Hugging Face's logo

john20031010
/
john_cpt

Model card Files
xet
Community
john20031010 commited on
Commit
0b7263b
·
verified ·
1 Parent(s): 357e3f3

Upload 10 files

Browse files
Files changed (10) hide show
  1. adapter_config.json +53 -0
  2. merges.txt +0 -0
  3. rng_state.pth +3 -0
  4. scheduler.pt +3 -0
  5. special_tokens_map.json +49 -0
  6. tokenizer.json +0 -0
  7. tokenizer_config.json +170 -0
  8. trainer_state.json +3079 -0
  9. training_args.bin +3 -0
  10. vocab.json +0 -0
adapter_config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": {
6
+ "base_model_class": "LlamaForCausalLM",
7
+ "parent_library": "transformers.models.llama.modeling_llama",
8
+ "unsloth_fixed": true
9
+ },
10
+ "base_model_name_or_path": "unsloth/smollm-135m-bnb-4bit",
11
+ "bias": "none",
12
+ "corda_config": null,
13
+ "ensure_weight_tying": false,
14
+ "eva_config": null,
15
+ "exclude_modules": null,
16
+ "fan_in_fan_out": false,
17
+ "inference_mode": true,
18
+ "init_lora_weights": true,
19
+ "layer_replication": null,
20
+ "layers_pattern": null,
21
+ "layers_to_transform": null,
22
+ "loftq_config": {},
23
+ "lora_alpha": 32,
24
+ "lora_bias": false,
25
+ "lora_dropout": 0,
26
+ "megatron_config": null,
27
+ "megatron_core": "megatron.core",
28
+ "modules_to_save": [
29
+ "embed_tokens",
30
+ "lm_head"
31
+ ],
32
+ "peft_type": "LORA",
33
+ "peft_version": "0.18.0",
34
+ "qalora_group_size": 16,
35
+ "r": 128,
36
+ "rank_pattern": {},
37
+ "revision": null,
38
+ "target_modules": [
39
+ "o_proj",
40
+ "k_proj",
41
+ "up_proj",
42
+ "v_proj",
43
+ "down_proj",
44
+ "q_proj",
45
+ "gate_proj"
46
+ ],
47
+ "target_parameters": null,
48
+ "task_type": "CAUSAL_LM",
49
+ "trainable_token_indices": null,
50
+ "use_dora": false,
51
+ "use_qalora": false,
52
+ "use_rslora": true
53
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89645f9f6a2a8823b334e0748e4f5ea683a27b3d02632f0a241397db55d82fad
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c4db86c21ca9370eecb96212ca97575cde37ddfcb1392b60052c88c1a7be21a
3
+ size 1465
special_tokens_map.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|endoftext|>",
4
+ "<|im_start|>",
5
+ "<|im_end|>",
6
+ "<repo_name>",
7
+ "<reponame>",
8
+ "<file_sep>",
9
+ "<filename>",
10
+ "<gh_stars>",
11
+ "<issue_start>",
12
+ "<issue_comment>",
13
+ "<issue_closed>",
14
+ "<jupyter_start>",
15
+ "<jupyter_text>",
16
+ "<jupyter_code>",
17
+ "<jupyter_output>",
18
+ "<jupyter_script>",
19
+ "<empty_output>"
20
+ ],
21
+ "bos_token": {
22
+ "content": "<|endoftext|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false
27
+ },
28
+ "eos_token": {
29
+ "content": "<|endoftext|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false
34
+ },
35
+ "pad_token": {
36
+ "content": "<empty_output>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false
41
+ },
42
+ "unk_token": {
43
+ "content": "<|endoftext|>",
44
+ "lstrip": false,
45
+ "normalized": false,
46
+ "rstrip": false,
47
+ "single_word": false
48
+ }
49
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,170 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<repo_name>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "4": {
37
+ "content": "<reponame>",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "5": {
45
+ "content": "<file_sep>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": true
51
+ },
52
+ "6": {
53
+ "content": "<filename>",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": true
59
+ },
60
+ "7": {
61
+ "content": "<gh_stars>",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": true
67
+ },
68
+ "8": {
69
+ "content": "<issue_start>",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "9": {
77
+ "content": "<issue_comment>",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "10": {
85
+ "content": "<issue_closed>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": true
91
+ },
92
+ "11": {
93
+ "content": "<jupyter_start>",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": true
99
+ },
100
+ "12": {
101
+ "content": "<jupyter_text>",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": true
107
+ },
108
+ "13": {
109
+ "content": "<jupyter_code>",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": true
115
+ },
116
+ "14": {
117
+ "content": "<jupyter_output>",
118
+ "lstrip": false,
119
+ "normalized": false,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": true
123
+ },
124
+ "15": {
125
+ "content": "<jupyter_script>",
126
+ "lstrip": false,
127
+ "normalized": false,
128
+ "rstrip": false,
129
+ "single_word": false,
130
+ "special": true
131
+ },
132
+ "16": {
133
+ "content": "<empty_output>",
134
+ "lstrip": false,
135
+ "normalized": false,
136
+ "rstrip": false,
137
+ "single_word": false,
138
+ "special": true
139
+ }
140
+ },
141
+ "additional_special_tokens": [
142
+ "<|endoftext|>",
143
+ "<|im_start|>",
144
+ "<|im_end|>",
145
+ "<repo_name>",
146
+ "<reponame>",
147
+ "<file_sep>",
148
+ "<filename>",
149
+ "<gh_stars>",
150
+ "<issue_start>",
151
+ "<issue_comment>",
152
+ "<issue_closed>",
153
+ "<jupyter_start>",
154
+ "<jupyter_text>",
155
+ "<jupyter_code>",
156
+ "<jupyter_output>",
157
+ "<jupyter_script>",
158
+ "<empty_output>"
159
+ ],
160
+ "bos_token": "<|endoftext|>",
161
+ "clean_up_tokenization_spaces": false,
162
+ "eos_token": "<|endoftext|>",
163
+ "extra_special_tokens": {},
164
+ "model_max_length": 2048,
165
+ "pad_token": "<empty_output>",
166
+ "padding_side": "right",
167
+ "tokenizer_class": "GPT2Tokenizer",
168
+ "unk_token": "<|endoftext|>",
169
+ "vocab_size": 49152
170
+ }
trainer_state.json ADDED
@@ -0,0 +1,3079 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 435,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.011510791366906475,
14
+ "grad_norm": 1.295614242553711,
15
+ "learning_rate": 0.0,
16
+ "loss": 2.0611,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.02302158273381295,
21
+ "grad_norm": 1.3193614482879639,
22
+ "learning_rate": 5.000000000000001e-07,
23
+ "loss": 2.2646,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.034532374100719423,
28
+ "grad_norm": 1.3311564922332764,
29
+ "learning_rate": 1.0000000000000002e-06,
30
+ "loss": 2.302,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.0460431654676259,
35
+ "grad_norm": 1.3216726779937744,
36
+ "learning_rate": 1.5e-06,
37
+ "loss": 2.2383,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.05755395683453238,
42
+ "grad_norm": 1.3751393556594849,
43
+ "learning_rate": 2.0000000000000003e-06,
44
+ "loss": 2.2098,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.06906474820143885,
49
+ "grad_norm": 1.3023695945739746,
50
+ "learning_rate": 2.5e-06,
51
+ "loss": 2.1331,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.08057553956834532,
56
+ "grad_norm": 1.271135926246643,
57
+ "learning_rate": 3e-06,
58
+ "loss": 2.2544,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.0920863309352518,
63
+ "grad_norm": 1.3196772336959839,
64
+ "learning_rate": 3.5e-06,
65
+ "loss": 2.3242,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.10359712230215827,
70
+ "grad_norm": 1.251950740814209,
71
+ "learning_rate": 4.000000000000001e-06,
72
+ "loss": 2.1899,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.11510791366906475,
77
+ "grad_norm": 1.2269312143325806,
78
+ "learning_rate": 4.5e-06,
79
+ "loss": 2.1637,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.12661870503597122,
84
+ "grad_norm": 1.3193552494049072,
85
+ "learning_rate": 5e-06,
86
+ "loss": 2.2172,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.1381294964028777,
91
+ "grad_norm": 1.2646515369415283,
92
+ "learning_rate": 4.988235294117647e-06,
