },
208
+ {
209
+ "epoch": 0.20027624309392264,
210
+ "grad_norm": 0.664196789264679,
211
+ "learning_rate": 2.9978699204255584e-05,
212
+ "loss": 1.0691,
213
+ "step": 145
214
+ },
215
+ {
216
+ "epoch": 0.20718232044198895,
217
+ "grad_norm": 0.6454105377197266,
218
+ "learning_rate": 2.9972180568722142e-05,
219
+ "loss": 1.1309,
220
+ "step": 150
221
+ },
222
+ {
223
+ "epoch": 0.21408839779005526,
224
+ "grad_norm": 0.631622850894928,
225
+ "learning_rate": 2.9964793924132924e-05,
226
+ "loss": 1.1142,
227
+ "step": 155
228
+ },
229
+ {
230
+ "epoch": 0.22099447513812154,
231
+ "grad_norm": 0.7202966213226318,
232
+ "learning_rate": 2.9956539698727114e-05,
233
+ "loss": 1.1244,
234
+ "step": 160
235
+ },
236
+ {
237
+ "epoch": 0.22790055248618785,
238
+ "grad_norm": 0.6964497566223145,
239
+ "learning_rate": 2.9947418371041712e-05,
240
+ "loss": 1.0802,
241
+ "step": 165
242
+ },
243
+ {
244
+ "epoch": 0.23480662983425415,
245
+ "grad_norm": 0.7859070897102356,
246
+ "learning_rate": 2.99374304698838e-05,
247
+ "loss": 1.0965,
248
+ "step": 170
249
+ },
250
+ {
251
+ "epoch": 0.24171270718232044,
252
+ "grad_norm": 0.6994317770004272,
253
+ "learning_rate": 2.9926576574299868e-05,
254
+ "loss": 1.0631,
255
+ "step": 175
256
+ },
257
+ {
258
+ "epoch": 0.24861878453038674,
259
+ "grad_norm": 0.7665867805480957,
260
+ "learning_rate": 2.9914857313542252e-05,
261
+ "loss": 1.0767,
262
+ "step": 180
263
+ },
264
+ {
265
+ "epoch": 0.255524861878453,
266
+ "grad_norm": 0.8231821060180664,
267
+ "learning_rate": 2.990227336703266e-05,
268
+ "loss": 0.9737,
269
+ "step": 185
270
+ },
271
+ {
272
+ "epoch": 0.26243093922651933,
273
+ "grad_norm": 0.7017345428466797,
274
+ "learning_rate": 2.9888825464322783e-05,
275
+ "loss": 0.9826,
276
+ "step": 190
277
+ },
278
+ {
279
+ "epoch": 0.26933701657458564,
280
+ "grad_norm": 0.7522607445716858,
281
+ "learning_rate": 2.9874514385051974e-05,
282
+ "loss": 1.024,
283
+ "step": 195
284
+ },
285
+ {
286
+ "epoch": 0.27624309392265195,
287
+ "grad_norm": 0.8091436624526978,
288
+ "learning_rate": 2.9859340958902086e-05,
289
+ "loss": 0.9739,
290
+ "step": 200
291
+ },
292
+ {
293
+ "epoch": 0.28314917127071826,
294
+ "grad_norm": 0.7903576493263245,
295
+ "learning_rate": 2.984330606554934e-05,
296
+ "loss": 0.9918,
297
+ "step": 205
298
+ },
299
+ {
300
+ "epoch": 0.2900552486187845,
301
+ "grad_norm": 0.7745909690856934,
302
+ "learning_rate": 2.9826410634613347e-05,
303
+ "loss": 0.9968,
304
+ "step": 210
305
+ },
306
+ {
307
+ "epoch": 0.2969613259668508,
308
+ "grad_norm": 0.809971034526825,
309
+ "learning_rate": 2.9808655645603205e-05,
310
+ "loss": 1.0183,
311
+ "step": 215
312
+ },
313
+ {
314
+ "epoch": 0.30386740331491713,
315
+ "grad_norm": 0.8054702877998352,
316
+ "learning_rate": 2.9790042127860703e-05,
317
+ "loss": 1.0071,
318
+ "step": 220
319
+ },
320
+ {
321
+ "epoch": 0.31077348066298344,
322
+ "grad_norm": 0.8940940499305725,
323
+ "learning_rate": 2.9770571160500668e-05,
324
+ "loss": 0.9229,
325
+ "step": 225
326
+ },
327
+ {
328
+ "epoch": 0.31767955801104975,
329
+ "grad_norm": 0.8612704277038574,
330
+ "learning_rate": 2.975024387234839e-05,
331
+ "loss": 0.9647,
332
+ "step": 230
333
+ },
334
+ {
335
+ "epoch": 0.324585635359116,
336
+ "grad_norm": 0.8931786417961121,
337
+ "learning_rate": 2.9729061441874164e-05,
338
+ "loss": 0.9055,
339
+ "step": 235
340
+ },
341
+ {
342
+ "epoch": 0.3314917127071823,
343
+ "grad_norm": 0.9774237275123596,
344
+ "learning_rate": 2.9707025097125e-05,
345
+ "loss": 0.9274,
346
+ "step": 240
347
+ },
348
+ {
349
+ "epoch": 0.3383977900552486,
350
+ "grad_norm": 0.8378453254699707,
351
+ "learning_rate": 2.96841361156534e-05,
352
+ "loss": 0.924,
353
+ "step": 245
354
+ },
355
+ {
356
+ "epoch": 0.3453038674033149,
357
+ "grad_norm": 0.8936626315116882,
358
+ "learning_rate": 2.96603958244433e-05,
359
+ "loss": 0.853,
360
+ "step": 250
361
+ },
362
+ {
363
+ "epoch": 0.35220994475138123,
364
+ "grad_norm": 0.9428366422653198,
365
+ "learning_rate": 2.9635805599833163e-05,
366
+ "loss": 0.915,
367
+ "step": 255
368
+ },
369
+ {
370
+ "epoch": 0.35911602209944754,
371
+ "grad_norm": 0.9703483581542969,
372
+ "learning_rate": 2.9610366867436135e-05,
373
+ "loss": 0.8444,
374
+ "step": 260
375
+ },
376
+ {
377
+ "epoch": 0.3660220994475138,
378
+ "grad_norm": 0.9037713408470154,
379
+ "learning_rate": 2.9584081102057443e-05,
380
+ "loss": 0.8645,
381
+ "step": 265
382
+ },
383
+ {
384
+ "epoch": 0.3729281767955801,
385
+ "grad_norm": 0.8838475942611694,
386
+ "learning_rate": 2.9556949827608868e-05,
387
+ "loss": 0.8835,
388
+ "step": 270
389
+ },
390
+ {
391
+ "epoch": 0.3798342541436464,
392
+ "grad_norm": 1.020803451538086,
393
+ "learning_rate": 2.9528974617020412e-05,
394
+ "loss": 0.8831,
395
+ "step": 275
396
+ },
397
+ {
398
+ "epoch": 0.3867403314917127,
399
+ "grad_norm": 0.9760249853134155,
400
+ "learning_rate": 2.9500157092149085e-05,
401
+ "loss": 0.7919,
402
+ "step": 280
403
+ },
404
+ {
405
+ "epoch": 0.393646408839779,
406
+ "grad_norm": 0.9419657588005066,
407
+ "learning_rate": 2.9470498923684902e-05,
408
+ "loss": 0.8455,
409
+ "step": 285
410
+ },
411
+ {
412
+ "epoch": 0.4005524861878453,
413
+ "grad_norm": 0.8996731638908386,
414
+ "learning_rate": 2.944000183105402e-05,
415
+ "loss": 0.8772,
416
+ "step": 290
417
+ },
418
+ {
419
+ "epoch": 0.4074585635359116,
420
+ "grad_norm": 1.0299211740493774,
421
+ "learning_rate": 2.9408667582319032e-05,
422
+ "loss": 0.8156,
423
+ "step": 295
424
+ },
425
+ {
426
+ "epoch": 0.4143646408839779,
427
+ "grad_norm": 1.1469181776046753,
428
+ "learning_rate": 2.9376497994076506e-05,
429
+ "loss": 0.8047,
430
+ "step": 300
431
+ },
432
+ {
433
+ "epoch": 0.4212707182320442,
434
+ "grad_norm": 0.9643601179122925,
435
+ "learning_rate": 2.934349493135163e-05,
436
+ "loss": 0.8013,
437
+ "step": 305
438
+ },
439
+ {
440
+ "epoch": 0.4281767955801105,
441
+ "grad_norm": 0.9881924986839294,
442
+ "learning_rate": 2.9309660307490096e-05,
443
+ "loss": 0.8039,
444
+ "step": 310
445
+ },
446
+ {
447
+ "epoch": 0.4350828729281768,
448
+ "grad_norm": 1.0978139638900757,
449
+ "learning_rate": 2.9274996084047196e-05,
450
+ "loss": 0.7978,
451
+ "step": 315
452
+ },
453
+ {
454
+ "epoch": 0.4419889502762431,
455
+ "grad_norm": 0.9411031603813171,
456
+ "learning_rate": 2.9239504270674075e-05,
457
+ "loss": 0.763,
458
+ "step": 320
459
+ },
460
+ {
461
+ "epoch": 0.4488950276243094,
462
+ "grad_norm": 1.0262761116027832,
463
+ "learning_rate": 2.920318692500124e-05,
464
+ "loss": 0.7677,
465
+ "step": 325
466
+ },
467
+ {
468
+ "epoch": 0.4558011049723757,
469
+ "grad_norm": 1.0894794464111328,
470
+ "learning_rate": 2.9166046152519247e-05,
471
+ "loss": 0.734,
472
+ "step": 330
473
+ },
474
+ {
475
+ "epoch": 0.462707182320442,
476
+ "grad_norm": 1.0422154664993286,
477
+ "learning_rate": 2.9128084106456662e-05,
478
+ "loss": 0.7612,
479
+ "step": 335
480
+ },
481
+ {
482
+ "epoch": 0.4696132596685083,
483
+ "grad_norm": 1.0151069164276123,
484
+ "learning_rate": 2.9089302987655214e-05,
485
+ "loss": 0.7593,
486
+ "step": 340
487
+ },
488
+ {
489
+ "epoch": 0.47651933701657456,
490
+ "grad_norm": 0.9965788722038269,
491
+ "learning_rate": 2.9049705044442207e-05,
492
+ "loss": 0.7611,
493
+ "step": 345
494
+ },
495
+ {
496
+ "epoch": 0.48342541436464087,
497
+ "grad_norm": 0.9909919500350952,
498
+ "learning_rate": 2.9009292572500155e-05,
499
+ "loss": 0.7749,
500
+ "step": 350
501
+ },
502
+ {
503
+ "epoch": 0.4903314917127072,
504
+ "grad_norm": 1.0382671356201172,
505
+ "learning_rate": 2.8968067914733734e-05,
506
+ "loss": 0.7264,
507
+ "step": 355
508
+ },
509
+ {
510
+ "epoch": 0.4972375690607735,
511
+ "grad_norm": 1.0746433734893799,
512
+ "learning_rate": 2.8926033461133895e-05,
513
+ "loss": 0.7682,
514
+ "step": 360
515
+ },
516
+ {
517
+ "epoch": 0.5041436464088398,
518
+ "grad_norm": 1.105519413948059,
519
+ "learning_rate": 2.8883191648639352e-05,
520
+ "loss": 0.6902,
521
+ "step": 365
522
+ },
523
+ {
524
+ "epoch": 0.511049723756906,
525
+ "grad_norm": 1.0307425260543823,
526
+ "learning_rate": 2.883954496099527e-05,
527
+ "loss": 0.7472,
528
+ "step": 370
529
+ },
530
+ {
531
+ "epoch": 0.5179558011049724,
532
+ "grad_norm": 1.0924628973007202,
533
+ "learning_rate": 2.8795095928609302e-05,
534
+ "loss": 0.7523,
535
+ "step": 375
536
+ },
537
+ {
538
+ "epoch": 0.5248618784530387,
539
+ "grad_norm": 1.0584372282028198,
540
+ "learning_rate": 2.8749847128404857e-05,
541
+ "loss": 0.7284,
542
+ "step": 380
543
+ },
544
+ {
545
+ "epoch": 0.5317679558011049,
546
+ "grad_norm": 1.0530767440795898,
547
+ "learning_rate": 2.870380118367171e-05,
548
+ "loss": 0.7381,
549
+ "step": 385
550
+ },
551
+ {
552
+ "epoch": 0.5386740331491713,
553
+ "grad_norm": 1.0420029163360596,
554
+ "learning_rate": 2.8656960763913944e-05,
555
+ "loss": 0.6886,
556
+ "step": 390
557
+ },
558
+ {
559
+ "epoch": 0.5455801104972375,
560
+ "grad_norm": 1.0296063423156738,
561
+ "learning_rate": 2.8609328584695155e-05,
562
+ "loss": 0.658,
563
+ "step": 395
564
+ },
565
+ {
566
+ "epoch": 0.5524861878453039,
567
+ "grad_norm": 1.056171178817749,
568
+ "learning_rate": 2.8560907407481027e-05,
569
+ "loss": 0.6526,
570
+ "step": 400
571
+ },
572
+ {
573
+ "epoch": 0.5593922651933702,
574
+ "grad_norm": 1.1418228149414062,
575
+ "learning_rate": 2.8511700039479244e-05,
576
+ "loss": 0.6538,
577