93
+ "loss": 2.1069,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.14964028776978416,
98
+ "grad_norm": 1.301425576210022,
99
+ "learning_rate": 4.976470588235294e-06,
100
+ "loss": 2.2276,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.16115107913669063,
105
+ "grad_norm": 1.1908143758773804,
106
+ "learning_rate": 4.964705882352942e-06,
107
+ "loss": 2.1751,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.17266187050359713,
112
+ "grad_norm": 1.1371599435806274,
113
+ "learning_rate": 4.9529411764705885e-06,
114
+ "loss": 2.2119,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.1841726618705036,
119
+ "grad_norm": 1.1369460821151733,
120
+ "learning_rate": 4.941176470588236e-06,
121
+ "loss": 2.1352,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.19568345323741007,
126
+ "grad_norm": 1.2164900302886963,
127
+ "learning_rate": 4.929411764705882e-06,
128
+ "loss": 2.1389,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.20719424460431654,
133
+ "grad_norm": 1.1954495906829834,
134
+ "learning_rate": 4.91764705882353e-06,
135
+ "loss": 2.1272,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.218705035971223,
140
+ "grad_norm": 1.077544927597046,
141
+ "learning_rate": 4.905882352941177e-06,
142
+ "loss": 2.2048,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.2302158273381295,
147
+ "grad_norm": 1.064244270324707,
148
+ "learning_rate": 4.894117647058824e-06,
149
+ "loss": 2.1051,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.24172661870503598,
154
+ "grad_norm": 1.045543909072876,
155
+ "learning_rate": 4.882352941176471e-06,
156
+ "loss": 2.1624,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.25323741007194245,
161
+ "grad_norm": 1.1219857931137085,
162
+ "learning_rate": 4.870588235294118e-06,
163
+ "loss": 2.0821,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.2647482014388489,
168
+ "grad_norm": 1.0319204330444336,
169
+ "learning_rate": 4.858823529411766e-06,
170
+ "loss": 2.2232,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.2762589928057554,
175
+ "grad_norm": 0.9106553792953491,
176
+ "learning_rate": 4.847058823529412e-06,
177
+ "loss": 2.0455,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.28776978417266186,
182
+ "grad_norm": 0.9222265481948853,
183
+ "learning_rate": 4.835294117647059e-06,
184
+ "loss": 2.0065,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.2992805755395683,
189
+ "grad_norm": 0.9210799336433411,
190
+ "learning_rate": 4.823529411764706e-06,
191
+ "loss": 2.2051,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.3107913669064748,
196
+ "grad_norm": 0.95929354429245,
197
+ "learning_rate": 4.811764705882354e-06,
198
+ "loss": 2.293,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.32230215827338127,
203
+ "grad_norm": 0.9271309971809387,
204
+ "learning_rate": 4.800000000000001e-06,
205
+ "loss": 2.1646,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.3338129496402878,
210
+ "grad_norm": 0.8561931848526001,
211
+ "learning_rate": 4.7882352941176475e-06,
212
+ "loss": 2.151,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.34532374100719426,
217
+ "grad_norm": 0.8737633228302002,
218
+ "learning_rate": 4.776470588235294e-06,
219
+ "loss": 2.2063,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.35683453237410073,
224
+ "grad_norm": 0.8623224496841431,
225
+ "learning_rate": 4.764705882352941e-06,
226
+ "loss": 2.0332,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.3683453237410072,
231
+ "grad_norm": 0.8182441592216492,
232
+ "learning_rate": 4.752941176470589e-06,
233
+ "loss": 2.1344,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.37985611510791367,
238
+ "grad_norm": 0.8933826088905334,
239
+ "learning_rate": 4.741176470588236e-06,
240
+ "loss": 2.2029,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.39136690647482014,
245
+ "grad_norm": 0.870568037033081,
246
+ "learning_rate": 4.729411764705883e-06,
247
+ "loss": 2.2155,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.4028776978417266,
252
+ "grad_norm": 0.8473733067512512,
253
+ "learning_rate": 4.717647058823529e-06,
254
+ "loss": 2.1522,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.4143884892086331,
259
+ "grad_norm": 0.8416939377784729,
260
+ "learning_rate": 4.705882352941177e-06,
261
+ "loss": 2.1937,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.42589928057553955,
266
+ "grad_norm": 0.7861284613609314,
267
+ "learning_rate": 4.694117647058824e-06,
268
+ "loss": 2.1359,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.437410071942446,
273
+ "grad_norm": 0.7433235049247742,
274
+ "learning_rate": 4.682352941176471e-06,
275
+ "loss": 2.1636,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.4489208633093525,
280
+ "grad_norm": 0.7175543308258057,
281
+ "learning_rate": 4.670588235294118e-06,
282
+ "loss": 1.9667,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.460431654676259,
287
+ "grad_norm": 0.7294153571128845,
288
+ "learning_rate": 4.658823529411765e-06,
289
+ "loss": 2.1362,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.4719424460431655,
294
+ "grad_norm": 0.7124460935592651,
295
+ "learning_rate": 4.647058823529412e-06,
296
+ "loss": 2.0463,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.48345323741007196,
301
+ "grad_norm": 0.7458817362785339,
302
+ "learning_rate": 4.635294117647059e-06,
303
+ "loss": 2.18,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.4949640287769784,
308
+ "grad_norm": 0.6650211215019226,
309
+ "learning_rate": 4.623529411764706e-06,
310
+ "loss": 2.0119,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.5064748201438849,
315
+ "grad_norm": 0.7400155067443848,
316
+ "learning_rate": 4.611764705882353e-06,
317
+ "loss": 2.0537,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.5179856115107914,
322
+ "grad_norm": 0.6916301250457764,
323
+ "learning_rate": 4.600000000000001e-06,
324
+ "loss": 2.0584,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.5294964028776978,
329
+ "grad_norm": 0.6942080855369568,
330
+ "learning_rate": 4.588235294117647e-06,
331
+ "loss": 2.1126,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.5410071942446043,
336
+ "grad_norm": 0.694041907787323,
337
+ "learning_rate": 4.5764705882352945e-06,
338
+ "loss": 2.1958,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.5525179856115108,
343
+ "grad_norm": 0.6456537842750549,
344
+ "learning_rate": 4.564705882352941e-06,
345
+ "loss": 2.212,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.5640287769784172,
350
+ "grad_norm": 0.663451075553894,
351
+ "learning_rate": 4.552941176470589e-06,
352
+ "loss": 2.1401,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.5755395683453237,
357
+ "grad_norm": 0.7182263135910034,
358
+ "learning_rate": 4.541176470588236e-06,
359
+ "loss": 2.137,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.5870503597122302,
364
+ "grad_norm": 0.6442857384681702,
365
+ "learning_rate": 4.529411764705883e-06,
366
+ "loss": 2.0755,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.5985611510791367,
371
+ "grad_norm": 0.6306608319282532,
372
+ "learning_rate": 4.51764705882353e-06,
373
+ "loss": 2.0456,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.6100719424460431,
378
+ "grad_norm": 0.628402054309845,
379
+ "learning_rate": 4.505882352941176e-06,
380
+ "loss": 2.0548,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.6215827338129496,
385
+ "grad_norm": 0.6245840191841125,
386
+ "learning_rate": 4.494117647058824e-06,
387
+ "loss": 2.176,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.6330935251798561,
392
+ "grad_norm": 0.5973242521286011,
393
+ "learning_rate": 4.482352941176471e-06,