+ "step": 405
578
+ },
579
+ {
580
+ "epoch": 0.5662983425414365,
581
+ "grad_norm": 1.093884825706482,
582
+ "learning_rate": 2.846170933347674e-05,
583
+ "loss": 0.6758,
584
+ "step": 410
585
+ },
586
+ {
587
+ "epoch": 0.5732044198895028,
588
+ "grad_norm": 1.155714750289917,
589
+ "learning_rate": 2.841093818767429e-05,
590
+ "loss": 0.6296,
591
+ "step": 415
592
+ },
593
+ {
594
+ "epoch": 0.580110497237569,
595
+ "grad_norm": 1.3182930946350098,
596
+ "learning_rate": 2.8359389545518534e-05,
597
+ "loss": 0.6535,
598
+ "step": 420
599
+ },
600
+ {
601
+ "epoch": 0.5870165745856354,
602
+ "grad_norm": 1.0659617185592651,
603
+ "learning_rate": 2.8307066395531273e-05,
604
+ "loss": 0.6463,
605
+ "step": 425
606
+ },
607
+ {
608
+ "epoch": 0.5939226519337016,
609
+ "grad_norm": 1.1250925064086914,
610
+ "learning_rate": 2.825397177113625e-05,
611
+ "loss": 0.689,
612
+ "step": 430
613
+ },
614
+ {
615
+ "epoch": 0.600828729281768,
616
+ "grad_norm": 1.1664258241653442,
617
+ "learning_rate": 2.8200108750483277e-05,
618
+ "loss": 0.6198,
619
+ "step": 435
620
+ },
621
+ {
622
+ "epoch": 0.6077348066298343,
623
+ "grad_norm": 1.2420212030410767,
624
+ "learning_rate": 2.814548045626978e-05,
625
+ "loss": 0.6201,
626
+ "step": 440
627
+ },
628
+ {
629
+ "epoch": 0.6146408839779005,
630
+ "grad_norm": 1.1614890098571777,
631
+ "learning_rate": 2.809009005555976e-05,
632
+ "loss": 0.5856,
633
+ "step": 445
634
+ },
635
+ {
636
+ "epoch": 0.6215469613259669,
637
+ "grad_norm": 1.0308414697647095,
638
+ "learning_rate": 2.803394075960018e-05,
639
+ "loss": 0.5769,
640
+ "step": 450
641
+ },
642
+ {
643
+ "epoch": 0.6284530386740331,
644
+ "grad_norm": 1.1657791137695312,
645
+ "learning_rate": 2.7977035823634803e-05,
646
+ "loss": 0.5552,
647
+ "step": 455
648
+ },
649
+ {
650
+ "epoch": 0.6353591160220995,
651
+ "grad_norm": 1.214986801147461,
652
+ "learning_rate": 2.791937854671545e-05,
653
+ "loss": 0.5985,
654
+ "step": 460
655
+ },
656
+ {
657
+ "epoch": 0.6422651933701657,
658
+ "grad_norm": 1.0663727521896362,
659
+ "learning_rate": 2.7860972271510783e-05,
660
+ "loss": 0.5929,
661
+ "step": 465
662
+ },
663
+ {
664
+ "epoch": 0.649171270718232,
665
+ "grad_norm": 1.0847468376159668,
666
+ "learning_rate": 2.7801820384112447e-05,
667
+ "loss": 0.5714,
668
+ "step": 470
669
+ },
670
+ {
671
+ "epoch": 0.6560773480662984,
672
+ "grad_norm": 1.2035939693450928,
673
+ "learning_rate": 2.7741926313838823e-05,
674
+ "loss": 0.5513,
675
+ "step": 475
676
+ },
677
+ {
678
+ "epoch": 0.6629834254143646,
679
+ "grad_norm": 1.1500974893569946,
680
+ "learning_rate": 2.7681293533036187e-05,
681
+ "loss": 0.575,
682
+ "step": 480
683
+ },
684
+ {
685
+ "epoch": 0.669889502762431,
686
+ "grad_norm": 1.170586109161377,
687
+ "learning_rate": 2.7619925556877394e-05,
688
+ "loss": 0.6289,
689
+ "step": 485
690
+ },
691
+ {
692
+ "epoch": 0.6767955801104972,
693
+ "grad_norm": 1.050852656364441,
694
+ "learning_rate": 2.7557825943158103e-05,
695
+ "loss": 0.5448,
696
+ "step": 490
697
+ },
698
+ {
699
+ "epoch": 0.6837016574585635,
700
+ "grad_norm": 1.1791731119155884,
701
+ "learning_rate": 2.7494998292090518e-05,
702
+ "loss": 0.5715,
703
+ "step": 495
704
+ },
705
+ {
706
+ "epoch": 0.6906077348066298,
707
+ "grad_norm": 1.0975124835968018,
708
+ "learning_rate": 2.743144624609464e-05,
709
+ "loss": 0.494,
710
+ "step": 500
711
+ },
712
+ {
713
+ "epoch": 0.6975138121546961,
714
+ "grad_norm": 1.2437090873718262,
715
+ "learning_rate": 2.7367173489587128e-05,
716
+ "loss": 0.5333,
717
+ "step": 505
718
+ },
719
+ {
720
+ "epoch": 0.7044198895027625,
721
+ "grad_norm": 1.225940227508545,
722
+ "learning_rate": 2.730218374876767e-05,
723
+ "loss": 0.5222,
724
+ "step": 510
725
+ },
726
+ {
727
+ "epoch": 0.7113259668508287,
728
+ "grad_norm": 1.2462373971939087,
729
+ "learning_rate": 2.7236480791402986e-05,
730
+ "loss": 0.5197,
731
+ "step": 515
732
+ },
733
+ {
734
+ "epoch": 0.7182320441988951,
735
+ "grad_norm": 1.1254507303237915,
736
+ "learning_rate": 2.717006842660837e-05,
737
+ "loss": 0.5589,
738
+ "step": 520
739
+ },
740
+ {
741
+ "epoch": 0.7251381215469613,
742
+ "grad_norm": 1.1946864128112793,
743
+ "learning_rate": 2.710295050462686e-05,
744
+ "loss": 0.5408,
745
+ "step": 525
746
+ },
747
+ {
748
+ "epoch": 0.7320441988950276,
749
+ "grad_norm": 1.2211350202560425,
750
+ "learning_rate": 2.703513091660603e-05,
751
+ "loss": 0.4679,
752
+ "step": 530
753
+ },
754
+ {
755
+ "epoch": 0.738950276243094,
756
+ "grad_norm": 1.4209901094436646,
757
+ "learning_rate": 2.69666135943724e-05,
758
+ "loss": 0.5483,
759
+ "step": 535
760
+ },
761
+ {
762
+ "epoch": 0.7458563535911602,
763
+ "grad_norm": 1.2153788805007935,
764
+ "learning_rate": 2.689740251020348e-05,
765
+ "loss": 0.536,
766
+ "step": 540
767
+ },
768
+ {
769
+ "epoch": 0.7527624309392266,
770
+ "grad_norm": 1.160967230796814,
771
+ "learning_rate": 2.682750167659748e-05,
772
+ "loss": 0.5351,
773
+ "step": 545
774
+ },
775
+ {
776
+ "epoch": 0.7596685082872928,
777
+ "grad_norm": 1.3381195068359375,
778
+ "learning_rate": 2.67569151460407e-05,
779
+ "loss": 0.4818,
780
+ "step": 550
781
+ },
782
+ {
783
+ "epoch": 0.7665745856353591,
784
+ "grad_norm": 1.2260489463806152,
785
+ "learning_rate": 2.6685647010772578e-05,
786
+ "loss": 0.5112,
787
+ "step": 555
788
+ },
789
+ {
790
+ "epoch": 0.7734806629834254,
791
+ "grad_norm": 1.2252088785171509,
792
+ "learning_rate": 2.6613701402548436e-05,
793
+ "loss": 0.4913,
794
+ "step": 560
795
+ },
796
+ {
797
+ "epoch": 0.7803867403314917,
798
+ "grad_norm": 1.1265445947647095,
799
+ "learning_rate": 2.6541082492399954e-05,
800
+ "loss": 0.5105,
801
+ "step": 565
802
+ },
803
+ {
804
+ "epoch": 0.787292817679558,
805
+ "grad_norm": 1.2992305755615234,
806
+ "learning_rate": 2.646779449039335e-05,
807
+ "loss": 0.4869,
808
+ "step": 570
809
+ },
810
+ {
811
+ "epoch": 0.7941988950276243,
812
+ "grad_norm": 1.074428915977478,
813
+ "learning_rate": 2.6393841645385302e-05,
814
+ "loss": 0.506,
815
+ "step": 575
816
+ },
817
+ {
818
+ "epoch": 0.8011049723756906,
819
+ "grad_norm": 1.3407135009765625,
820
+ "learning_rate": 2.6319228244776632e-05,
821
+ "loss": 0.4437,
822
+ "step": 580
823
+ },
824
+ {
825
+ "epoch": 0.8080110497237569,
826
+ "grad_norm": 1.1925331354141235,
827
+ "learning_rate": 2.624395861426372e-05,
828
+ "loss": 0.4637,
829
+ "step": 585
830
+ },
831
+ {
832
+ "epoch": 0.8149171270718232,
833
+ "grad_norm": 1.1516950130462646,
834
+ "learning_rate": 2.6168037117587745e-05,
835
+ "loss": 0.431,
836
+ "step": 590
837
+ },
838
+ {
839
+ "epoch": 0.8218232044198895,
840
+ "grad_norm": 1.217863917350769,
841
+ "learning_rate": 2.6091468156281707e-05,
842
+ "loss": 0.446,
843
+ "step": 595
844
+ },
845
+ {
846
+ "epoch": 0.8287292817679558,
847
+ "grad_norm": 1.1536725759506226,
848
+ "learning_rate": 2.60142561694152e-05,
849
+ "loss": 0.4393,
850
+ "step": 600
851
+ },
852
+ {
853
+ "epoch": 0.835635359116022,
854
+ "grad_norm": 1.159977912902832,
855
+ "learning_rate": 2.5936405633337128e-05,
856
+ "loss": 0.4561,
857
+ "step": 605
858
+ },
859
+ {
860
+ "epoch": 0.8425414364640884,
861
+ "grad_norm": 1.1989026069641113,
862
+ "learning_rate": 2.5857921061416122e-05,
863
+ "loss": 0.44,
864
+ "step": 610
865
+ },
866
+ {
867
+ "epoch": 0.8494475138121547,
868
+ "grad_norm": 1.294961929321289,
869
+ "learning_rate": 2.577880700377893e-05,
870
+ "loss": 0.4455,
871
+ "step": 615
872
+ },
873
+ {
874
+ "epoch": 0.856353591160221,
875
+ "grad_norm": 1.1323007345199585,
876
+ "learning_rate": 2.5699068047046594e-05,
877
+ "loss": 0.4266,
878
+ "step": 620
879
+ },
880
+ {
881
+ "epoch": 0.8632596685082873,
882
+ "grad_norm": 1.2258801460266113,
883
+ "learning_rate": 2.5618708814068547e-05,
884
+ "loss": 0.4205,
885
+ "step": 625
886
+ },
887
+ {
888
+ "epoch": 0.8701657458563536,
889
+ "grad_norm": 1.1914846897125244,
890
+ "learning_rate": 2.553773396365462e-05,
891
+ "loss": 0.4585,
892
+ "step": 630
893
+ },
894
+ {
895
+ "epoch": 0.8770718232044199,
896
+ "grad_norm": 1.1909990310668945,
897
+ "learning_rate": 2.5456148190304935e-05,
898
+ "loss": 0.4201,
899
+ "step": 635
900
+ },
901
+ {
902
+ "epoch": 0.8839779005524862,
903
+ "grad_norm": 1.2360750436782837,
904
+ "learning_rate": 2.537395622393773e-05,
905
+ "loss": 0.4812,
906
+ "step": 640
907
+ },
908
+ {
909
+ "epoch": 0.8908839779005525,
910
+ "grad_norm": 1.2781263589859009,
911
+ "learning_rate": 2.5291162829615177e-05,
912
+ "loss": 0.4057,
913
+ "step": 645
914
+ },
915
+ {
916
+ "epoch": 0.8977900552486188,
917
+ "grad_norm": 1.1696977615356445,
918
+ "learning_rate": 2.5207772807267094e-05,
919
+ "loss": 0.422,
920
+ "step": 650
921
+ },
922
+ {
923
+ "epoch": 0.9046961325966851,
924
+ "grad_norm": 1.157449722290039,
925
+ "learning_rate": 2.51237909914127e-05,
926
+ "loss": 0.4071,
927
+ "step": 655
928
+ },
929
+ {
930
+ "epoch": 0.9116022099447514,
931
+ "grad_norm": 1.2058300971984863,
932
+ "learning_rate": 2.5039222250880286e-05,
933
+ "loss": 0.3967,
934
+ "step": 660
935
+ },
936
+ {
937
+ "epoch": 0.9185082872928176,
938
+ "grad_norm": 1.1892244815826416,
939
+ "learning_rate": 2.495407148852503e-05,
940
+ "loss": 0.3988,
941
+ "step": 665
942
+ },
943
+ {
944
+ "epoch": 0.925414364640884,
945
+ "grad_norm": 1.0561643838882446,
946
+ "learning_rate": 2.486834364094466e-05,
947
+ "loss": 0.4343,
948
+ "step": 670
949
+ },
950
+ {
951
+ "epoch": 0.9323204419889503,
952
+ "grad_norm": 1.2995855808258057,
953
+ "learning_rate": 2.4782043678193326e-05,
954
+ "loss": 0.3922,
955
+ "step": 675
956
+ },
957
+ {
958
+ "epoch": 0.9392265193370166,
959
+ "grad_norm": 1.197883129119873,
960
+ "learning_rate": 2.4695176603493416e-05,
961
+ "loss": 0.3679,
962
+ "step": 680
963
+ },
964
+ {
965
+ "epoch": 0.9461325966850829,
966
+ "grad_norm": 1.1414422988891602,
967
+ "learning_rate": 2.4607747452945532e-05,
968
+ "loss": 0.3956,
969
+ "step": 685
970
+ },
971
+ {
972
+ "epoch": 0.9530386740331491,
973
+ "grad_norm": 1.080524206161499,
974
+ "learning_rate": 2.4519761295236504e-05,
975
+ "loss": 0.4072,
976
+ "step": 690
977
+ },
978
+ {
979
+ "epoch": 0.9599447513812155,
980
+ "grad_norm": 1.1876850128173828,
981
+ "learning_rate": 2.443122323134553e-05,
982
+ "loss": 0.3705,
983
+ "step": 695
984
+ },
985
+ {
986
+ "epoch": 0.9668508287292817,
987
+ "grad_norm": 1.2733322381973267,
988
+ "learning_rate": 2.4342138394248456e-05,
989
+ "loss": 0.3987,
990
+ "step": 700
991
+ },
992
+ {
993
+ "epoch": 0.9737569060773481,
994
+ "grad_norm": 1.4294981956481934,
995
+ "learning_rate": 2.425251194862019e-05,
996
+ "loss": 0.3966,
997
+ "step": 705
998
+ },
999
+ {
1000
+ "epoch": 0.9806629834254144,
1001
+ "grad_norm": 1.0828477144241333,
1002
+ "learning_rate": 2.4162349090535294e-05,
1003
+ "loss": 0.3323,
1004
+ "step": 710
1005
+ },
1006
+ {
1007
+ "epoch": 0.9875690607734806,
1008
+ "grad_norm": 1.1359148025512695,
1009
+ "learning_rate": 2.407165504716671e-05,
1010
+ "loss": 0.3689,
1011
+ "step": 715
1012
+ },
1013
+ {
1014
+ "epoch": 0.994475138121547,
1015
+ "grad_norm": 1.3517268896102905,
1016
+ "learning_rate": 2.3980435076482753e-05,
1017
+ "loss": 0.3453,
1018
+ "step": 720
1019
+ },
1020
+ {
1021
+ "epoch": 1.0013812154696133,
1022
+ "grad_norm": 1.1152492761611938,
1023
+ "learning_rate": 2.3888694466942254e-05,
1024
+ "loss": 0.3435,
1025
+ "step": 725
1026
+ },
1027
+ {
1028
+ "epoch": 1.0082872928176796,
1029
+ "grad_norm": 1.3809410333633423,
1030
+ "learning_rate": 2.3796438537187985e-05,
1031
+ "loss": 0.2837,
1032
+ "step": 730
1033
+ },
1034
+ {
1035
+ "epoch": 1.0151933701657458,
1036
+ "grad_norm": 1.061593770980835,
1037
+ "learning_rate": 2.3703672635738286e-05,
1038
+ "loss": 0.3307,
1039
+ "step": 735
1040
+ },
1041
+ {
1042
+ "epoch": 1.022099447513812,
1043
+ "grad_norm": 1.2451817989349365,
1044
+ "learning_rate": 2.3610402140677016e-05,
1045
+ "loss": 0.2913,
1046
+ "step": 740
1047
+ },
1048
+ {
1049
+ "epoch": 1.0290055248618784,
1050
+ "grad_norm": 1.108031153678894,
1051
+ "learning_rate": 2.3516632459341727e-05,
1052
+ "loss": 0.3268,
1053
+ "step": 745
1054
+ },
1055
+ {
1056
+ "epoch": 1.0359116022099448,
1057
+ "grad_norm": 1.3022626638412476,
1058
+ "learning_rate": 2.342236902801021e-05,
1059
+ "loss": 0.2776,
1060
+ "step": 750
1061
+ },
1062
+ {
1063
+ "epoch": 1.042817679558011,
1064
+ "grad_norm": 1.308383584022522,
1065
+ "learning_rate": 2.3327617311585284e-05,
1066
+ "loss": 0.2995,
1067
+ "step": 755
1068
+ },
1069
+ {
1070
+ "epoch": 1.0497237569060773,
1071
+ "grad_norm": 1.1599491834640503,
1072
+ "learning_rate": 2.3232382803278023e-05,
1073
+ "loss": 0.2807,
1074
+ "step": 760
1075
+ },
1076
+ {
1077
+ "epoch": 1.0566298342541436,
1078
+ "grad_norm": 1.1750526428222656,
1079
+ "learning_rate": 2.3136671024289245e-05,
1080
+ "loss": 0.2898,
1081
+ "step": 765
1082
+ },
1083
+ {
1084
+ "epoch": 1.06353591160221,
1085
+ "grad_norm": 1.1713992357254028,
1086
+ "learning_rate": 2.3040487523489436e-05,
1087
+ "loss": 0.2651,
1088
+ "step": 770
1089
+ },
1090
+ {
1091
+ "epoch": 1.0704419889502763,
1092
+ "grad_norm": 1.3281424045562744,
1093
+ "learning_rate": 2.2943837877097057e-05,
1094
+ "loss": 0.2746,
1095
+ "step": 775
1096
+ },
1097
+ {
1098
+ "epoch": 1.0773480662983426,
1099
+ "grad_norm": 1.2191420793533325,
1100
+ "learning_rate": 2.284672768835527e-05,
1101
+ "loss": 0.2975,
1102
+ "step": 780
1103
+ },
1104
+ {
1105
+ "epoch": 1.0842541436464088,
1106
+ "grad_norm": 1.143297553062439,
1107
+ "learning_rate": 2.274916258720707e-05,
1108
+ "loss": 0.2548,
1109
+ "step": 785
1110
+ },
1111
+ {
1112
+ "epoch": 1.091160220994475,
1113
+ "grad_norm": 1.0258440971374512,
1114
+ "learning_rate": 2.2651148229968915e-05,
1115
+ "loss": 0.3112,
1116
+ "step": 790
1117
+ },
1118
+ {
1119
+ "epoch": 1.0980662983425415,
1120
+ "grad_norm": 1.2485140562057495,
1121
+ "learning_rate": 2.25526902990028e-05,
1122
+ "loss": 0.2673,
1123
+ "step": 795
1124
+ },
1125
+ {
1126
+ "epoch": 1.1049723756906078,
1127
+ "grad_norm": 1.3313682079315186,
1128
+ "learning_rate": 2.245379450238679e-05,
1129
+ "loss": 0.2507,
1130
+ "step": 800
1131
+ },
1132
+ {
1133
+ "epoch": 1.111878453038674,
1134
+ "grad_norm": 1.054181456565857,
1135
+ "learning_rate": 2.2354466573584133e-05,
1136
+ "loss": 0.2849,
1137
+ "step": 805
1138
+ },
1139
+ {
1140
+ "epoch": 1.1187845303867403,
1141
+ "grad_norm": 1.1910122632980347,
1142
+ "learning_rate": 2.2254712271110864e-05,
1143
+ "loss": 0.3027,
1144
+ "step": 810
1145
+ },
1146
+ {
1147
+ "epoch": 1.1256906077348066,
1148
+ "grad_norm": 1.2852493524551392,
1149
+ "learning_rate": 2.2154537378201916e-05,
1150
+ "loss": 0.3069,
1151
+ "step": 815
1152
+ },
1153
+ {
1154
+ "epoch": 1.132596685082873,
1155
+ "grad_norm": 1.2807539701461792,
1156
+ "learning_rate": 2.205394770247589e-05,
1157
+ "loss": 0.2659,
1158
+ "step": 820
1159
+ },
1160
+ {
1161
+ "epoch": 1.1395027624309393,
1162
+ "grad_norm": 1.1665374040603638,
1163
+ "learning_rate": 2.1952949075598316e-05,
1164
+ "loss": 0.2731,
1165
+ "step": 825
1166
+ },
1167
+ {
1168
+ "epoch": 1.1464088397790055,
1169
+ "grad_norm": 1.146697759628296,
1170
+ "learning_rate": 2.185154735294359e-05,
1171
+ "loss": 0.2725,
1172
+ "step": 830
1173
+ },
1174
+ {
1175
+ "epoch": 1.1533149171270718,
1176
+ "grad_norm": 1.172950029373169,
1177
+ "learning_rate": 2.1749748413255507e-05,
1178
+ "loss": 0.2701,
1179
+ "step": 835
1180
+ },
1181
+ {
1182
+ "epoch": 1.160220994475138,
1183
+ "grad_norm": 1.1988505125045776,
1184
+ "learning_rate": 2.1647558158306427e-05,
1185
+ "loss": 0.2668,
1186
+ "step": 840
1187
+ },
1188
+ {
1189
+ "epoch": 1.1671270718232045,
1190
+ "grad_norm": 1.1398203372955322,
1191
+ "learning_rate": 2.1544982512555142e-05,
1192
+ "loss": 0.2821,
1193
+ "step": 845
1194
+ },
1195
+ {
1196
+ "epoch": 1.1740331491712708,
1197
+ "grad_norm": 1.280297875404358,
1198
+ "learning_rate": 2.144202742280338e-05,
1199
+ "loss": 0.2874,
1200
+ "step": 850
1201
+ },
1202
+ {
1203
+ "epoch": 1.180939226519337,
1204
+ "grad_norm": 1.0032289028167725,
1205
+ "learning_rate": 2.1338698857851073e-05,
1206
+ "loss": 0.216,
1207
+ "step": 855
1208
+ },
1209
+ {
1210
+ "epoch": 1.1878453038674033,
1211
+ "grad_norm": 1.215258240699768,
1212
+ "learning_rate": 2.1235002808150287e-05,
1213
+ "loss": 0.2316,
1214
+ "step": 860
1215
+ },
1216
+ {
1217
+ "epoch": 1.1947513812154695,
1218
+ "grad_norm": 1.1415657997131348,
1219
+ "learning_rate": 2.1130945285457948e-05,
1220
+ "loss": 0.2411,
1221
+ "step": 865
1222
+ },
1223
+ {
1224
+ "epoch": 1.201657458563536,
1225
+ "grad_norm": 1.1622593402862549,
1226
+ "learning_rate": 2.10265323224873e-05,
1227
+ "loss": 0.254,
1228
+ "step": 870
1229
+ },
1230
+ {
1231
+ "epoch": 1.2085635359116023,
1232
+ "grad_norm": 1.4223426580429077,
1233
+ "learning_rate": 2.0921769972558163e-05,
1234
+ "loss": 0.2296,
1235
+ "step": 875
1236
+ },
1237
+ {
1238
+ "epoch": 1.2154696132596685,
1239
+ "grad_norm": 1.1282471418380737,
1240
+ "learning_rate": 2.0816664309245995e-05,
1241
+ "loss": 0.2661,
1242
+ "step": 880
1243
+ },
1244
+ {
1245
+ "epoch": 1.2223756906077348,
1246
+ "grad_norm": 1.2149296998977661,
1247
+ "learning_rate": 2.071122142602977e-05,
1248
+ "loss": 0.2129,
1249
+ "step": 885
1250
+ },
1251
+ {
1252
+ "epoch": 1.229281767955801,
1253
+ "grad_norm": 1.0837780237197876,
1254
+ "learning_rate": 2.0605447435938733e-05,
1255
+ "loss": 0.277,
1256
+ "step": 890
1257
+ },
1258
+ {
1259
+ "epoch": 1.2361878453038675,
1260
+ "grad_norm": 1.1397072076797485,
1261
+ "learning_rate": 2.0499348471197957e-05,
1262
+ "loss": 0.232,
1263
+ "step": 895
1264
+ },
1265
+ {
1266
+ "epoch": 1.2430939226519337,
1267
+ "grad_norm": 1.0269798040390015,
1268
+ "learning_rate": 2.0392930682872878e-05,
1269
+ "loss": 0.2389,
1270
+ "step": 900
1271
+ },
1272
+ {
1273
+ "epoch": 1.25,
1274
+ "grad_norm": 1.221785306930542,
1275
+ "learning_rate": 2.0286200240512636e-05,
1276
+ "loss": 0.2227,
1277
+ "step": 905
1278
+ },
1279
+ {
1280
+ "epoch": 1.2569060773480663,
1281
+ "grad_norm": 1.2008177042007446,
1282
+ "learning_rate": 2.017916333179245e-05,
1283
+ "loss": 0.2759,
1284
+ "step": 910
1285
+ },
1286
+ {
1287
+ "epoch": 1.2638121546961325,
1288
+ "grad_norm": 1.2568888664245605,
1289
+ "learning_rate": 2.0071826162154836e-05,
1290
+ "loss": 0.2315,
1291
+ "step": 915
1292
+ },
1293
+ {
1294
+ "epoch": 1.270718232044199,
1295
+ "grad_norm": 1.2279225587844849,
1296
+ "learning_rate": 1.9964194954449896e-05,
1297
+ "loss": 0.2126,
1298
+ "step": 920
1299
+ },
1300
+ {
1301
+ "epoch": 1.2776243093922652,
1302
+ "grad_norm": 1.1100877523422241,
1303
+ "learning_rate": 1.985627594857453e-05,
1304
+ "loss": 0.2345,
1305
+ "step": 925
1306
+ },
1307
+ {
1308
+ "epoch": 1.2845303867403315,
1309
+ "grad_norm": 1.225351333618164,