394
+ "loss": 2.004,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.6446043165467625,
399
+ "grad_norm": 0.6911327242851257,
400
+ "learning_rate": 4.4705882352941184e-06,
401
+ "loss": 2.0991,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.6561151079136691,
406
+ "grad_norm": 0.6195106506347656,
407
+ "learning_rate": 4.458823529411765e-06,
408
+ "loss": 2.0232,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.6676258992805756,
413
+ "grad_norm": 0.5956724882125854,
414
+ "learning_rate": 4.447058823529412e-06,
415
+ "loss": 2.1384,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.679136690647482,
420
+ "grad_norm": 0.6167479753494263,
421
+ "learning_rate": 4.435294117647059e-06,
422
+ "loss": 2.0524,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.6906474820143885,
427
+ "grad_norm": 0.5958898663520813,
428
+ "learning_rate": 4.423529411764707e-06,
429
+ "loss": 2.0854,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.702158273381295,
434
+ "grad_norm": 0.6348150372505188,
435
+ "learning_rate": 4.411764705882353e-06,
436
+ "loss": 2.1037,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.7136690647482015,
441
+ "grad_norm": 0.5777585506439209,
442
+ "learning_rate": 4.4e-06,
443
+ "loss": 2.0463,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.7251798561151079,
448
+ "grad_norm": 0.5744509696960449,
449
+ "learning_rate": 4.388235294117648e-06,
450
+ "loss": 2.0368,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.7366906474820144,
455
+ "grad_norm": 0.6182202696800232,
456
+ "learning_rate": 4.376470588235294e-06,
457
+ "loss": 2.1305,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.7482014388489209,
462
+ "grad_norm": 0.6051247715950012,
463
+ "learning_rate": 4.3647058823529415e-06,
464
+ "loss": 1.9982,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.7597122302158273,
469
+ "grad_norm": 0.585530161857605,
470
+ "learning_rate": 4.352941176470588e-06,
471
+ "loss": 2.1311,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.7712230215827338,
476
+ "grad_norm": 0.5718886256217957,
477
+ "learning_rate": 4.341176470588236e-06,
478
+ "loss": 2.0899,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.7827338129496403,
483
+ "grad_norm": 0.5871637463569641,
484
+ "learning_rate": 4.329411764705883e-06,
485
+ "loss": 2.1422,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.7942446043165468,
490
+ "grad_norm": 0.5348390340805054,
491
+ "learning_rate": 4.31764705882353e-06,
492
+ "loss": 2.0414,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.8057553956834532,
497
+ "grad_norm": 0.5360246896743774,
498
+ "learning_rate": 4.305882352941177e-06,
499
+ "loss": 2.0697,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.8172661870503597,
504
+ "grad_norm": 0.575326144695282,
505
+ "learning_rate": 4.294117647058823e-06,
506
+ "loss": 2.126,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.8287769784172662,
511
+ "grad_norm": 0.5389431118965149,
512
+ "learning_rate": 4.282352941176471e-06,
513
+ "loss": 1.9699,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.8402877697841726,
518
+ "grad_norm": 0.5181302428245544,
519
+ "learning_rate": 4.270588235294118e-06,
520
+ "loss": 2.0431,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.8517985611510791,
525
+ "grad_norm": 0.5258436799049377,
526
+ "learning_rate": 4.2588235294117655e-06,
527
+ "loss": 2.1398,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.8633093525179856,
532
+ "grad_norm": 0.5759520530700684,
533
+ "learning_rate": 4.247058823529412e-06,
534
+ "loss": 2.1257,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.874820143884892,
539
+ "grad_norm": 0.5312909483909607,
540
+ "learning_rate": 4.235294117647059e-06,
541
+ "loss": 2.0202,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.8863309352517985,
546
+ "grad_norm": 0.6128862500190735,
547
+ "learning_rate": 4.223529411764706e-06,
548
+ "loss": 2.1821,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.897841726618705,
553
+ "grad_norm": 0.5947574377059937,
554
+ "learning_rate": 4.211764705882354e-06,
555
+ "loss": 2.0586,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.9093525179856116,
560
+ "grad_norm": 0.5092775225639343,
561
+ "learning_rate": 4.2000000000000004e-06,
562
+ "loss": 2.0783,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.920863309352518,
567
+ "grad_norm": 0.5344525575637817,
568
+ "learning_rate": 4.188235294117647e-06,
569
+ "loss": 2.1457,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.9323741007194245,
574
+ "grad_norm": 0.5249314308166504,
575
+ "learning_rate": 4.176470588235295e-06,
576
+ "loss": 1.9904,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.943884892086331,
581
+ "grad_norm": 0.5562986135482788,
582
+ "learning_rate": 4.164705882352941e-06,
583
+ "loss": 2.053,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.9553956834532374,
588
+ "grad_norm": 0.5227307081222534,
589
+ "learning_rate": 4.152941176470589e-06,
590
+ "loss": 1.9463,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.9669064748201439,
595
+ "grad_norm": 0.5479752421379089,
596
+ "learning_rate": 4.141176470588235e-06,
597
+ "loss": 2.1546,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.9784172661870504,
602
+ "grad_norm": 0.49207690358161926,
603
+ "learning_rate": 4.129411764705883e-06,
604
+ "loss": 1.9769,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.9899280575539569,
609
+ "grad_norm": 0.5519751310348511,
610
+ "learning_rate": 4.11764705882353e-06,
611
+ "loss": 2.0985,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 1.0,
616
+ "grad_norm": 0.5100968480110168,
617
+ "learning_rate": 4.105882352941177e-06,
618
+ "loss": 2.0412,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 1.0115107913669066,
623
+ "grad_norm": 0.5197513103485107,
624
+ "learning_rate": 4.094117647058824e-06,
625
+ "loss": 2.0366,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 1.023021582733813,
630
+ "grad_norm": 0.475099116563797,
631
+ "learning_rate": 4.082352941176471e-06,
632
+ "loss": 1.9594,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 1.0345323741007195,
637
+ "grad_norm": 0.4655166566371918,
638
+ "learning_rate": 4.070588235294118e-06,
639
+ "loss": 1.988,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 1.0460431654676259,
644
+ "grad_norm": 0.47748327255249023,
645
+ "learning_rate": 4.058823529411765e-06,
646
+ "loss": 2.0573,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 1.0575539568345325,
651
+ "grad_norm": 0.4752672612667084,
652
+ "learning_rate": 4.0470588235294125e-06,
653
+ "loss": 2.0761,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 1.0690647482014388,
658
+ "grad_norm": 0.42830637097358704,
659
+ "learning_rate": 4.0352941176470585e-06,
660
+ "loss": 1.9,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 1.0805755395683454,
665
+ "grad_norm": 0.509665310382843,
666
+ "learning_rate": 4.023529411764706e-06,
667
+ "loss": 2.1022,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 1.0920863309352518,
672
+ "grad_norm": 0.4985044300556183,
673
+ "learning_rate": 4.011764705882353e-06,
674
+ "loss": 2.0004,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 1.1035971223021583,
679
+ "grad_norm": 0.4855203330516815,
680
+ "learning_rate": 4.000000000000001e-06,
681
+ "loss": 1.9497,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 1.1151079136690647,
686
+ "grad_norm": 0.5421211123466492,
687
+ "learning_rate": 3.9882352941176475e-06,
688
+ "loss": 2.0288,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 1.1266187050359713,
693
+ "grad_norm": 0.4415021538734436,
694
+ "learning_rate": 3.976470588235294e-06,