1310
+ "learning_rate": 1.9748075401110653e-05,
1311
+ "loss": 0.2687,
1312
+ "step": 930
1313
+ },
1314
+ {
1315
+ "epoch": 1.2914364640883977,
1316
+ "grad_norm": 1.134172797203064,
1317
+ "learning_rate": 1.9639599584962517e-05,
1318
+ "loss": 0.231,
1319
+ "step": 935
1320
+ },
1321
+ {
1322
+ "epoch": 1.298342541436464,
1323
+ "grad_norm": 1.1137300729751587,
1324
+ "learning_rate": 1.9530854788993003e-05,
1325
+ "loss": 0.2305,
1326
+ "step": 940
1327
+ },
1328
+ {
1329
+ "epoch": 1.3052486187845305,
1330
+ "grad_norm": 1.0298051834106445,
1331
+ "learning_rate": 1.9421847317659062e-05,
1332
+ "loss": 0.2216,
1333
+ "step": 945
1334
+ },
1335
+ {
1336
+ "epoch": 1.3121546961325967,
1337
+ "grad_norm": 1.0633655786514282,
1338
+ "learning_rate": 1.9312583490646175e-05,
1339
+ "loss": 0.2239,
1340
+ "step": 950
1341
+ },
1342
+ {
1343
+ "epoch": 1.319060773480663,
1344
+ "grad_norm": 1.0870096683502197,
1345
+ "learning_rate": 1.9203069642501996e-05,
1346
+ "loss": 0.224,
1347
+ "step": 955
1348
+ },
1349
+ {
1350
+ "epoch": 1.3259668508287292,
1351
+ "grad_norm": 1.2183704376220703,
1352
+ "learning_rate": 1.9093312122269106e-05,
1353
+ "loss": 0.2323,
1354
+ "step": 960
1355
+ },
1356
+ {
1357
+ "epoch": 1.3328729281767955,
1358
+ "grad_norm": 1.1188147068023682,
1359
+ "learning_rate": 1.8983317293116922e-05,
1360
+ "loss": 0.224,
1361
+ "step": 965
1362
+ },
1363
+ {
1364
+ "epoch": 1.339779005524862,
1365
+ "grad_norm": 1.2784143686294556,
1366
+ "learning_rate": 1.88730915319728e-05,
1367
+ "loss": 0.2616,
1368
+ "step": 970
1369
+ },
1370
+ {
1371
+ "epoch": 1.3466850828729282,
1372
+ "grad_norm": 1.2629567384719849,
1373
+ "learning_rate": 1.8762641229152317e-05,
1374
+ "loss": 0.2071,
1375
+ "step": 975
1376
+ },
1377
+ {
1378
+ "epoch": 1.3535911602209945,
1379
+ "grad_norm": 1.4207789897918701,
1380
+ "learning_rate": 1.8651972787988822e-05,
1381
+ "loss": 0.2079,
1382
+ "step": 980
1383
+ },
1384
+ {
1385
+ "epoch": 1.3604972375690607,
1386
+ "grad_norm": 1.2269219160079956,
1387
+ "learning_rate": 1.8541092624462185e-05,
1388
+ "loss": 0.2216,
1389
+ "step": 985
1390
+ },
1391
+ {
1392
+ "epoch": 1.367403314917127,
1393
+ "grad_norm": 1.1409672498703003,
1394
+ "learning_rate": 1.8430007166826833e-05,
1395
+ "loss": 0.1588,
1396
+ "step": 990
1397
+ },
1398
+ {
1399
+ "epoch": 1.3743093922651934,
1400
+ "grad_norm": 1.315989375114441,
1401
+ "learning_rate": 1.8318722855239075e-05,
1402
+ "loss": 0.1898,
1403
+ "step": 995
1404
+ },
1405
+ {
1406
+ "epoch": 1.3812154696132597,
1407
+ "grad_norm": 0.9964866638183594,
1408
+ "learning_rate": 1.8207246141383744e-05,
1409
+ "loss": 0.1823,
1410
+ "step": 1000
1411
+ },
1412
+ {
1413
+ "epoch": 1.388121546961326,
1414
+ "grad_norm": 1.0690737962722778,
1415
+ "learning_rate": 1.809558348810015e-05,
1416
+ "loss": 0.2222,
1417
+ "step": 1005
1418
+ },
1419
+ {
1420
+ "epoch": 1.3950276243093922,
1421
+ "grad_norm": 1.1185473203659058,
1422
+ "learning_rate": 1.7983741369007404e-05,
1423
+ "loss": 0.2034,
1424
+ "step": 1010
1425
+ },
1426
+ {
1427
+ "epoch": 1.4019337016574585,
1428
+ "grad_norm": 1.1302683353424072,
1429
+ "learning_rate": 1.78717262681291e-05,
1430
+ "loss": 0.2019,
1431
+ "step": 1015
1432
+ },
1433
+ {
1434
+ "epoch": 1.408839779005525,
1435
+ "grad_norm": 1.0584911108016968,
1436
+ "learning_rate": 1.775954467951743e-05,
1437
+ "loss": 0.1984,
1438
+ "step": 1020
1439
+ },
1440
+ {
1441
+ "epoch": 1.4157458563535912,
1442
+ "grad_norm": 1.137606143951416,
1443
+ "learning_rate": 1.7647203106876673e-05,
1444
+ "loss": 0.2038,
1445
+ "step": 1025
1446
+ },
1447
+ {
1448
+ "epoch": 1.4226519337016574,
1449
+ "grad_norm": 1.0054763555526733,
1450
+ "learning_rate": 1.7534708063186146e-05,
1451
+ "loss": 0.1836,
1452
+ "step": 1030
1453
+ },
1454
+ {
1455
+ "epoch": 1.4295580110497237,
1456
+ "grad_norm": 1.0451409816741943,
1457
+ "learning_rate": 1.7422066070322614e-05,
1458
+ "loss": 0.1936,
1459
+ "step": 1035
1460
+ },
1461
+ {
1462
+ "epoch": 1.43646408839779,
1463
+ "grad_norm": 1.2669540643692017,
1464
+ "learning_rate": 1.73092836586822e-05,
1465
+ "loss": 0.1706,
1466
+ "step": 1040
1467
+ },
1468
+ {
1469
+ "epoch": 1.4433701657458564,
1470
+ "grad_norm": 1.1040784120559692,
1471
+ "learning_rate": 1.719636736680177e-05,
1472
+ "loss": 0.1816,
1473
+ "step": 1045
1474
+ },
1475
+ {
1476
+ "epoch": 1.4502762430939227,
1477
+ "grad_norm": 1.049950122833252,
1478
+ "learning_rate": 1.7083323740979875e-05,
1479
+ "loss": 0.1813,
1480
+ "step": 1050
1481
+ },
1482
+ {
1483
+ "epoch": 1.457182320441989,
1484
+ "grad_norm": 0.8987106084823608,
1485
+ "learning_rate": 1.697015933489721e-05,
1486
+ "loss": 0.2,
1487
+ "step": 1055
1488
+ },
1489
+ {
1490
+ "epoch": 1.4640883977900552,
1491
+ "grad_norm": 1.0366822481155396,
1492
+ "learning_rate": 1.6856880709236706e-05,
1493
+ "loss": 0.1913,
1494
+ "step": 1060
1495
+ },
1496
+ {
1497
+ "epoch": 1.4709944751381214,
1498
+ "grad_norm": 1.1850148439407349,
1499
+ "learning_rate": 1.6743494431303122e-05,
1500
+ "loss": 0.1724,
1501
+ "step": 1065
1502
+ },
1503
+ {
1504
+ "epoch": 1.477900552486188,
1505
+ "grad_norm": 1.1060631275177002,
1506
+ "learning_rate": 1.6630007074642353e-05,
1507
+ "loss": 0.2095,
1508
+ "step": 1070
1509
+ },
1510
+ {
1511
+ "epoch": 1.4848066298342542,
1512
+ "grad_norm": 1.00900399684906,
1513
+ "learning_rate": 1.651642521866031e-05,
1514
+ "loss": 0.1779,
1515
+ "step": 1075
1516
+ },
1517
+ {
1518
+ "epoch": 1.4917127071823204,
1519
+ "grad_norm": 1.204003095626831,
1520
+ "learning_rate": 1.6402755448241476e-05,
1521
+ "loss": 0.1706,
1522
+ "step": 1080
1523
+ },
1524
+ {
1525
+ "epoch": 1.4986187845303867,
1526
+ "grad_norm": 1.1007627248764038,
1527
+ "learning_rate": 1.6289004353367164e-05,
1528
+ "loss": 0.1807,
1529
+ "step": 1085
1530
+ },
1531
+ {
1532
+ "epoch": 1.505524861878453,
1533
+ "grad_norm": 1.0451252460479736,
1534
+ "learning_rate": 1.6175178528733443e-05,
1535
+ "loss": 0.1822,
1536
+ "step": 1090
1537
+ },
1538
+ {
1539
+ "epoch": 1.5124309392265194,
1540
+ "grad_norm": 1.0837717056274414,
1541
+ "learning_rate": 1.6061284573368844e-05,
1542
+ "loss": 0.1892,
1543
+ "step": 1095
1544
+ },
1545
+ {
1546
+ "epoch": 1.5193370165745856,
1547
+ "grad_norm": 1.2764201164245605,
1548
+ "learning_rate": 1.5947329090251738e-05,
1549
+ "loss": 0.1978,
1550
+ "step": 1100
1551
+ },
1552
+ {
1553
+ "epoch": 1.526243093922652,
1554
+ "grad_norm": 1.0126081705093384,
1555
+ "learning_rate": 1.583331868592757e-05,
1556
+ "loss": 0.1533,
1557
+ "step": 1105
1558
+ },
1559
+ {
1560
+ "epoch": 1.5331491712707184,
1561
+ "grad_norm": 1.2582359313964844,
1562
+ "learning_rate": 1.5719259970125837e-05,
1563
+ "loss": 0.1693,
1564
+ "step": 1110
1565
+ },
1566
+ {
1567
+ "epoch": 1.5400552486187844,
1568
+ "grad_norm": 0.9745519161224365,
1569
+ "learning_rate": 1.5605159555376866e-05,
1570
+ "loss": 0.1766,
1571
+ "step": 1115
1572
+ },
1573
+ {
1574
+ "epoch": 1.5469613259668509,
1575
+ "grad_norm": 1.211788296699524,
1576
+ "learning_rate": 1.5491024056628493e-05,
1577
+ "loss": 0.167,
1578
+ "step": 1120
1579
+ },
1580
+ {
1581
+ "epoch": 1.5538674033149171,
1582
+ "grad_norm": 1.0196974277496338,
1583
+ "learning_rate": 1.5376860090862507e-05,
1584
+ "loss": 0.1787,
1585
+ "step": 1125
1586
+ },
1587
+ {
1588
+ "epoch": 1.5607734806629834,
1589
+ "grad_norm": 1.2439481019973755,
1590
+ "learning_rate": 1.526267427671111e-05,
1591
+ "loss": 0.171,
1592
+ "step": 1130
1593
+ },
1594
+ {
1595
+ "epoch": 1.5676795580110499,
1596
+ "grad_norm": 1.1908317804336548,
1597
+ "learning_rate": 1.5148473234073129e-05,
1598
+ "loss": 0.1688,
1599
+ "step": 1135
1600
+ },
1601
+ {
1602
+ "epoch": 1.5745856353591159,
1603
+ "grad_norm": 1.1063194274902344,
1604
+ "learning_rate": 1.5034263583730272e-05,
1605
+ "loss": 0.1612,
1606
+ "step": 1140
1607
+ },
1608
+ {
1609
+ "epoch": 1.5814917127071824,
1610
+ "grad_norm": 1.1663639545440674,
1611
+ "learning_rate": 1.4920051946963264e-05,
1612
+ "loss": 0.1778,
1613
+ "step": 1145
1614
+ },
1615
+ {
1616
+ "epoch": 1.5883977900552486,
1617
+ "grad_norm": 0.9558233618736267,
1618
+ "learning_rate": 1.4805844945168013e-05,
1619
+ "loss": 0.1445,
1620
+ "step": 1150
1621
+ },
1622
+ {
1623
+ "epoch": 1.5953038674033149,
1624
+ "grad_norm": 1.1863877773284912,
1625
+ "learning_rate": 1.4691649199471705e-05,
1626
+ "loss": 0.1803,
1627
+ "step": 1155
1628
+ },
1629
+ {
1630
+ "epoch": 1.6022099447513813,
1631
+ "grad_norm": 1.0897351503372192,
1632
+ "learning_rate": 1.4577471330348951e-05,
1633
+ "loss": 0.1756,
1634
+ "step": 1160
1635
+ },
1636
+ {
1637
+ "epoch": 1.6091160220994474,
1638
+ "grad_norm": 1.0444782972335815,
1639
+ "learning_rate": 1.4463317957237977e-05,
1640
+ "loss": 0.1498,
1641
+ "step": 1165
1642
+ },
1643
+ {
1644
+ "epoch": 1.6160220994475138,
1645
+ "grad_norm": 1.215334415435791,
1646
+ "learning_rate": 1.4349195698156855e-05,
1647
+ "loss": 0.1573,
1648
+ "step": 1170
1649
+ },
1650
+ {
1651
+ "epoch": 1.62292817679558,
1652
+ "grad_norm": 0.9446357488632202,
1653
+ "learning_rate": 1.4235111169319827e-05,
1654
+ "loss": 0.1457,
1655
+ "step": 1175
1656
+ },
1657
+ {
1658
+ "epoch": 1.6298342541436464,
1659
+ "grad_norm": 1.0680475234985352,
1660
+ "learning_rate": 1.4121070984753742e-05,
1661
+ "loss": 0.1502,
1662
+ "step": 1180
1663
+ },
1664
+ {
1665
+ "epoch": 1.6367403314917128,
1666