695
+ "loss": 1.9209,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 1.1381294964028776,
700
+ "grad_norm": 0.5151281356811523,
701
+ "learning_rate": 3.964705882352942e-06,
702
+ "loss": 2.0827,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 1.1496402877697842,
707
+ "grad_norm": 0.4920862913131714,
708
+ "learning_rate": 3.952941176470588e-06,
709
+ "loss": 2.058,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 1.1611510791366906,
714
+ "grad_norm": 0.4780770540237427,
715
+ "learning_rate": 3.941176470588236e-06,
716
+ "loss": 2.0177,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 1.1726618705035972,
721
+ "grad_norm": 0.5100148916244507,
722
+ "learning_rate": 3.9294117647058824e-06,
723
+ "loss": 2.0284,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 1.1841726618705035,
728
+ "grad_norm": 0.49509063363075256,
729
+ "learning_rate": 3.91764705882353e-06,
730
+ "loss": 1.9646,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 1.19568345323741,
735
+ "grad_norm": 0.518322229385376,
736
+ "learning_rate": 3.905882352941177e-06,
737
+ "loss": 1.9543,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 1.2071942446043165,
742
+ "grad_norm": 0.49948441982269287,
743
+ "learning_rate": 3.894117647058824e-06,
744
+ "loss": 1.9506,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 1.218705035971223,
749
+ "grad_norm": 0.4965578317642212,
750
+ "learning_rate": 3.882352941176471e-06,
751
+ "loss": 1.9751,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 1.2302158273381294,
756
+ "grad_norm": 0.4927343428134918,
757
+ "learning_rate": 3.870588235294118e-06,
758
+ "loss": 2.0053,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 1.241726618705036,
763
+ "grad_norm": 0.5349815487861633,
764
+ "learning_rate": 3.858823529411765e-06,
765
+ "loss": 1.9543,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 1.2532374100719426,
770
+ "grad_norm": 0.47634226083755493,
771
+ "learning_rate": 3.847058823529412e-06,
772
+ "loss": 2.0552,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 1.264748201438849,
777
+ "grad_norm": 0.4937780797481537,
778
+ "learning_rate": 3.8352941176470596e-06,
779
+ "loss": 2.0465,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 1.2762589928057553,
784
+ "grad_norm": 0.49782195687294006,
785
+ "learning_rate": 3.8235294117647055e-06,
786
+ "loss": 2.0029,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 1.2877697841726619,
791
+ "grad_norm": 0.4511786103248596,
792
+ "learning_rate": 3.8117647058823532e-06,
793
+ "loss": 2.0275,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 1.2992805755395684,
798
+ "grad_norm": 0.5700087547302246,
799
+ "learning_rate": 3.8000000000000005e-06,
800
+ "loss": 2.1454,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 1.3107913669064748,
805
+ "grad_norm": 0.46322473883628845,
806
+ "learning_rate": 3.7882352941176477e-06,
807
+ "loss": 2.1094,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 1.3223021582733812,
812
+ "grad_norm": 0.48486411571502686,
813
+ "learning_rate": 3.776470588235294e-06,
814
+ "loss": 1.9504,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 1.3338129496402877,
819
+ "grad_norm": 0.5094816088676453,
820
+ "learning_rate": 3.7647058823529414e-06,
821
+ "loss": 2.0014,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 1.3453237410071943,
826
+ "grad_norm": 0.4983859360218048,
827
+ "learning_rate": 3.7529411764705886e-06,
828
+ "loss": 2.0876,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 1.3568345323741007,
833
+ "grad_norm": 0.46220195293426514,
834
+ "learning_rate": 3.741176470588236e-06,
835
+ "loss": 1.9317,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 1.3683453237410073,
840
+ "grad_norm": 0.5100168585777283,
841
+ "learning_rate": 3.7294117647058827e-06,
842
+ "loss": 2.0247,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 1.3798561151079136,
847
+ "grad_norm": 0.45993947982788086,
848
+ "learning_rate": 3.71764705882353e-06,
849
+ "loss": 2.1151,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 1.3913669064748202,
854
+ "grad_norm": 0.45173344016075134,
855
+ "learning_rate": 3.7058823529411767e-06,
856
+ "loss": 1.9784,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 1.4028776978417266,
861
+ "grad_norm": 0.5567683577537537,
862
+ "learning_rate": 3.6941176470588236e-06,
863
+ "loss": 1.9727,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 1.4143884892086331,
868
+ "grad_norm": 0.5246084928512573,
869
+ "learning_rate": 3.682352941176471e-06,
870
+ "loss": 1.9025,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 1.4258992805755395,
875
+ "grad_norm": 0.4744240343570709,
876
+ "learning_rate": 3.670588235294118e-06,
877
+ "loss": 1.9713,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 1.437410071942446,
882
+ "grad_norm": 0.4525962769985199,
883
+ "learning_rate": 3.6588235294117653e-06,
884
+ "loss": 1.9845,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 1.4489208633093524,
889
+ "grad_norm": 0.5006890296936035,
890
+ "learning_rate": 3.6470588235294117e-06,
891
+ "loss": 1.9638,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 1.460431654676259,
896
+ "grad_norm": 0.4702132046222687,
897
+ "learning_rate": 3.635294117647059e-06,
898
+ "loss": 1.9444,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 1.4719424460431654,
903
+ "grad_norm": 0.5455424189567566,
904
+ "learning_rate": 3.623529411764706e-06,
905
+ "loss": 1.9387,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 1.483453237410072,
910
+ "grad_norm": 0.6158833503723145,
911
+ "learning_rate": 3.6117647058823534e-06,
912
+ "loss": 2.0112,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 1.4949640287769785,
917
+ "grad_norm": 0.4596414268016815,
918
+ "learning_rate": 3.6000000000000003e-06,
919
+ "loss": 2.0323,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 1.506474820143885,
924
+ "grad_norm": 0.46785110235214233,
925
+ "learning_rate": 3.5882352941176475e-06,
926
+ "loss": 2.065,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 1.5179856115107913,
931
+ "grad_norm": 0.46331682801246643,
932
+ "learning_rate": 3.5764705882352948e-06,
933
+ "loss": 1.93,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 1.5294964028776978,
938
+ "grad_norm": 0.47538137435913086,
939
+ "learning_rate": 3.564705882352941e-06,
940
+ "loss": 2.0613,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 1.5410071942446044,
945
+ "grad_norm": 0.5323189496994019,
946
+ "learning_rate": 3.5529411764705884e-06,
947
+ "loss": 2.0175,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 1.5525179856115108,
952
+ "grad_norm": 0.4505595266819,
953
+ "learning_rate": 3.5411764705882356e-06,
954
+ "loss": 1.9124,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 1.5640287769784171,
959
+ "grad_norm": 0.4327537417411804,
960
+ "learning_rate": 3.529411764705883e-06,
961
+ "loss": 1.9633,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 1.5755395683453237,
966
+ "grad_norm": 0.497213214635849,
967
+ "learning_rate": 3.5176470588235297e-06,
968
+ "loss": 1.9673,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 1.5870503597122303,
973
+ "grad_norm": 0.4779418706893921,
974
+ "learning_rate": 3.5058823529411765e-06,
975
+ "loss": 2.0615,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 1.5985611510791367,
980
+ "grad_norm": 0.45400741696357727,
981
+ "learning_rate": 3.4941176470588238e-06,
982
+ "loss": 2.0023,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 1.610071942446043,
987
+ "grad_norm": 0.4582999646663666,
988
+ "learning_rate": 3.4823529411764706e-06,
989
+ "loss": 2.104,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 1.6215827338129496,
994
+ "grad_norm": 0.436869353055954,
995
+ "learning_rate": 3.470588235294118e-06,
996
+ "loss": 2.0327,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 1.6330935251798562,
1001
+ "grad_norm": 0.49151161313056946,
1002
+ "learning_rate": 3.458823529411765e-06,
1003
+ "loss": 1.8815,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 1.6446043165467625,
1008
+ "grad_norm": 0.4952029585838318,
1009
+ "learning_rate": 3.4470588235294123e-06,
1010
+ "loss": 1.9371,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 1.6561151079136691,
1015
+ "grad_norm": 0.4617934226989746,
1016
+ "learning_rate": 3.4352941176470587e-06,
1017
+ "loss": 2.0492,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 1.6676258992805755,
1022
+ "grad_norm": 0.45421457290649414,
1023
+ "learning_rate": 3.423529411764706e-06,
1024
+ "loss": 1.9799,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 1.679136690647482,
1029
+ "grad_norm": 0.45641016960144043,
1030
+ "learning_rate": 3.4117647058823532e-06,
1031
+ "loss": 2.1129,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 1.6906474820143886,
1036
+ "grad_norm": 0.4528232514858246,
1037
+ "learning_rate": 3.4000000000000005e-06,
1038
+ "loss": 1.9968,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 1.702158273381295,
1043
+ "grad_norm": 0.425224632024765,
1044
+ "learning_rate": 3.3882352941176473e-06,
1045
+ "loss": 2.0766,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 1.7136690647482014,
1050
+ "grad_norm": 0.44971901178359985,
1051
+ "learning_rate": 3.3764705882352946e-06,
1052
+ "loss": 1.951,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 1.725179856115108,
1057
+ "grad_norm": 0.5100486278533936,
1058
+ "learning_rate": 3.364705882352942e-06,
1059
+ "loss": 2.0126,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 1.7366906474820145,
1064
+ "grad_norm": 0.43650567531585693,
1065
+ "learning_rate": 3.352941176470588e-06,
1066
+ "loss": 2.0752,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 1.7482014388489209,
1071
+ "grad_norm": 0.4769532382488251,
1072
+ "learning_rate": 3.3411764705882354e-06,
1073
+ "loss": 2.005,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 1.7597122302158272,
1078
+ "grad_norm": 0.4705875813961029,
1079
+ "learning_rate": 3.3294117647058827e-06,
1080
+ "loss": 2.0744,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 1.7712230215827338,
1085
+ "grad_norm": 0.4628766179084778,
1086
+ "learning_rate": 3.31764705882353e-06,
1087
+ "loss": 2.108,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 1.7827338129496404,
1092
+ "grad_norm": 0.4407738149166107,
1093
+ "learning_rate": 3.3058823529411763e-06,
1094
+ "loss": 1.9779,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 1.7942446043165468,
1099
+ "grad_norm": 0.500913679599762,
1100
+ "learning_rate": 3.2941176470588236e-06,
1101
+ "loss": 1.8435,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 1.8057553956834531,
1106
+ "grad_norm": 0.4094080626964569,
1107
+ "learning_rate": 3.282352941176471e-06,
1108
+ "loss": 1.972,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 1.8172661870503597,
1113
+ "grad_norm": 0.46937379240989685,
1114
+ "learning_rate": 3.270588235294118e-06,
1115
+ "loss": 1.8564,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 1.8287769784172663,
1120
+ "grad_norm": 0.46270328760147095,
1121
+ "learning_rate": 3.258823529411765e-06,
1122
+ "loss": 2.0629,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 1.8402877697841726,
1127
+ "grad_norm": 0.452856183052063,
1128
+ "learning_rate": 3.247058823529412e-06,
1129
+ "loss": 2.0334,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 1.851798561151079,
1134
+ "grad_norm": 0.4600447714328766,
1135
+ "learning_rate": 3.2352941176470594e-06,
1136
+ "loss": 1.9833,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 1.8633093525179856,
1141
+ "grad_norm": 0.5156863927841187,
1142
+ "learning_rate": 3.2235294117647058e-06,
1143
+ "loss": 2.1083,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 1.8748201438848922,
1148
+ "grad_norm": 0.4730561077594757,
1149
+ "learning_rate": 3.211764705882353e-06,
1150
+ "loss": 2.016,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 1.8863309352517985,
1155
+ "grad_norm": 0.4580685496330261,
1156
+ "learning_rate": 3.2000000000000003e-06,
1157
+ "loss": 2.0209,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 1.8978417266187049,
1162
+ "grad_norm": 0.438728004693985,
1163
+ "learning_rate": 3.1882352941176475e-06,
1164
+ "loss": 1.9078,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 1.9093525179856115,
1169
+ "grad_norm": 0.4413028061389923,
1170
+ "learning_rate": 3.1764705882352943e-06,
1171
+ "loss": 1.8775,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 1.920863309352518,
1176
+ "grad_norm": 0.4299080967903137,
1177
+ "learning_rate": 3.1647058823529416e-06,
1178
+ "loss": 2.0034,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 1.9323741007194246,
1183
+ "grad_norm": 0.47266408801078796,
1184
+ "learning_rate": 3.1529411764705884e-06,
1185
+ "loss": 1.8302,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 1.943884892086331,
1190
+ "grad_norm": 0.4524175226688385,
1191
+ "learning_rate": 3.1411764705882357e-06,
1192
+ "loss": 1.967,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 1.9553956834532373,
1197
+ "grad_norm": 0.39914512634277344,
1198
+ "learning_rate": 3.1294117647058825e-06,
1199
+ "loss": 2.0307,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 1.966906474820144,
1204
+ "grad_norm": 0.47265124320983887,
1205
+ "learning_rate": 3.1176470588235297e-06,
1206
+ "loss": 2.0066,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 1.9784172661870505,
1211
+ "grad_norm": 0.3866676986217499,
1212
+ "learning_rate": 3.105882352941177e-06,
1213
+ "loss": 2.0306,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 1.9899280575539569,
1218
+ "grad_norm": 0.4875778555870056,
1219
+ "learning_rate": 3.0941176470588234e-06,
1220
+ "loss": 1.926,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 2.0,
1225
+ "grad_norm": 0.467013418674469,
1226
+ "learning_rate": 3.0823529411764706e-06,
1227
+ "loss": 2.0708,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 2.0115107913669066,
1232
+ "grad_norm": 0.4312443435192108,
1233
+ "learning_rate": 3.070588235294118e-06,
1234
+ "loss": 1.9997,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 2.023021582733813,
1239
+ "grad_norm": 0.44231534004211426,
1240
+ "learning_rate": 3.058823529411765e-06,
1241
+ "loss": 2.0316,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 2.0345323741007193,
1246
+ "grad_norm": 0.43418049812316895,
1247
+ "learning_rate": 3.047058823529412e-06,
1248
+ "loss": 1.898,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 2.046043165467626,
1253
+ "grad_norm": 0.42247310280799866,
1254
+ "learning_rate": 3.035294117647059e-06,
1255
+ "loss": 1.9524,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 2.0575539568345325,
1260
+ "grad_norm": 0.47551754117012024,
1261
+ "learning_rate": 3.0235294117647064e-06,
1262
+ "loss": 1.9299,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 2.069064748201439,
1267
+ "grad_norm": 0.39652958512306213,
1268
+ "learning_rate": 3.011764705882353e-06,
1269
+ "loss": 2.0617,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 2.080575539568345,
1274
+ "grad_norm": 0.41868913173675537,
1275
+ "learning_rate": 3e-06,
1276
+ "loss": 1.9897,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 2.0920863309352518,
1281
+ "grad_norm": 0.39825567603111267,
1282
+ "learning_rate": 2.9882352941176473e-06,
1283