+ "grad_norm": 1.0287423133850098,
1667
+ "learning_rate": 1.4007081755914589e-05,
1668
+ "loss": 0.1395,
1669
+ "step": 1185
1670
+ },
1671
+ {
1672
+ "epoch": 1.6436464088397789,
1673
+ "grad_norm": 1.0201785564422607,
1674
+ "learning_rate": 1.3893150091304212e-05,
1675
+ "loss": 0.145,
1676
+ "step": 1190
1677
+ },
1678
+ {
1679
+ "epoch": 1.6505524861878453,
1680
+ "grad_norm": 1.0086338520050049,
1681
+ "learning_rate": 1.3779282596087174e-05,
1682
+ "loss": 0.1612,
1683
+ "step": 1195
1684
+ },
1685
+ {
1686
+ "epoch": 1.6574585635359116,
1687
+ "grad_norm": 1.0178710222244263,
1688
+ "learning_rate": 1.3665485871707848e-05,
1689
+ "loss": 0.1214,
1690
+ "step": 1200
1691
+ },
1692
+ {
1693
+ "epoch": 1.6643646408839778,
1694
+ "grad_norm": 1.079098105430603,
1695
+ "learning_rate": 1.3551766515507663e-05,
1696
+ "loss": 0.1496,
1697
+ "step": 1205
1698
+ },
1699
+ {
1700
+ "epoch": 1.6712707182320443,
1701
+ "grad_norm": 1.4140187501907349,
1702
+ "learning_rate": 1.3438131120342652e-05,
1703
+ "loss": 0.1515,
1704
+ "step": 1210
1705
+ },
1706
+ {
1707
+ "epoch": 1.6781767955801103,
1708
+ "grad_norm": 1.0219125747680664,
1709
+ "learning_rate": 1.3324586274201232e-05,
1710
+ "loss": 0.1373,
1711
+ "step": 1215
1712
+ },
1713
+ {
1714
+ "epoch": 1.6850828729281768,
1715
+ "grad_norm": 1.0244609117507935,
1716
+ "learning_rate": 1.3211138559822254e-05,
1717
+ "loss": 0.1411,
1718
+ "step": 1220
1719
+ },
1720
+ {
1721
+ "epoch": 1.691988950276243,
1722
+ "grad_norm": 0.8596904873847961,
1723
+ "learning_rate": 1.3097794554313364e-05,
1724
+ "loss": 0.1474,
1725
+ "step": 1225
1726
+ },
1727
+ {
1728
+ "epoch": 1.6988950276243093,
1729
+ "grad_norm": 0.9590954780578613,
1730
+ "learning_rate": 1.298456082876972e-05,
1731
+ "loss": 0.1211,
1732
+ "step": 1230
1733
+ },
1734
+ {
1735
+ "epoch": 1.7058011049723758,
1736
+ "grad_norm": 1.010059118270874,
1737
+ "learning_rate": 1.2871443947893026e-05,
1738
+ "loss": 0.1544,
1739
+ "step": 1235
1740
+ },
1741
+ {
1742
+ "epoch": 1.7127071823204418,
1743
+ "grad_norm": 1.023232340812683,
1744
+ "learning_rate": 1.2758450469610926e-05,
1745
+ "loss": 0.1491,
1746
+ "step": 1240
1747
+ },
1748
+ {
1749
+ "epoch": 1.7196132596685083,
1750
+ "grad_norm": 0.8564914464950562,
1751
+ "learning_rate": 1.2645586944696833e-05,
1752
+ "loss": 0.1353,
1753
+ "step": 1245
1754
+ },
1755
+ {
1756
+ "epoch": 1.7265193370165746,
1757
+ "grad_norm": 1.1296415328979492,
1758
+ "learning_rate": 1.2532859916390142e-05,
1759
+ "loss": 0.1346,
1760
+ "step": 1250
1761
+ },
1762
+ {
1763
+ "epoch": 1.7334254143646408,
1764
+ "grad_norm": 0.9086131453514099,
1765
+ "learning_rate": 1.2420275920016888e-05,
1766
+ "loss": 0.1327,
1767
+ "step": 1255
1768
+ },
1769
+ {
1770
+ "epoch": 1.7403314917127073,
1771
+ "grad_norm": 0.9245094060897827,
1772
+ "learning_rate": 1.2307841482610847e-05,
1773
+ "loss": 0.1343,
1774
+ "step": 1260
1775
+ },
1776
+ {
1777
+ "epoch": 1.7472375690607733,
1778
+ "grad_norm": 1.055877685546875,
1779
+ "learning_rate": 1.2195563122535155e-05,
1780
+ "loss": 0.1322,
1781
+ "step": 1265
1782
+ },
1783
+ {
1784
+ "epoch": 1.7541436464088398,
1785
+ "grad_norm": 0.9924134016036987,
1786
+ "learning_rate": 1.2083447349104398e-05,
1787
+ "loss": 0.1183,
1788
+ "step": 1270
1789
+ },
1790
+ {
1791
+ "epoch": 1.761049723756906,
1792
+ "grad_norm": 1.0010970830917358,
1793
+ "learning_rate": 1.1971500662207228e-05,
1794
+ "loss": 0.1406,
1795
+ "step": 1275
1796
+ },
1797
+ {
1798
+ "epoch": 1.7679558011049723,
1799
+ "grad_norm": 0.9735561013221741,
1800
+ "learning_rate": 1.1859729551929549e-05,
1801
+ "loss": 0.1305,
1802
+ "step": 1280
1803
+ },
1804
+ {
1805
+ "epoch": 1.7748618784530388,
1806
+ "grad_norm": 0.9438261389732361,
1807
+ "learning_rate": 1.1748140498178228e-05,
1808
+ "loss": 0.1339,
1809
+ "step": 1285
1810
+ },
1811
+ {
1812
+ "epoch": 1.7817679558011048,
1813
+ "grad_norm": 1.0416855812072754,
1814
+ "learning_rate": 1.1636739970305477e-05,
1815
+ "loss": 0.1241,
1816
+ "step": 1290
1817
+ },
1818
+ {
1819
+ "epoch": 1.7886740331491713,
1820
+ "grad_norm": 0.8868617415428162,
1821
+ "learning_rate": 1.152553442673373e-05,
1822
+ "loss": 0.1116,
1823
+ "step": 1295
1824
+ },
1825
+ {
1826
+ "epoch": 1.7955801104972375,
1827
+ "grad_norm": 1.1853326559066772,
1828
+ "learning_rate": 1.1414530314581258e-05,
1829
+ "loss": 0.1468,
1830
+ "step": 1300
1831
+ },
1832
+ {
1833
+ "epoch": 1.8024861878453038,
1834
+ "grad_norm": 0.9709151983261108,
1835
+ "learning_rate": 1.1303734069288379e-05,
1836
+ "loss": 0.1246,
1837
+ "step": 1305
1838
+ },
1839
+ {
1840
+ "epoch": 1.8093922651933703,
1841
+ "grad_norm": 0.8692839741706848,
1842
+ "learning_rate": 1.1193152114244387e-05,
1843
+ "loss": 0.1355,
1844
+ "step": 1310
1845
+ },
1846
+ {
1847
+ "epoch": 1.8162983425414365,
1848
+ "grad_norm": 1.0272483825683594,
1849
+ "learning_rate": 1.1082790860415135e-05,
1850
+ "loss": 0.132,
1851
+ "step": 1315
1852
+ },
1853
+ {
1854
+ "epoch": 1.8232044198895028,
1855
+ "grad_norm": 1.0036672353744507,
1856
+ "learning_rate": 1.0972656705971367e-05,
1857
+ "loss": 0.1272,
1858
+ "step": 1320
1859
+ },
1860
+ {
1861
+ "epoch": 1.830110497237569,
1862
+ "grad_norm": 1.0183753967285156,
1863
+ "learning_rate": 1.0862756035917787e-05,
1864
+ "loss": 0.11,
1865
+ "step": 1325
1866
+ },
1867
+ {
1868
+ "epoch": 1.8370165745856353,
1869
+ "grad_norm": 0.9497151970863342,
1870
+ "learning_rate": 1.0753095221722899e-05,
1871
+ "loss": 0.1246,
1872
+ "step": 1330
1873
+ },
1874
+ {
1875
+ "epoch": 1.8439226519337018,
1876
+ "grad_norm": 1.0131932497024536,
1877
+ "learning_rate": 1.0643680620949611e-05,
1878
+ "loss": 0.124,
1879
+ "step": 1335
1880
+ },
1881
+ {
1882
+ "epoch": 1.850828729281768,
1883
+ "grad_norm": 0.9499644041061401,
1884
+ "learning_rate": 1.0534518576886645e-05,
1885
+ "loss": 0.1179,
1886
+ "step": 1340
1887
+ },
1888
+ {
1889
+ "epoch": 1.8577348066298343,
1890
+ "grad_norm": 0.8952851891517639,
1891
+ "learning_rate": 1.0425615418180818e-05,
1892
+ "loss": 0.1071,
1893
+ "step": 1345
1894
+ },
1895
+ {
1896
+ "epoch": 1.8646408839779005,
1897
+ "grad_norm": 0.891074001789093,
1898
+ "learning_rate": 1.0316977458470117e-05,
1899
+ "loss": 0.1326,
1900
+ "step": 1350
1901
+ },
1902
+ {
1903
+ "epoch": 1.8715469613259668,
1904
+ "grad_norm": 1.0223931074142456,
1905
+ "learning_rate": 1.0208610996017665e-05,
1906
+ "loss": 0.1186,
1907
+ "step": 1355
1908
+ },
1909
+ {
1910
+ "epoch": 1.8784530386740332,
1911
+ "grad_norm": 0.9757610559463501,
1912
+ "learning_rate": 1.010052231334659e-05,
1913
+ "loss": 0.1193,
1914
+ "step": 1360
1915
+ },
1916
+ {
1917
+ "epoch": 1.8853591160220995,
1918
+ "grad_norm": 0.8758223652839661,
1919
+ "learning_rate": 9.99271767687581e-06,
1920
+ "loss": 0.1226,
1921
+ "step": 1365
1922
+ },
1923
+ {
1924
+ "epoch": 1.8922651933701657,
1925
+ "grad_norm": 0.8537967801094055,
1926
+ "learning_rate": 9.885203336556708e-06,
1927
+ "loss": 0.1167,
1928
+ "step": 1370
1929
+ },
1930
+ {
1931
+ "epoch": 1.899171270718232,
1932
+ "grad_norm": 1.0821012258529663,
1933
+ "learning_rate": 9.777985525510807e-06,
1934
+ "loss": 0.1282,
1935
+ "step": 1375
1936
+ },
1937
+ {
1938
+ "epoch": 1.9060773480662982,
1939
+ "grad_norm": 0.7928299307823181,
1940
+ "learning_rate": 9.671070459668413e-06,
1941
+ "loss": 0.1123,
1942
+ "step": 1380
1943
+ },
1944
+ {
1945
+ "epoch": 1.9129834254143647,
1946
+ "grad_norm": 0.850980281829834,
1947
+ "learning_rate": 9.564464337408249e-06,
1948
+ "loss": 0.1224,
1949
+ "step": 1385
1950
+ },
1951
+ {
1952
+ "epoch": 1.919889502762431,
1953
+ "grad_norm": 1.0882021188735962,
1954
+ "learning_rate": 9.458173339198079e-06,
1955
+ "loss": 0.1135,
1956
+ "step": 1390
1957
+ },
1958
+ {
1959
+ "epoch": 1.9267955801104972,
1960
+ "grad_norm": 0.8873913884162903,
1961
+ "learning_rate": 9.352203627236427e-06,
1962
+ "loss": 0.1044,
1963
+ "step": 1395
1964
+ },
1965
+ {
1966
+ "epoch": 1.9337016574585635,
1967
+ "grad_norm": 0.9549221396446228,
1968
+ "learning_rate": 9.246561345095321e-06,
1969
+ "loss": 0.1043,
1970
+ "step": 1400
1971
+ },
1972
+ {
1973
+ "epoch": 1.9406077348066297,
1974
+ "grad_norm": 0.9950575232505798,
1975
+ "learning_rate": 9.141252617364103e-06,
1976
+ "loss": 0.1081,
1977
+ "step": 1405
1978
+ },
1979
+ {
1980
+ "epoch": 1.9475138121546962,
1981
+ "grad_norm": 0.8856183290481567,
1982
+ "learning_rate": 9.036283549294365e-06,
1983
+ "loss": 0.1098,
1984
+ "step": 1410
1985
+ },
1986
+ {
1987
+ "epoch": 1.9544198895027625,
1988
+ "grad_norm": 0.9136974811553955,
1989
+ "learning_rate": 8.931660226446008e-06,
1990
+ "loss": 0.1196,
1991
+ "step": 1415
1992
+ },
1993
+ {
1994
+ "epoch": 1.9613259668508287,
1995
+ "grad_norm": 0.8464009761810303,
1996
+ "learning_rate": 8.827388714334428e-06,
1997
+ "loss": 0.1033,
1998
+ "step": 1420
1999
+ },
2000
+ {
2001
+ "epoch": 1.9682320441988952,
2002
+ "grad_norm": 0.8959425687789917,
2003
+ "learning_rate": 8.723475058078856e-06,
2004
+ "loss": 0.114,
2005
+ "step": 1425
2006
+ },
2007
+ {
2008
+ "epoch": 1.9751381215469612,
2009
+ "grad_norm": 1.1590272188186646,
2010
+ "learning_rate": 8.619925282051906e-06,
2011
+ "loss": 0.1062,
2012
+ "step": 1430
2013
+ },
2014
+ {
2015
+ "epoch": 1.9820441988950277,
2016
+ "grad_norm": 0.8596269488334656,
2017
+ "learning_rate": 8.516745389530318e-06,
2018
+ "loss": 0.1037,
2019
+ "step": 1435
2020
+ },
2021
+ {