+ "loss": 1.9624,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 2.1035971223021583,
1288
+ "grad_norm": 0.4380688965320587,
1289
+ "learning_rate": 2.9764705882352946e-06,
1290
+ "loss": 1.9904,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 2.115107913669065,
1295
+ "grad_norm": 0.39882150292396545,
1296
+ "learning_rate": 2.9647058823529414e-06,
1297
+ "loss": 1.9084,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 2.126618705035971,
1302
+ "grad_norm": 0.4040652811527252,
1303
+ "learning_rate": 2.9529411764705882e-06,
1304
+ "loss": 2.1459,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 2.1381294964028776,
1309
+ "grad_norm": 0.4166700839996338,
1310
+ "learning_rate": 2.9411764705882355e-06,
1311
+ "loss": 2.0026,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 2.149640287769784,
1316
+ "grad_norm": 0.41756269335746765,
1317
+ "learning_rate": 2.9294117647058827e-06,
1318
+ "loss": 2.0675,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 2.161151079136691,
1323
+ "grad_norm": 0.41332557797431946,
1324
+ "learning_rate": 2.9176470588235295e-06,
1325
+ "loss": 1.9391,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 2.172661870503597,
1330
+ "grad_norm": 0.4298039972782135,
1331
+ "learning_rate": 2.9058823529411768e-06,
1332
+ "loss": 1.8671,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 2.1841726618705035,
1337
+ "grad_norm": 0.45794206857681274,
1338
+ "learning_rate": 2.894117647058824e-06,
1339
+ "loss": 1.9466,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 2.19568345323741,
1344
+ "grad_norm": 0.3940126597881317,
1345
+ "learning_rate": 2.8823529411764704e-06,
1346
+ "loss": 1.9557,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 2.2071942446043167,
1351
+ "grad_norm": 0.45463988184928894,
1352
+ "learning_rate": 2.8705882352941177e-06,
1353
+ "loss": 1.9659,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 2.218705035971223,
1358
+ "grad_norm": 0.4931933879852295,
1359
+ "learning_rate": 2.858823529411765e-06,
1360
+ "loss": 1.9761,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 2.2302158273381294,
1365
+ "grad_norm": 0.42029982805252075,
1366
+ "learning_rate": 2.847058823529412e-06,
1367
+ "loss": 2.0658,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 2.241726618705036,
1372
+ "grad_norm": 0.4499173164367676,
1373
+ "learning_rate": 2.835294117647059e-06,
1374
+ "loss": 2.0426,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 2.2532374100719426,
1379
+ "grad_norm": 0.42561689019203186,
1380
+ "learning_rate": 2.8235294117647062e-06,
1381
+ "loss": 1.9447,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 2.2647482014388487,
1386
+ "grad_norm": 0.4154174327850342,
1387
+ "learning_rate": 2.8117647058823535e-06,
1388
+ "loss": 1.8771,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 2.2762589928057553,
1393
+ "grad_norm": 0.40316736698150635,
1394
+ "learning_rate": 2.8000000000000003e-06,
1395
+ "loss": 2.0181,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 2.287769784172662,
1400
+ "grad_norm": 0.4346086084842682,
1401
+ "learning_rate": 2.788235294117647e-06,
1402
+ "loss": 1.9526,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 2.2992805755395684,
1407
+ "grad_norm": 0.4031694829463959,
1408
+ "learning_rate": 2.7764705882352944e-06,
1409
+ "loss": 2.0371,
1410
+ "step": 200
1411
+ },
1412
+ {
1413
+ "epoch": 2.310791366906475,
1414
+ "grad_norm": 0.46281856298446655,
1415
+ "learning_rate": 2.7647058823529416e-06,
1416
+ "loss": 1.9795,
1417
+ "step": 201
1418
+ },
1419
+ {
1420
+ "epoch": 2.322302158273381,
1421
+ "grad_norm": 0.41117939352989197,
1422
+ "learning_rate": 2.7529411764705884e-06,
1423
+ "loss": 1.93,
1424
+ "step": 202
1425
+ },
1426
+ {
1427
+ "epoch": 2.3338129496402877,
1428
+ "grad_norm": 0.4585975110530853,
1429
+ "learning_rate": 2.7411764705882353e-06,
1430
+ "loss": 2.0624,
1431
+ "step": 203
1432
+ },
1433
+ {
1434
+ "epoch": 2.3453237410071943,
1435
+ "grad_norm": 0.43110424280166626,
1436
+ "learning_rate": 2.7294117647058825e-06,
1437
+ "loss": 1.9551,
1438
+ "step": 204
1439
+ },
1440
+ {
1441
+ "epoch": 2.356834532374101,
1442
+ "grad_norm": 0.44794800877571106,
1443
+ "learning_rate": 2.7176470588235297e-06,
1444
+ "loss": 1.9633,
1445
+ "step": 205
1446
+ },
1447
+ {
1448
+ "epoch": 2.368345323741007,
1449
+ "grad_norm": 0.44358718395233154,
1450
+ "learning_rate": 2.7058823529411766e-06,
1451
+ "loss": 1.992,
1452
+ "step": 206
1453
+ },
1454
+ {
1455
+ "epoch": 2.3798561151079136,
1456
+ "grad_norm": 0.3666572868824005,
1457
+ "learning_rate": 2.694117647058824e-06,
1458
+ "loss": 1.9791,
1459
+ "step": 207
1460
+ },
1461
+ {
1462
+ "epoch": 2.39136690647482,
1463
+ "grad_norm": 0.4338827431201935,
1464
+ "learning_rate": 2.682352941176471e-06,
1465
+ "loss": 1.9804,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 2.402877697841727,
1470
+ "grad_norm": 0.3984374701976776,
1471
+ "learning_rate": 2.6705882352941175e-06,
1472
+ "loss": 2.0183,
1473
+ "step": 209
1474
+ },
1475
+ {
1476
+ "epoch": 2.414388489208633,
1477
+ "grad_norm": 0.44970276951789856,
1478
+ "learning_rate": 2.6588235294117647e-06,
1479
+ "loss": 1.9441,
1480
+ "step": 210
1481
+ },
1482
+ {
1483
+ "epoch": 2.4258992805755395,
1484
+ "grad_norm": 0.38036495447158813,
1485
+ "learning_rate": 2.647058823529412e-06,
1486
+ "loss": 2.0069,
1487
+ "step": 211
1488
+ },
1489
+ {
1490
+ "epoch": 2.437410071942446,
1491
+ "grad_norm": 0.40496423840522766,
1492
+ "learning_rate": 2.635294117647059e-06,
1493
+ "loss": 2.0429,
1494
+ "step": 212
1495
+ },
1496
+ {
1497
+ "epoch": 2.4489208633093527,
1498
+ "grad_norm": 0.4080514907836914,
1499
+ "learning_rate": 2.623529411764706e-06,
1500
+ "loss": 1.9265,
1501
+ "step": 213
1502
+ },
1503
+ {
1504
+ "epoch": 2.460431654676259,
1505
+ "grad_norm": 0.4119996130466461,
1506
+ "learning_rate": 2.6117647058823533e-06,
1507
+ "loss": 2.0675,
1508
+ "step": 214
1509
+ },
1510
+ {
1511
+ "epoch": 2.4719424460431654,
1512
+ "grad_norm": 0.4465183615684509,
1513
+ "learning_rate": 2.6e-06,
1514
+ "loss": 1.9325,
1515
+ "step": 215
1516
+ },
1517
+ {
1518
+ "epoch": 2.483453237410072,
1519
+ "grad_norm": 0.4121825397014618,
1520
+ "learning_rate": 2.5882352941176473e-06,
1521
+ "loss": 1.9198,
1522
+ "step": 216
1523
+ },
1524
+ {
1525
+ "epoch": 2.4949640287769785,
1526
+ "grad_norm": 0.41370537877082825,
1527
+ "learning_rate": 2.576470588235294e-06,
1528
+ "loss": 2.0955,
1529
+ "step": 217
1530
+ },
1531
+ {
1532
+ "epoch": 2.506474820143885,
1533
+ "grad_norm": 0.44053828716278076,
1534
+ "learning_rate": 2.5647058823529414e-06,
1535
+ "loss": 2.0383,
1536
+ "step": 218
1537
+ },
1538
+ {
1539
+ "epoch": 2.5179856115107913,
1540
+ "grad_norm": 0.42168861627578735,
1541
+ "learning_rate": 2.5529411764705887e-06,
1542
+ "loss": 1.9718,
1543
+ "step": 219
1544
+ },
1545
+ {
1546
+ "epoch": 2.529496402877698,
1547
+ "grad_norm": 0.4006345570087433,
1548
+ "learning_rate": 2.541176470588235e-06,
1549
+ "loss": 1.9606,
1550
+ "step": 220
1551
+ },
1552
+ {
1553
+ "epoch": 2.5410071942446044,
1554
+ "grad_norm": 0.40757259726524353,
1555
+ "learning_rate": 2.5294117647058823e-06,
1556
+ "loss": 2.0171,
1557
+ "step": 221
1558
+ },
1559
+ {
1560
+ "epoch": 2.5525179856115106,
1561
+ "grad_norm": 0.40500617027282715,
1562
+ "learning_rate": 2.5176470588235295e-06,
1563
+ "loss": 1.9153,
1564
+ "step": 222
1565
+ },
1566
+ {
1567
+ "epoch": 2.564028776978417,
1568
+ "grad_norm": 0.36510592699050903,
1569
+ "learning_rate": 2.505882352941177e-06,
1570
+ "loss": 1.8415,
1571
+ "step": 223
1572
+ },
1573
+ {
1574
+ "epoch": 2.5755395683453237,
1575
+ "grad_norm": 0.44662347435951233,
1576
+ "learning_rate": 2.4941176470588236e-06,
1577
+ "loss": 1.9275,
1578
+ "step": 224
1579
+ },
1580
+ {
1581
+ "epoch": 2.5870503597122303,
1582
+ "grad_norm": 0.4446622431278229,
1583
+ "learning_rate": 2.482352941176471e-06,
1584
+ "loss": 2.002,
1585
+ "step": 225
1586
+ },
1587
+ {
1588
+ "epoch": 2.598561151079137,
1589
+ "grad_norm": 0.46108031272888184,
1590
+ "learning_rate": 2.470588235294118e-06,
1591
+ "loss": 1.9739,
1592
+ "step": 226
1593
+ },
1594
+ {
1595
+ "epoch": 2.610071942446043,
1596
+ "grad_norm": 0.42075315117836,
1597
+ "learning_rate": 2.458823529411765e-06,
1598
+ "loss": 1.93,
1599
+ "step": 227
1600
+ },
1601
+ {
1602
+ "epoch": 2.6215827338129496,
1603
+ "grad_norm": 0.4763178527355194,
1604
+ "learning_rate": 2.447058823529412e-06,
1605
+ "loss": 1.8973,
1606
+ "step": 228
1607
+ },
1608
+ {
1609
+ "epoch": 2.633093525179856,
1610
+ "grad_norm": 0.39183807373046875,
1611
+ "learning_rate": 2.435294117647059e-06,
1612
+ "loss": 1.9485,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 2.6446043165467623,
1617
+ "grad_norm": 0.4360307455062866,
1618
+ "learning_rate": 2.423529411764706e-06,
1619
+ "loss": 1.902,
1620
+ "step": 230
1621
+ },
1622
+ {
1623
+ "epoch": 2.656115107913669,
1624
+ "grad_norm": 0.46272391080856323,
1625
+ "learning_rate": 2.411764705882353e-06,
1626
+ "loss": 1.9535,
1627
+ "step": 231
1628
+ },
1629
+ {
1630
+ "epoch": 2.6676258992805755,
1631
+ "grad_norm": 0.43482983112335205,
1632
+ "learning_rate": 2.4000000000000003e-06,
1633
+ "loss": 1.9474,
1634
+ "step": 232
1635
+ },
1636
+ {
1637
+ "epoch": 2.679136690647482,
1638
+ "grad_norm": 0.38162457942962646,
1639
+ "learning_rate": 2.388235294117647e-06,
1640
+ "loss": 1.935,
1641
+ "step": 233
1642
+ },
1643
+ {
1644
+ "epoch": 2.6906474820143886,
1645
+ "grad_norm": 0.4534943401813507,
1646
+ "learning_rate": 2.3764705882352944e-06,
1647
+ "loss": 2.0316,
1648
+ "step": 234
1649
+ },
1650
+ {
1651
+ "epoch": 2.702158273381295,
1652
+ "grad_norm": 0.3955429792404175,
1653
+ "learning_rate": 2.3647058823529416e-06,
1654
+ "loss": 1.9407,
1655
+ "step": 235
1656
+ },
1657
+ {
1658
+ "epoch": 2.7136690647482014,
1659
+ "grad_norm": 0.3862835466861725,
1660
+ "learning_rate": 2.3529411764705885e-06,
1661
+ "loss": 1.7705,
1662
+ "step": 236
1663
+ },
1664
+ {
1665
+ "epoch": 2.725179856115108,
1666
+ "grad_norm": 0.42534133791923523,
1667
+ "learning_rate": 2.3411764705882357e-06,
1668
+ "loss": 1.9679,
1669
+ "step": 237
1670
+ },
1671
+ {
1672
+ "epoch": 2.7366906474820145,
1673
+ "grad_norm": 0.42100828886032104,
1674
+ "learning_rate": 2.3294117647058825e-06,
1675
+ "loss": 1.8997,
1676
+ "step": 238
1677
+ },
1678
+ {
1679
+ "epoch": 2.7482014388489207,
1680
+ "grad_norm": 0.4313439726829529,
1681
+ "learning_rate": 2.3176470588235293e-06,
1682
+ "loss": 1.865,
1683
+ "step": 239
1684
+ },
1685
+ {
1686
+ "epoch": 2.7597122302158272,
1687
+ "grad_norm": 0.42461809515953064,
1688
+ "learning_rate": 2.3058823529411766e-06,
1689
+ "loss": 1.8169,
1690
+ "step": 240
1691
+ },
1692
+ {
1693
+ "epoch": 2.771223021582734,
1694
+ "grad_norm": 0.39618465304374695,
1695
+ "learning_rate": 2.2941176470588234e-06,
1696
+ "loss": 1.8453,
1697
+ "step": 241
1698
+ },
1699
+ {
1700
+ "epoch": 2.7827338129496404,
1701
+ "grad_norm": 0.4130411148071289,
1702
+ "learning_rate": 2.2823529411764707e-06,
1703
+ "loss": 1.901,
1704
+ "step": 242
1705
+ },
1706
+ {
1707
+ "epoch": 2.794244604316547,
1708
+ "grad_norm": 0.40345144271850586,
1709
+ "learning_rate": 2.270588235294118e-06,
1710
+ "loss": 1.9913,
1711
+ "step": 243
1712
+ },
1713
+ {
1714
+ "epoch": 2.805755395683453,
1715
+ "grad_norm": 0.4513832926750183,
1716
+ "learning_rate": 2.258823529411765e-06,
1717
+ "loss": 1.9226,
1718
+ "step": 244
1719
+ },
1720
+ {
1721
+ "epoch": 2.8172661870503597,
1722
+ "grad_norm": 0.4180731475353241,
1723
+ "learning_rate": 2.247058823529412e-06,
1724
+ "loss": 1.9364,
1725
+ "step": 245
1726
+ },
1727
+ {
1728
+ "epoch": 2.8287769784172663,
1729
+ "grad_norm": 0.40142112970352173,
1730
+ "learning_rate": 2.2352941176470592e-06,
1731
+ "loss": 2.0484,
1732
+ "step": 246
1733
+ },
1734
+ {
1735
+ "epoch": 2.8402877697841724,
1736
+ "grad_norm": 0.42125147581100464,
1737
+ "learning_rate": 2.223529411764706e-06,
1738
+ "loss": 1.8964,
1739
+ "step": 247
1740
+ },
1741
+ {
1742
+ "epoch": 2.851798561151079,
1743
+ "grad_norm": 0.4866432249546051,
1744
+ "learning_rate": 2.2117647058823533e-06,
1745
+ "loss": 1.8468,
1746
+ "step": 248
1747
+ },
1748
+ {
1749
+ "epoch": 2.8633093525179856,
1750
+ "grad_norm": 0.43598270416259766,
1751
+ "learning_rate": 2.2e-06,
1752
+ "loss": 1.8239,
1753
+ "step": 249
1754
+ },
1755
+ {
1756
+ "epoch": 2.874820143884892,
1757
+ "grad_norm": 0.39764901995658875,
1758
+ "learning_rate": 2.188235294117647e-06,
1759
+ "loss": 1.9577,
1760
+ "step": 250
1761
+ },
1762
+ {
1763
+ "epoch": 2.8863309352517987,
1764
+ "grad_norm": 0.43190810084342957,
1765
+ "learning_rate": 2.176470588235294e-06,
1766
+ "loss": 1.8573,
1767
+ "step": 251
1768
+ },
1769
+ {
1770
+ "epoch": 2.897841726618705,
1771
+ "grad_norm": 0.43591439723968506,
1772
+ "learning_rate": 2.1647058823529414e-06,
1773
+ "loss": 1.9726,
1774
+ "step": 252
1775
+ },
1776
+ {
1777
+ "epoch": 2.9093525179856115,
1778
+ "grad_norm": 0.37023991346359253,
1779
+ "learning_rate": 2.1529411764705887e-06,
1780
+ "loss": 1.9219,
1781
+ "step": 253
1782
+ },
1783
+ {
1784
+ "epoch": 2.920863309352518,
1785
+ "grad_norm": 0.4382263422012329,
1786
+ "learning_rate": 2.1411764705882355e-06,
1787
+ "loss": 2.0037,
1788
+ "step": 254
1789
+ },
1790
+ {
1791
+ "epoch": 2.9323741007194246,
1792
+ "grad_norm": 0.4027315378189087,
1793
+ "learning_rate": 2.1294117647058827e-06,
1794
+ "loss": 1.7579,
1795
+ "step": 255
1796
+ },
1797
+ {
1798
+ "epoch": 2.9438848920863308,
1799
+ "grad_norm": 0.41298389434814453,
1800
+ "learning_rate": 2.1176470588235296e-06,
1801
+ "loss": 1.895,
1802
+ "step": 256
1803
+ },
1804
+ {
1805
+ "epoch": 2.9553956834532373,
1806
+ "grad_norm": 0.39726293087005615,
1807
+ "learning_rate": 2.105882352941177e-06,
1808
+ "loss": 1.9757,
1809
+ "step": 257
1810
+ },
1811
+ {
1812
+ "epoch": 2.966906474820144,
1813
+ "grad_norm": 0.4210617244243622,
1814
+ "learning_rate": 2.0941176470588236e-06,
1815
+ "loss": 1.9105,
1816
+ "step": 258
1817
+ },
1818
+ {
1819
+ "epoch": 2.9784172661870505,
1820
+ "grad_norm": 0.4142284095287323,
1821
+ "learning_rate": 2.0823529411764705e-06,
1822
+ "loss": 1.8864,
1823
+ "step": 259
1824
+ },
1825
+ {
1826
+ "epoch": 2.989928057553957,
1827
+ "grad_norm": 0.3998337686061859,
1828
+ "learning_rate": 2.0705882352941177e-06,
1829
+ "loss": 1.9273,
1830
+ "step": 260
1831
+ },
1832
+ {
1833
+ "epoch": 3.0,
1834
+ "grad_norm": 0.5197608470916748,
1835
+ "learning_rate": 2.058823529411765e-06,
1836
+ "loss": 1.9653,
1837
+ "step": 261
1838
+ },
1839
+ {
1840
+ "epoch": 3.0115107913669066,
1841
+ "grad_norm": 0.41634294390678406,
1842
+ "learning_rate": 2.047058823529412e-06,
1843
+ "loss": 2.0373,
1844
+ "step": 262
1845
+ },
1846
+ {
1847
+ "epoch": 3.023021582733813,
1848
+ "grad_norm": 0.45880216360092163,
1849
+ "learning_rate": 2.035294117647059e-06,
1850
+ "loss": 2.0428,
1851
+ "step": 263
1852
+ },
1853
+ {
1854
+ "epoch": 3.0345323741007193,
1855
+ "grad_norm": 0.4628404974937439,
1856
+ "learning_rate": 2.0235294117647063e-06,
1857
+ "loss": 1.9155,
1858
+ "step": 264
1859
+ },
1860
+ {
1861
+ "epoch": 3.046043165467626,
1862
+ "grad_norm": 0.434393972158432,
1863
+ "learning_rate": 2.011764705882353e-06,
1864
+ "loss": 1.9765,
1865
+ "step": 265
1866
+ },
1867
+ {
1868
+ "epoch": 3.0575539568345325,