2022
+ "epoch": 1.988950276243094,
2023
+ "grad_norm": 0.8075308799743652,
2024
+ "learning_rate": 8.41394136234692e-06,
2025
+ "loss": 0.113,
2026
+ "step": 1440
2027
+ },
2028
+ {
2029
+ "epoch": 1.9958563535911602,
2030
+ "grad_norm": 0.7233743071556091,
2031
+ "learning_rate": 8.311519160543806e-06,
2032
+ "loss": 0.0987,
2033
+ "step": 1445
2034
+ },
2035
+ {
2036
+ "epoch": 2.0027624309392267,
2037
+ "grad_norm": 0.6330487728118896,
2038
+ "learning_rate": 8.209484722026827e-06,
2039
+ "loss": 0.0945,
2040
+ "step": 1450
2041
+ },
2042
+ {
2043
+ "epoch": 2.0096685082872927,
2044
+ "grad_norm": 0.7824728488922119,
2045
+ "learning_rate": 8.107843962221359e-06,
2046
+ "loss": 0.0875,
2047
+ "step": 1455
2048
+ },
2049
+ {
2050
+ "epoch": 2.016574585635359,
2051
+ "grad_norm": 0.9059628248214722,
2052
+ "learning_rate": 8.006602773729308e-06,
2053
+ "loss": 0.0893,
2054
+ "step": 1460
2055
+ },
2056
+ {
2057
+ "epoch": 2.023480662983425,
2058
+ "grad_norm": 0.7491239309310913,
2059
+ "learning_rate": 7.905767025987528e-06,
2060
+ "loss": 0.082,
2061
+ "step": 1465
2062
+ },
2063
+ {
2064
+ "epoch": 2.0303867403314917,
2065
+ "grad_norm": 0.7340039610862732,
2066
+ "learning_rate": 7.805342564927534e-06,
2067
+ "loss": 0.0781,
2068
+ "step": 1470
2069
+ },
2070
+ {
2071
+ "epoch": 2.037292817679558,
2072
+ "grad_norm": 0.8720238208770752,
2073
+ "learning_rate": 7.705335212636586e-06,
2074
+ "loss": 0.0887,
2075
+ "step": 1475
2076
+ },
2077
+ {
2078
+ "epoch": 2.044198895027624,
2079
+ "grad_norm": 0.7781177163124084,
2080
+ "learning_rate": 7.605750767020134e-06,
2081
+ "loss": 0.0791,
2082
+ "step": 1480
2083
+ },
2084
+ {
2085
+ "epoch": 2.0511049723756907,
2086
+ "grad_norm": 0.7472957968711853,
2087
+ "learning_rate": 7.506595001465704e-06,
2088
+ "loss": 0.0884,
2089
+ "step": 1485
2090
+ },
2091
+ {
2092
+ "epoch": 2.0580110497237567,
2093
+ "grad_norm": 0.9449889659881592,
2094
+ "learning_rate": 7.407873664508202e-06,
2095
+ "loss": 0.0859,
2096
+ "step": 1490
2097
+ },
2098
+ {
2099
+ "epoch": 2.064917127071823,
2100
+ "grad_norm": 0.7017861008644104,
2101
+ "learning_rate": 7.309592479496607e-06,
2102
+ "loss": 0.0876,
2103
+ "step": 1495
2104
+ },
2105
+ {
2106
+ "epoch": 2.0718232044198897,
2107
+ "grad_norm": 0.7882440686225891,
2108
+ "learning_rate": 7.211757144262182e-06,
2109
+ "loss": 0.0808,
2110
+ "step": 1500
2111
+ },
2112
+ {
2113
+ "epoch": 2.0787292817679557,
2114
+ "grad_norm": 0.910998523235321,
2115
+ "learning_rate": 7.1143733307881595e-06,
2116
+ "loss": 0.0884,
2117
+ "step": 1505
2118
+ },
2119
+ {
2120
+ "epoch": 2.085635359116022,
2121
+ "grad_norm": 0.6744552850723267,
2122
+ "learning_rate": 7.017446684880869e-06,
2123
+ "loss": 0.0774,
2124
+ "step": 1510
2125
+ },
2126
+ {
2127
+ "epoch": 2.092541436464088,
2128
+ "grad_norm": 0.7734429836273193,
2129
+ "learning_rate": 6.920982825842464e-06,
2130
+ "loss": 0.0802,
2131
+ "step": 1515
2132
+ },
2133
+ {
2134
+ "epoch": 2.0994475138121547,
2135
+ "grad_norm": 0.8410816192626953,
2136
+ "learning_rate": 6.824987346145111e-06,
2137
+ "loss": 0.0839,
2138
+ "step": 1520
2139
+ },
2140
+ {
2141
+ "epoch": 2.106353591160221,
2142
+ "grad_norm": 0.6995574235916138,
2143
+ "learning_rate": 6.729465811106795e-06,
2144
+ "loss": 0.0925,
2145
+ "step": 1525
2146
+ },
2147
+ {
2148
+ "epoch": 2.113259668508287,
2149
+ "grad_norm": 0.6755858063697815,
2150
+ "learning_rate": 6.634423758568649e-06,
2151
+ "loss": 0.0721,
2152
+ "step": 1530
2153
+ },
2154
+ {
2155
+ "epoch": 2.1201657458563536,
2156
+ "grad_norm": 0.6981886029243469,
2157
+ "learning_rate": 6.539866698573902e-06,
2158
+ "loss": 0.0833,
2159
+ "step": 1535
2160
+ },
2161
+ {
2162
+ "epoch": 2.12707182320442,
2163
+ "grad_norm": 0.767754077911377,
2164
+ "learning_rate": 6.445800113048459e-06,
2165
+ "loss": 0.0702,
2166
+ "step": 1540
2167
+ },
2168
+ {
2169
+ "epoch": 2.133977900552486,
2170
+ "grad_norm": 0.7646163702011108,
2171
+ "learning_rate": 6.3522294554830525e-06,
2172
+ "loss": 0.0872,
2173
+ "step": 1545
2174
+ },
2175
+ {
2176
+ "epoch": 2.1408839779005526,
2177
+ "grad_norm": 0.7515497803688049,
2178
+ "learning_rate": 6.259160150617108e-06,
2179
+ "loss": 0.0754,
2180
+ "step": 1550
2181
+ },
2182
+ {
2183
+ "epoch": 2.1477900552486187,
2184
+ "grad_norm": 0.7609333992004395,
2185
+ "learning_rate": 6.166597594124219e-06,
2186
+ "loss": 0.0855,
2187
+ "step": 1555
2188
+ },
2189
+ {
2190
+ "epoch": 2.154696132596685,
2191
+ "grad_norm": 0.8277834057807922,
2192
+ "learning_rate": 6.074547152299364e-06,
2193
+ "loss": 0.0802,
2194
+ "step": 1560
2195
+ },
2196
+ {
2197
+ "epoch": 2.1616022099447516,
2198
+ "grad_norm": 0.8085483312606812,
2199
+ "learning_rate": 5.983014161747765e-06,
2200
+ "loss": 0.0825,
2201
+ "step": 1565
2202
+ },
2203
+ {
2204
+ "epoch": 2.1685082872928176,
2205
+ "grad_norm": 0.9735024571418762,
2206
+ "learning_rate": 5.892003929075514e-06,
2207
+ "loss": 0.0861,
2208
+ "step": 1570
2209
+ },
2210
+ {
2211
+ "epoch": 2.175414364640884,
2212
+ "grad_norm": 0.837413489818573,
2213
+ "learning_rate": 5.80152173058194e-06,
2214
+ "loss": 0.0813,
2215
+ "step": 1575
2216
+ },
2217
+ {
2218
+ "epoch": 2.18232044198895,
2219
+ "grad_norm": 0.7000280618667603,
2220
+ "learning_rate": 5.711572811953686e-06,
2221
+ "loss": 0.0728,
2222
+ "step": 1580
2223
+ },
2224
+ {
2225
+ "epoch": 2.1892265193370166,
2226
+ "grad_norm": 0.558982253074646,
2227
+ "learning_rate": 5.622162387960601e-06,
2228
+ "loss": 0.0677,
2229
+ "step": 1585
2230
+ },
2231
+ {
2232
+ "epoch": 2.196132596685083,
2233
+ "grad_norm": 0.639984667301178,
2234
+ "learning_rate": 5.533295642153433e-06,
2235
+ "loss": 0.0728,
2236
+ "step": 1590
2237
+ },
2238
+ {
2239
+ "epoch": 2.203038674033149,
2240
+ "grad_norm": 0.650260865688324,
2241
+ "learning_rate": 5.4449777265633034e-06,
2242
+ "loss": 0.0752,
2243
+ "step": 1595
2244
+ },
2245
+ {
2246
+ "epoch": 2.2099447513812156,
2247
+ "grad_norm": 0.6584851741790771,
2248
+ "learning_rate": 5.357213761403002e-06,
2249
+ "loss": 0.0816,
2250
+ "step": 1600
2251
+ },
2252
+ {
2253
+ "epoch": 2.2168508287292816,
2254
+ "grad_norm": 0.7054372429847717,
2255
+ "learning_rate": 5.270008834770153e-06,
2256
+ "loss": 0.0727,
2257
+ "step": 1605
2258
+ },
2259
+ {
2260
+ "epoch": 2.223756906077348,
2261
+ "grad_norm": 0.6314348578453064,
2262
+ "learning_rate": 5.183368002352263e-06,
2263
+ "loss": 0.0826,
2264
+ "step": 1610
2265
+ },
2266
+ {
2267
+ "epoch": 2.2306629834254146,
2268
+ "grad_norm": 0.6414453387260437,
2269
+ "learning_rate": 5.097296287133573e-06,
2270
+ "loss": 0.0693,
2271
+ "step": 1615
2272
+ },
2273
+ {
2274
+ "epoch": 2.2375690607734806,
2275
+ "grad_norm": 0.6908891797065735,
2276
+ "learning_rate": 5.011798679103871e-06,
2277
+ "loss": 0.0856,
2278
+ "step": 1620
2279
+ },
2280
+ {
2281
+ "epoch": 2.244475138121547,
2282
+ "grad_norm": 0.5416004061698914,
2283
+ "learning_rate": 4.926880134969216e-06,
2284
+ "loss": 0.0688,
2285
+ "step": 1625
2286
+ },
2287
+ {
2288
+ "epoch": 2.251381215469613,
2289
+ "grad_norm": 0.6058601140975952,
2290
+ "learning_rate": 4.842545577864552e-06,
2291
+ "loss": 0.0828,
2292
+ "step": 1630
2293
+ },
2294
+ {
2295
+ "epoch": 2.2582872928176796,
2296
+ "grad_norm": 0.9427646398544312,
2297
+ "learning_rate": 4.758799897068293e-06,
2298
+ "loss": 0.0844,
2299
+ "step": 1635
2300
+ },
2301
+ {
2302
+ "epoch": 2.265193370165746,
2303
+ "grad_norm": 0.765407383441925,
2304
+ "learning_rate": 4.675647947718868e-06,
2305
+ "loss": 0.0807,
2306
+ "step": 1640
2307
+ },
2308
+ {
2309
+ "epoch": 2.272099447513812,
2310
+ "grad_norm": 0.5569551587104797,
2311
+ "learning_rate": 4.593094550533251e-06,
2312
+ "loss": 0.069,
2313
+ "step": 1645
2314
+ },
2315
+ {
2316
+ "epoch": 2.2790055248618786,
2317
+ "grad_norm": 0.7765437960624695,
2318
+ "learning_rate": 4.511144491527491e-06,
2319
+ "loss": 0.0674,
2320
+ "step": 1650
2321
+ },
2322
+ {
2323
+ "epoch": 2.2859116022099446,
2324
+ "grad_norm": 0.6243528127670288,
2325
+ "learning_rate": 4.429802521739215e-06,
2326
+ "loss": 0.0737,
2327
+ "step": 1655
2328
+ },
2329
+ {
2330
+ "epoch": 2.292817679558011,
2331
+ "grad_norm": 0.8506344556808472,
2332
+ "learning_rate": 4.349073356952203e-06,
2333
+ "loss": 0.0734,
2334
+ "step": 1660
2335
+ },
2336
+ {
2337
+ "epoch": 2.2997237569060776,
2338
+ "grad_norm": 0.6179936528205872,
2339
+ "learning_rate": 4.2689616774230175e-06,
2340
+ "loss": 0.0757,
2341
+ "step": 1665
2342
+ },
2343
+ {
2344
+ "epoch": 2.3066298342541436,
2345
+ "grad_norm": 0.9602283835411072,
2346
+ "learning_rate": 4.189472127609616e-06,
2347
+ "loss": 0.0719,
2348
+ "step": 1670
2349
+ },
2350
+ {
2351
+ "epoch": 2.31353591160221,
2352
+ "grad_norm": 0.6802037954330444,
2353
+ "learning_rate": 4.110609315902121e-06,
2354
+ "loss": 0.0758,
2355
+ "step": 1675
2356
+ },
2357
+ {
2358
+ "epoch": 2.320441988950276,
2359
+ "grad_norm": 0.5111964344978333,
2360
+ "learning_rate": 4.032377814355644e-06,
2361
+ "loss": 0.0717,
2362
+ "step": 1680
2363
+ },
2364
+ {
2365
+ "epoch": 2.3273480662983426,
2366
+ "grad_norm": 0.5520031452178955,
2367
+ "learning_rate": 3.9547821584252305e-06,
2368
+ "loss": 0.0636,
2369
+ "step": 1685
2370
+ },
2371
+ {
2372
+ "epoch": 2.334254143646409,
2373
+ "grad_norm": 0.6408013105392456,
2374
+ "learning_rate": 3.877826846702897e-06,
2375
+ "loss": 0.0738,
2376
+ "step": 1690
2377
+ },
2378
+ {
2379
+ "epoch": 2.341160220994475,
2380
+ "grad_norm": 0.6354919672012329,
2381
+ "learning_rate": 3.8015163406568354e-06,
2382
+ "loss": 0.0717,
2383
+ "step": 1695
2384
+ },
2385
+ {
2386
+ "epoch": 2.3480662983425415,
2387
+ "grad_norm": 0.44999581575393677,
2388
+ "learning_rate": 3.725855064372776e-06,
2389
+ "loss": 0.0625,
2390
+ "step": 1700
2391
+ },
2392
+ {
2393
+ "epoch": 2.3549723756906076,
2394
+ "grad_norm": 0.7536104917526245,
2395
+ "learning_rate": 3.6508474042974854e-06,
2396
+ "loss": 0.0761,
2397
+ "step": 1705
2398
+ },
2399
+ {
2400
+ "epoch": 2.361878453038674,
2401
+ "grad_norm": 0.8556480407714844,
2402
+ "learning_rate": 3.576497708984457e-06,
2403
+ "loss": 0.0767,
2404
+ "step": 1710
2405
+ },
2406
+ {
2407
+ "epoch": 2.3687845303867405,
2408
+ "grad_norm": 0.6432438492774963,
2409
+ "learning_rate": 3.5028102888418146e-06,
2410
+ "loss": 0.0724,
2411
+ "step": 1715
2412
+ },
2413
+ {
2414
+ "epoch": 2.3756906077348066,
2415
+ "grad_norm": 0.657943606376648,
2416
+ "learning_rate": 3.4297894158824317e-06,
2417
+ "loss": 0.0656,
2418
+ "step": 1720
2419
+ },
2420
+ {
2421
+ "epoch": 2.382596685082873,
2422
+ "grad_norm": 0.7321274280548096,
2423
+ "learning_rate": 3.3574393234762314e-06,
2424
+ "loss": 0.067,
2425
+ "step": 1725
2426
+ },
2427
+ {
2428
+ "epoch": 2.389502762430939,
2429
+ "grad_norm": 0.5283787846565247,
2430
+ "learning_rate": 3.285764206104777e-06,
2431
+ "loss": 0.0687,
2432
+ "step": 1730
2433
+ },
2434
+ {
2435
+ "epoch": 2.3964088397790055,
2436
+ "grad_norm": 0.5984429717063904,
2437
+ "learning_rate": 3.2147682191181e-06,
2438
+ "loss": 0.0695,
2439
+ "step": 1735
2440
+ },
2441
+ {
2442
+ "epoch": 2.403314917127072,
2443
+ "grad_norm": 0.7319976687431335,
2444
+ "learning_rate": 3.14445547849379e-06,
2445
+ "loss": 0.0801,
2446
+ "step": 1740
2447
+ },
2448
+ {
2449
+ "epoch": 2.410220994475138,
2450
+ "grad_norm": 0.663022518157959,
2451
+ "learning_rate": 3.074830060598362e-06,
2452
+ "loss": 0.0607,
2453
+ "step": 1745
2454
+ },
2455
+ {
2456
+ "epoch": 2.4171270718232045,
2457
+ "grad_norm": 0.653674840927124,
2458
+ "learning_rate": 3.005896001950941e-06,
2459
+ "loss": 0.0737,
2460
+ "step": 1750
2461
+ },
2462
+ {
2463
+ "epoch": 2.4240331491712706,
2464
+ "grad_norm": 0.6952985525131226,
2465
+ "learning_rate": 2.9376572989892538e-06,
2466
+ "loss": 0.067,
2467
+ "step": 1755
2468
+ },
2469
+ {
2470
+ "epoch": 2.430939226519337,
2471
+ "grad_norm": 0.7008129954338074,
2472
+ "learning_rate": 2.8701179078379135e-06,
2473
+ "loss": 0.0788,
2474
+ "step": 1760
2475
+ },
2476
+ {
2477
+ "epoch": 2.4378453038674035,
2478
+ "grad_norm": 0.8141315579414368,
2479
+ "learning_rate": 2.8032817440790766e-06,
2480
+ "loss": 0.0611,
2481
+ "step": 1765
2482
+ },
2483
+ {
2484
+ "epoch": 2.4447513812154695,
2485
+ "grad_norm": 0.6668171882629395,
2486
+ "learning_rate": 2.7371526825254445e-06,
2487
+ "loss": 0.0646,
2488
+ "step": 1770
2489
+ },
2490
+ {
2491
+ "epoch": 2.451657458563536,
2492
+ "grad_norm": 0.6932879090309143,
2493
+ "learning_rate": 2.671734556995619e-06,
2494
+ "loss": 0.0714,
2495
+ "step": 1775
2496
+ },
2497
+ {
2498
+ "epoch": 2.458563535911602,
2499
+ "grad_norm": 0.5816868543624878,
2500
+ "learning_rate": 2.607031160091821e-06,
2501
+ "loss": 0.0775,
2502
+ "step": 1780
2503
+ },
2504
+ {
2505
+ "epoch": 2.4654696132596685,
2506
+ "grad_norm": 0.6177311539649963,
2507
+ "learning_rate": 2.543046242980032e-06,
2508
+ "loss": 0.0738,
2509
+ "step": 1785
2510
+ },
2511
+ {
2512
+ "epoch": 2.472375690607735,
2513
+ "grad_norm": 0.5263634920120239,
2514
+ "learning_rate": 2.4797835151725275e-06,
2515
+ "loss": 0.0688,
2516
+ "step": 1790
2517
+ },
2518
+ {
2519
+ "epoch": 2.479281767955801,
2520
+ "grad_norm": 0.8297523260116577,
2521
+ "learning_rate": 2.4172466443127937e-06,
2522
+ "loss": 0.0666,
2523
+ "step": 1795
2524
+ },
2525
+ {
2526
+ "epoch": 2.4861878453038675,
2527
+ "grad_norm": 0.5330537557601929,
2528
+ "learning_rate": 2.3554392559629152e-06,
2529
+ "loss": 0.0724,
2530
+ "step": 1800
2531
+ },
2532
+ {
2533
+ "epoch": 2.4930939226519335,
2534
+ "grad_norm": 0.7814313173294067,
2535
+ "learning_rate": 2.2943649333933864e-06,
2536
+ "loss": 0.0591,
2537
+ "step": 1805
2538
+ },
2539
+ {
2540
+ "epoch": 2.5,
2541
+ "grad_norm": 0.5160076022148132,
2542
+ "learning_rate": 2.234027217375364e-06,
2543
+ "loss": 0.0654,
2544
+ "step": 1810
2545
+ },
2546
+ {
2547
+ "epoch": 2.5069060773480665,
2548
+ "grad_norm": 0.6160582900047302,
2549
+ "learning_rate": 2.1744296059753843e-06,
2550
+ "loss": 0.076,
2551
+ "step": 1815
2552
+ },
2553
+ {
2554
+ "epoch": 2.5138121546961325,
2555
+ "grad_norm": 0.604662299156189,
2556
+ "learning_rate": 2.115575554352578e-06,
2557
+ "loss": 0.0695,
2558
+ "step": 1820
2559
+ },
2560
+ {
2561
+ "epoch": 2.520718232044199,
2562
+ "grad_norm": 0.5971505641937256,
2563
+ "learning_rate": 2.057468474558356e-06,
2564
+ "loss": 0.0684,
2565
+ "step": 1825
2566
+ },
2567
+ {
2568
+ "epoch": 2.527624309392265,
2569
+ "grad_norm": 0.6619613170623779,
2570
+ "learning_rate": 2.000111735338588e-06,
2571
+ "loss": 0.0667,
2572
+ "step": 1830
2573
+ },
2574
+ {
2575
+ "epoch": 2.5345303867403315,
2576
+ "grad_norm": 0.6380955576896667,
2577
+ "learning_rate": 1.943508661938301e-06,
2578
+ "loss": 0.0724,
2579
+ "step": 1835
2580
+ },
2581
+ {
2582
+ "epoch": 2.541436464088398,
2583
+ "grad_norm": 0.6917901635169983,
2584
+ "learning_rate": 1.8876625359089139e-06,
2585
+ "loss": 0.0723,
2586
+ "step": 1840
2587
+ },
2588
+ {
2589
+ "epoch": 2.548342541436464,
2590
+ "grad_norm": 0.6325092315673828,
2591
+ "learning_rate": 1.8325765949179697e-06,
2592
+ "loss": 0.0707,
2593
+ "step": 1845
2594
+ },
2595
+ {
2596
+ "epoch": 2.5552486187845305,
2597
+ "grad_norm": 0.6682673692703247,
2598
+ "learning_rate": 1.7782540325614472e-06,
2599
+ "loss": 0.0669,
2600
+ "step": 1850
2601
+ },
2602
+ {
2603
+ "epoch": 2.5621546961325965,
2604
+ "grad_norm": 0.5539576411247253,
2605
+ "learning_rate": 1.724697998178601e-06,
2606
+ "loss": 0.0694,
2607
+ "step": 1855
2608
+ },
2609
+ {
2610
+ "epoch": 2.569060773480663,
2611
+ "grad_norm": 0.5859360694885254,
2612
+ "learning_rate": 1.6719115966693965e-06,
2613
+ "loss": 0.0677,
2614
+ "step": 1860
2615
+ },
2616
+ {
2617
+ "epoch": 2.5759668508287294,
2618
+ "grad_norm": 0.6111728549003601,
2619
+ "learning_rate": 1.6198978883144844e-06,
2620
+ "loss": 0.0725,
2621
+ "step": 1865
2622
+ },
2623
+ {
2624
+ "epoch": 2.5828729281767955,
2625
+ "grad_norm": 0.5038866996765137,
2626
+ "learning_rate": 1.5686598885977904e-06,
2627
+ "loss": 0.0605,
2628
+ "step": 1870
2629
+ },
2630
+ {
2631
+ "epoch": 2.589779005524862,
2632
+ "grad_norm": 0.6684066653251648,
2633
+ "learning_rate": 1.518200568031699e-06,
2634
+ "loss": 0.0729,
2635
+ "step": 1875
2636
+ },
2637
+ {
2638
+ "epoch": 2.596685082872928,
2639
+ "grad_norm": 0.6127700805664062,
2640
+ "learning_rate": 1.468522851984826e-06,
2641
+ "loss": 0.0729,
2642
+ "step": 1880
2643
+ },
2644
+ {
2645
+ "epoch": 2.6035911602209945,
2646
+ "grad_norm": 0.41123858094215393,
2647
+ "learning_rate": 1.4196296205124372e-06,
2648
+ "loss": 0.0673,
2649
+ "step": 1885
2650
+ },
2651
+ {
2652
+ "epoch": 2.610497237569061,
2653
+ "grad_norm": 0.5337295532226562,
2654
+ "learning_rate": 1.3715237081894556e-06,
2655
+ "loss": 0.0668,
2656
+ "step": 1890
2657
+ },
2658
+ {
2659
+ "epoch": 2.617403314917127,
2660
+ "grad_norm": 0.5636597871780396,
2661
+ "learning_rate": 1.3242079039461529e-06,
2662
+ "loss": 0.0595,
2663
+ "step": 1895
2664
+ },
2665
+ {
2666
+ "epoch": 2.6243093922651934,
2667
+ "grad_norm": 0.5281133651733398,
2668
+ "learning_rate": 1.277684950906441e-06,
2669
+ "loss": 0.0592,
2670
+ "step": 1900
2671
+ },
2672
+ {
2673
+ "epoch": 2.6312154696132595,
2674
+ "grad_norm": 0.67793869972229,
2675
+ "learning_rate": 1.231957546228849e-06,
2676
+ "loss": 0.0723,
2677
+ "step": 1905
2678
+ },
2679
+ {
2680
+ "epoch": 2.638121546961326,
2681
+ "grad_norm": 0.7178230285644531,
2682
+ "learning_rate": 1.187028340950156e-06,
2683
+ "loss": 0.0646,
2684
+ "step": 1910
2685
+ },
2686
+ {
2687
+ "epoch": 2.6450276243093924,
2688
+ "grad_norm": 0.6277831792831421,
2689
+ "learning_rate": 1.1428999398316997e-06,
2690
+ "loss": 0.067,
2691
+ "step": 1915
2692
+ },
2693
+ {
2694
+ "epoch": 2.6519337016574585,
2695
+ "grad_norm": 0.6713959574699402,
2696
+ "learning_rate": 1.0995749012083578e-06,
2697
+ "loss": 0.0685,
2698
+ "step": 1920
2699
+ },
2700
+ {
2701
+ "epoch": 2.658839779005525,
2702
+ "grad_norm": 0.6218628287315369,
2703
+ "learning_rate": 1.0570557368402384e-06,
2704
+ "loss": 0.0617,
2705
+ "step": 1925
2706
+ },
2707
+ {
2708
+ "epoch": 2.665745856353591,
2709
+ "grad_norm": 0.518380880355835,
2710
+ "learning_rate": 1.0153449117670606e-06,
2711
+ "loss": 0.0692,
2712
+ "step": 1930
2713
+ },
2714
+ {
2715
+ "epoch": 2.6726519337016574,
2716
+ "grad_norm": 0.6599457263946533,
2717
+ "learning_rate": 9.744448441652315e-07,
2718
+ "loss": 0.0642,
2719
+ "step": 1935
2720
+ },
2721
+ {
2722
+ "epoch": 2.679558011049724,
2723
+ "grad_norm": 0.5130138993263245,
2724
+ "learning_rate": 9.343579052076678e-07,
2725
+ "loss": 0.0663,
2726
+ "step": 1940
2727
+ },
2728
+ {
2729
+ "epoch": 2.68646408839779,
2730
+ "grad_norm": 0.5128707885742188,
2731
+ "learning_rate": 8.950864189263214e-07,
2732
+ "loss": 0.0646,
2733
+ "step": 1945
2734
+ },
2735
+ {
2736
+ "epoch": 2.6933701657458564,
2737
+ "grad_norm": 0.6543782949447632,
2738
+ "learning_rate": 8.566326620774473e-07,
2739
+ "loss": 0.0621,
2740
+ "step": 1950
2741
+ },
2742
+ {
2743
+ "epoch": 2.7002762430939224,
2744
+ "grad_norm": 0.4967004656791687,
2745
+ "learning_rate": 8.189988640095997e-07,
2746
+ "loss": 0.0696,
2747
+ "step": 1955
2748
+ },
2749
+ {
2750
+ "epoch": 2.707182320441989,
2751
+ "grad_norm": 0.5444068312644958,
2752
+ "learning_rate": 7.821872065344032e-07,
2753
+ "loss": 0.0623,
2754
+ "step": 1960
2755
+ },
2756
+ {
2757
+ "epoch": 2.7140883977900554,
2758
+ "grad_norm": 0.6234949827194214,
2759
+ "learning_rate": 7.461998238000412e-07,
2760
+ "loss": 0.0654,
2761
+ "step": 1965
2762
+ },
2763
+ {
2764
+ "epoch": 2.7209944751381214,
2765
+ "grad_norm": 0.6768404245376587,
2766
+ "learning_rate": 7.110388021675523e-07,
2767
+ "loss": 0.066,
2768
+ "step": 1970
2769
+ },
2770
+ {
2771
+ "epoch": 2.727900552486188,
2772
+ "grad_norm": 0.572343647480011,
2773
+ "learning_rate": 6.767061800898517e-07,
2774
+ "loss": 0.0649,
2775
+ "step": 1975
2776
+ },
2777
+ {
2778
+ "epoch": 2.734806629834254,
2779
+ "grad_norm": 0.5168081521987915,
2780
+ "learning_rate": 6.432039479935642e-07,
2781
+ "loss": 0.0651,
2782
+ "step": 1980
2783
+ },
2784
+ {
2785
+ "epoch": 2.7417127071823204,
2786
+ "grad_norm": 0.660479724407196,
2787
+ "learning_rate": 6.105340481636363e-07,
2788
+ "loss": 0.069,
2789
+ "step": 1985
2790
+ },
2791
+ {
2792
+ "epoch": 2.748618784530387,
2793
+ "grad_norm": 0.5443263649940491,
2794
+ "learning_rate": 5.786983746307167e-07,
2795
+ "loss": 0.0596,
2796
+ "step": 1990
2797
+ },
2798
+ {
2799
+ "epoch": 2.755524861878453,
2800
+ "grad_norm": 0.6075274348258972,
2801
+ "learning_rate": 5.47698773061357e-07,
2802
+ "loss": 0.065,
2803
+ "step": 1995
2804
+ },
2805
+ {
2806
+ "epoch": 2.7624309392265194,
2807
+ "grad_norm": 0.5168625712394714,
2808
+ "learning_rate": 5.17537040651016e-07,
2809
+ "loss": 0.0585,
2810
+ "step": 2000
2811
+ },
2812
+ {
2813
+ "epoch": 2.7693370165745854,
2814
+ "grad_norm": 0.6499465107917786,
2815
+ "learning_rate": 4.882149260198593e-07,
2816
+ "loss": 0.0637,
2817
+ "step": 2005
2818
+ },
2819
+ {
2820
+ "epoch": 2.776243093922652,
2821
+ "grad_norm": 0.6499086022377014,
2822
+ "learning_rate": 4.597341291113877e-07,
2823
+ "loss": 0.0655,
2824
+ "step": 2010
2825
+ },
2826
+ {
2827
+ "epoch": 2.7831491712707184,
2828
+ "grad_norm": 0.4925879240036011,
2829
+ "learning_rate": 4.3209630109387834e-07,
2830
+ "loss": 0.0542,
2831
+ "step": 2015
2832
+ },
2833
+ {
2834
+ "epoch": 2.7900552486187844,
2835
+ "grad_norm": 0.6256985068321228,
2836
+ "learning_rate": 4.05303044264666e-07,
2837
+ "loss": 0.0636,
2838
+ "step": 2020
2839
+ },
2840
+ {
2841
+ "epoch": 2.796961325966851,
2842
+ "grad_norm": 0.5748464465141296,
2843
+ "learning_rate": 3.7935591195724593e-07,
2844
+ "loss": 0.0649,
2845
+ "step": 2025
2846
+ },
2847
+ {
2848
+ "epoch": 2.803867403314917,
2849
+ "grad_norm": 0.5975679159164429,
2850
+ "learning_rate": 3.5425640845121566e-07,
2851
+ "loss": 0.0581,
2852
+ "step": 2030
2853
+ },
2854
+ {
2855
+ "epoch": 2.8107734806629834,
2856
+ "grad_norm": 0.4849628210067749,
2857
+ "learning_rate": 3.300059888850765e-07,
2858
+ "loss": 0.0593,
2859
+ "step": 2035
2860
+ },
2861
+ {
2862
+ "epoch": 2.81767955801105,
2863
+ "grad_norm": 0.5387313961982727,
2864
+ "learning_rate": 3.066060591718628e-07,
2865
+ "loss": 0.0634,
2866
+ "step": 2040
2867
+ },
2868
+ {
2869
+ "epoch": 2.824585635359116,
2870
+ "grad_norm": 0.5569940209388733,
2871
+ "learning_rate": 2.8405797591763037e-07,
2872
+ "loss": 0.0654,
2873
+ "step": 2045
2874
+ },
2875
+ {
2876
+ "epoch": 2.8314917127071824,
2877
+ "grad_norm": 0.5501409769058228,
2878
+ "learning_rate": 2.623630463428195e-07,
2879
+ "loss": 0.0638,
2880
+ "step": 2050
2881
+ },
2882
+ {
2883
+ "epoch": 2.8383977900552484,
2884
+ "grad_norm": 0.5223251581192017,
2885
+ "learning_rate": 2.415225282064637e-07,
2886
+ "loss": 0.0722,
2887
+ "step": 2055
2888
+ },
2889
+ {
2890
+ "epoch": 2.845303867403315,
2891
+ "grad_norm": 0.5327146053314209,
2892
+ "learning_rate": 2.2153762973326664e-07,
2893
+ "loss": 0.0604,
2894
+ "step": 2060
2895
+ },
2896
+ {
2897
+ "epoch": 2.8522099447513813,
2898
+ "grad_norm": 0.6037606596946716,
2899
+ "learning_rate": 2.0240950954356464e-07,
2900
+ "loss": 0.067,
2901
+ "step": 2065
2902
+ },
2903
+ {
2904
+ "epoch": 2.8591160220994474,
2905
+ "grad_norm": 0.5121417045593262,
2906
+ "learning_rate": 1.8413927658614714e-07,
2907
+ "loss": 0.0597,
2908
+ "step": 2070
2909
+ },
2910
+ {
2911
+ "epoch": 2.866022099447514,
2912
+ "grad_norm": 0.539158046245575,
2913
+ "learning_rate": 1.667279900739732e-07,
2914
+ "loss": 0.0579,
2915
+ "step": 2075
2916
+ },
2917
+ {
2918
+ "epoch": 2.87292817679558,
2919
+ "grad_norm": 0.5444369316101074,
2920
+ "learning_rate": 1.5017665942275882e-07,
2921
+ "loss": 0.0632,
2922
+ "step": 2080
2923
+ },
2924
+ {
2925
+ "epoch": 2.8798342541436464,
2926
+ "grad_norm": 0.6921401023864746,
2927
+ "learning_rate": 1.3448624419245382e-07,
2928
+ "loss": 0.0609,
2929
+ "step": 2085
2930
+ },
2931
+ {
2932
+ "epoch": 2.886740331491713,
2933
+ "grad_norm": 0.5181390643119812,
2934
+ "learning_rate": 1.1965765403162465e-07,
2935
+ "loss": 0.0609,
2936
+ "step": 2090
2937
+ },
2938
+ {
2939
+ "epoch": 2.893646408839779,
2940
+ "grad_norm": 0.5733912587165833,
2941
+ "learning_rate": 1.056917486247e-07,
2942
+ "loss": 0.0554,
2943
+ "step": 2095
2944
+ },
2945
+ {
2946
+ "epoch": 2.9005524861878453,
2947
+ "grad_norm": 0.6415034532546997,
2948
+ "learning_rate": 9.258933764214051e-08,
2949
+ "loss": 0.0592,
2950
+ "step": 2100
2951
+ },
2952
+ {
2953
+ "epoch": 2.9074585635359114,
2954
+ "grad_norm": 0.6035104990005493,
2955
+ "learning_rate": 8.035118069350311e-08,
2956
+ "loss": 0.0624,
2957
+ "step": 2105
2958
+ },
2959
+ {
2960
+ "epoch": 2.914364640883978,
2961
+ "grad_norm": 0.6024800539016724,
2962
+ "learning_rate": 6.897798728338955e-08,
2963
+ "loss": 0.0638,
2964
+ "step": 2110
2965
+ },
2966
+ {
2967
+ "epoch": 2.9212707182320443,
2968
+ "grad_norm": 0.5170609951019287,
2969
+ "learning_rate": 5.8470416770324274e-08,
2970
+ "loss": 0.0625,
2971
+ "step": 2115
2972
+ },
2973
+ {
2974
+ "epoch": 2.9281767955801103,
2975
+ "grad_norm": 0.5538831949234009,
2976
+ "learning_rate": 4.882907832852335e-08,
2977
+ "loss": 0.0654,
2978
+ "step": 2120
2979
+ },
2980
+ {
2981
+ "epoch": 2.935082872928177,
2982
+ "grad_norm": 0.673328161239624,
2983
+ "learning_rate": 4.005453091257938e-08,
2984
+ "loss": 0.0617,
2985
+ "step": 2125
2986
+ },
2987
+ {
2988
+ "epoch": 2.941988950276243,
2989
+ "grad_norm": 0.6543774008750916,
2990
+ "learning_rate": 3.214728322505245e-08,
2991
+ "loss": 0.0653,
2992
+ "step": 2130
2993
+ },
2994
+ {
2995
+ "epoch": 2.9488950276243093,
2996
+ "grad_norm": 0.4813374876976013,
2997
+ "learning_rate": 2.5107793686985346e-08,
2998
+ "loss": 0.0643,
2999
+ "step": 2135
3000
+ },
3001
+ {
3002
+ "epoch": 2.955801104972376,
3003
+ "grad_norm": 0.597122311592102,
3004
+ "learning_rate": 1.893647041131985e-08,
3005
+ "loss": 0.0615,
3006
+ "step": 2140
3007
+ },
3008
+ {
3009
+ "epoch": 2.962707182320442,
3010
+ "grad_norm": 0.5757369995117188,
3011
+ "learning_rate": 1.3633671179238972e-08,
3012
+ "loss": 0.055,
3013
+ "step": 2145
3014
+ },
3015
+ {
3016
+ "epoch": 2.9696132596685083,
3017
+ "grad_norm": 0.5769405961036682,
3018
+ "learning_rate": 9.199703419426885e-09,
3019
+ "loss": 0.0617,
3020
+ "step": 2150
3021
+ },
3022
+ {
3023
+ "epoch": 2.9765193370165743,
3024
+ "grad_norm": 0.4745386838912964,
3025
+ "learning_rate": 5.634824190243193e-09,
3026
+ "loss": 0.0533,
3027
+ "step": 2155
3028
+ },
3029
+ {
3030
+ "epoch": 2.983425414364641,
3031
+ "grad_norm": 0.5590378046035767,
3032
+ "learning_rate": 2.939240164818169e-09,
3033
+ "loss": 0.0702,
3034
+ "step": 2160
3035
+ },
3036
+ {
3037
+ "epoch": 2.9903314917127073,
3038
+ "grad_norm": 0.6124696731567383,
3039
+ "learning_rate": 1.1131076190756818e-09,
3040
+ "loss": 0.0687,
3041
+ "step": 2165
3042
+ },
3043
+ {
3044
+ "epoch": 2.9972375690607733,
3045
+ "grad_norm": 0.6062889695167542,
3046
+ "learning_rate": 1.5653242267210655e-10,
3047
+ "loss": 0.0572,
3048
+ "step": 2170
3049
+ },
3050
+ {
3051
+ "epoch": 3.0,
3052
+ "step": 2172,
3053
+ "total_flos": 2.72283124759108e+18,
3054
+ "train_loss": 0.35041305541470785,
3055
+ "train_runtime": 1331.4434,
3056
+ "train_samples_per_second": 52.173,
3057
+ "train_steps_per_second": 1.631
3058
+ }
3059
+ ],
3060
+ "logging_steps": 5,
3061
+ "max_steps": 2172,
3062
+ "num_input_tokens_seen": 0,
3063
+ "num_train_epochs": 3,
3064
+ "save_steps": 20000,
3065
+ "stateful_callbacks": {
3066
+ "TrainerControl": {
3067
+ "args": {
3068
+ "should_epoch_stop": false,
3069
+ "should_evaluate": false,
3070
+ "should_log": false,
3071
+ "should_save": false,
3072
+ "should_training_stop": false
3073
+ },
3074
+ "attributes": {}
3075
+ }
3076
+ },
3077
+ "total_flos": 2.72283124759108e+18,
3078
+ "train_batch_size": 2,
3079
+ "trial_name": null,
3080
+ "trial_params": null
3081
+ }
instruct/102_128_e3_3e-5/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3e59ad54ba7ffd2ac358783617c9d09fe50b1be2699427766f43ca962661042
3
+ size 8273
instruct/102_128_e3_3e-5/vocab.json ADDED
The diff for this file is too large to render. See raw diff