1869
+ "grad_norm": 0.3955315053462982,
1870
+ "learning_rate": 2.0000000000000003e-06,
1871
+ "loss": 1.8562,
1872
+ "step": 266
1873
+ },
1874
+ {
1875
+ "epoch": 3.069064748201439,
1876
+ "grad_norm": 0.38154056668281555,
1877
+ "learning_rate": 1.988235294117647e-06,
1878
+ "loss": 1.9882,
1879
+ "step": 267
1880
+ },
1881
+ {
1882
+ "epoch": 3.080575539568345,
1883
+ "grad_norm": 0.39588648080825806,
1884
+ "learning_rate": 1.976470588235294e-06,
1885
+ "loss": 1.966,
1886
+ "step": 268
1887
+ },
1888
+ {
1889
+ "epoch": 3.0920863309352518,
1890
+ "grad_norm": 0.3876536786556244,
1891
+ "learning_rate": 1.9647058823529412e-06,
1892
+ "loss": 1.9866,
1893
+ "step": 269
1894
+ },
1895
+ {
1896
+ "epoch": 3.1035971223021583,
1897
+ "grad_norm": 0.4039998948574066,
1898
+ "learning_rate": 1.9529411764705885e-06,
1899
+ "loss": 1.8997,
1900
+ "step": 270
1901
+ },
1902
+ {
1903
+ "epoch": 3.115107913669065,
1904
+ "grad_norm": 0.4637863039970398,
1905
+ "learning_rate": 1.9411764705882353e-06,
1906
+ "loss": 1.8496,
1907
+ "step": 271
1908
+ },
1909
+ {
1910
+ "epoch": 3.126618705035971,
1911
+ "grad_norm": 0.37885233759880066,
1912
+ "learning_rate": 1.9294117647058825e-06,
1913
+ "loss": 1.9608,
1914
+ "step": 272
1915
+ },
1916
+ {
1917
+ "epoch": 3.1381294964028776,
1918
+ "grad_norm": 0.39651158452033997,
1919
+ "learning_rate": 1.9176470588235298e-06,
1920
+ "loss": 1.9621,
1921
+ "step": 273
1922
+ },
1923
+ {
1924
+ "epoch": 3.149640287769784,
1925
+ "grad_norm": 0.4272053837776184,
1926
+ "learning_rate": 1.9058823529411766e-06,
1927
+ "loss": 2.0647,
1928
+ "step": 274
1929
+ },
1930
+ {
1931
+ "epoch": 3.161151079136691,
1932
+ "grad_norm": 0.36566904187202454,
1933
+ "learning_rate": 1.8941176470588239e-06,
1934
+ "loss": 1.9796,
1935
+ "step": 275
1936
+ },
1937
+ {
1938
+ "epoch": 3.172661870503597,
1939
+ "grad_norm": 0.4320291578769684,
1940
+ "learning_rate": 1.8823529411764707e-06,
1941
+ "loss": 1.8802,
1942
+ "step": 276
1943
+ },
1944
+ {
1945
+ "epoch": 3.1841726618705035,
1946
+ "grad_norm": 0.43051987886428833,
1947
+ "learning_rate": 1.870588235294118e-06,
1948
+ "loss": 1.8472,
1949
+ "step": 277
1950
+ },
1951
+ {
1952
+ "epoch": 3.19568345323741,
1953
+ "grad_norm": 0.40524744987487793,
1954
+ "learning_rate": 1.858823529411765e-06,
1955
+ "loss": 1.978,
1956
+ "step": 278
1957
+ },
1958
+ {
1959
+ "epoch": 3.2071942446043167,
1960
+ "grad_norm": 0.40227210521698,
1961
+ "learning_rate": 1.8470588235294118e-06,
1962
+ "loss": 1.8278,
1963
+ "step": 279
1964
+ },
1965
+ {
1966
+ "epoch": 3.218705035971223,
1967
+ "grad_norm": 0.4132155478000641,
1968
+ "learning_rate": 1.835294117647059e-06,
1969
+ "loss": 1.8649,
1970
+ "step": 280
1971
+ },
1972
+ {
1973
+ "epoch": 3.2302158273381294,
1974
+ "grad_norm": 0.44230976700782776,
1975
+ "learning_rate": 1.8235294117647058e-06,
1976
+ "loss": 2.0472,
1977
+ "step": 281
1978
+ },
1979
+ {
1980
+ "epoch": 3.241726618705036,
1981
+ "grad_norm": 0.41526058316230774,
1982
+ "learning_rate": 1.811764705882353e-06,
1983
+ "loss": 1.9339,
1984
+ "step": 282
1985
+ },
1986
+ {
1987
+ "epoch": 3.2532374100719426,
1988
+ "grad_norm": 0.4362848401069641,
1989
+ "learning_rate": 1.8000000000000001e-06,
1990
+ "loss": 2.0415,
1991
+ "step": 283
1992
+ },
1993
+ {
1994
+ "epoch": 3.2647482014388487,
1995
+ "grad_norm": 0.36819082498550415,
1996
+ "learning_rate": 1.7882352941176474e-06,
1997
+ "loss": 1.9617,
1998
+ "step": 284
1999
+ },
2000
+ {
2001
+ "epoch": 3.2762589928057553,
2002
+ "grad_norm": 0.3629878759384155,
2003
+ "learning_rate": 1.7764705882352942e-06,
2004
+ "loss": 1.9392,
2005
+ "step": 285
2006
+ },
2007
+ {
2008
+ "epoch": 3.287769784172662,
2009
+ "grad_norm": 0.39181169867515564,
2010
+ "learning_rate": 1.7647058823529414e-06,
2011
+ "loss": 1.9113,
2012
+ "step": 286
2013
+ },
2014
+ {
2015
+ "epoch": 3.2992805755395684,
2016
+ "grad_norm": 0.38480955362319946,
2017
+ "learning_rate": 1.7529411764705883e-06,
2018
+ "loss": 1.796,
2019
+ "step": 287
2020
+ },
2021
+ {
2022
+ "epoch": 3.310791366906475,
2023
+ "grad_norm": 0.3901033103466034,
2024
+ "learning_rate": 1.7411764705882353e-06,
2025
+ "loss": 1.8803,
2026
+ "step": 288
2027
+ },
2028
+ {
2029
+ "epoch": 3.322302158273381,
2030
+ "grad_norm": 0.43883225321769714,
2031
+ "learning_rate": 1.7294117647058825e-06,
2032
+ "loss": 1.9635,
2033
+ "step": 289
2034
+ },
2035
+ {
2036
+ "epoch": 3.3338129496402877,
2037
+ "grad_norm": 0.36902785301208496,
2038
+ "learning_rate": 1.7176470588235294e-06,
2039
+ "loss": 1.9661,
2040
+ "step": 290
2041
+ },
2042
+ {
2043
+ "epoch": 3.3453237410071943,
2044
+ "grad_norm": 0.39565619826316833,
2045
+ "learning_rate": 1.7058823529411766e-06,
2046
+ "loss": 2.0416,
2047
+ "step": 291
2048
+ },
2049
+ {
2050
+ "epoch": 3.356834532374101,
2051
+ "grad_norm": 0.4097813367843628,
2052
+ "learning_rate": 1.6941176470588237e-06,
2053
+ "loss": 1.9529,
2054
+ "step": 292
2055
+ },
2056
+ {
2057
+ "epoch": 3.368345323741007,
2058
+ "grad_norm": 0.43599432706832886,
2059
+ "learning_rate": 1.682352941176471e-06,
2060
+ "loss": 2.011,
2061
+ "step": 293
2062
+ },
2063
+ {
2064
+ "epoch": 3.3798561151079136,
2065
+ "grad_norm": 0.3928837180137634,
2066
+ "learning_rate": 1.6705882352941177e-06,
2067
+ "loss": 1.8357,
2068
+ "step": 294
2069
+ },
2070
+ {
2071
+ "epoch": 3.39136690647482,
2072
+ "grad_norm": 0.40635019540786743,
2073
+ "learning_rate": 1.658823529411765e-06,
2074
+ "loss": 1.8737,
2075
+ "step": 295
2076
+ },
2077
+ {
2078
+ "epoch": 3.402877697841727,
2079
+ "grad_norm": 0.43016231060028076,
2080
+ "learning_rate": 1.6470588235294118e-06,
2081
+ "loss": 1.9542,
2082
+ "step": 296
2083
+ },
2084
+ {
2085
+ "epoch": 3.414388489208633,
2086
+ "grad_norm": 0.478292852640152,
2087
+ "learning_rate": 1.635294117647059e-06,
2088
+ "loss": 1.8912,
2089
+ "step": 297
2090
+ },
2091
+ {
2092
+ "epoch": 3.4258992805755395,
2093
+ "grad_norm": 0.40000081062316895,
2094
+ "learning_rate": 1.623529411764706e-06,
2095
+ "loss": 1.9357,
2096
+ "step": 298
2097
+ },
2098
+ {
2099
+ "epoch": 3.437410071942446,
2100
+ "grad_norm": 0.4033874273300171,
2101
+ "learning_rate": 1.6117647058823529e-06,
2102
+ "loss": 1.906,
2103
+ "step": 299
2104
+ },
2105
+ {
2106
+ "epoch": 3.4489208633093527,
2107
+ "grad_norm": 0.40947261452674866,
2108
+ "learning_rate": 1.6000000000000001e-06,
2109
+ "loss": 1.8173,
2110
+ "step": 300
2111
+ },
2112
+ {
2113
+ "epoch": 3.460431654676259,
2114
+ "grad_norm": 0.4071550965309143,
2115
+ "learning_rate": 1.5882352941176472e-06,
2116
+ "loss": 1.9455,
2117
+ "step": 301
2118
+ },
2119
+ {
2120
+ "epoch": 3.4719424460431654,
2121
+ "grad_norm": 0.4430578052997589,
2122
+ "learning_rate": 1.5764705882352942e-06,
2123
+ "loss": 1.8767,
2124
+ "step": 302
2125
+ },
2126
+ {
2127
+ "epoch": 3.483453237410072,
2128
+ "grad_norm": 0.41929903626441956,
2129
+ "learning_rate": 1.5647058823529412e-06,
2130
+ "loss": 1.9888,
2131
+ "step": 303
2132
+ },
2133
+ {
2134
+ "epoch": 3.4949640287769785,
2135
+ "grad_norm": 0.3640955090522766,
2136
+ "learning_rate": 1.5529411764705885e-06,
2137
+ "loss": 1.8597,
2138
+ "step": 304
2139
+ },
2140
+ {
2141
+ "epoch": 3.506474820143885,
2142
+ "grad_norm": 0.4024275839328766,
2143
+ "learning_rate": 1.5411764705882353e-06,
2144
+ "loss": 1.8771,
2145
+ "step": 305
2146
+ },
2147
+ {
2148
+ "epoch": 3.5179856115107913,
2149
+ "grad_norm": 0.38399186730384827,
2150
+ "learning_rate": 1.5294117647058826e-06,
2151
+ "loss": 1.9956,
2152
+ "step": 306
2153
+ },
2154
+ {
2155
+ "epoch": 3.529496402877698,
2156
+ "grad_norm": 0.39859694242477417,
2157
+ "learning_rate": 1.5176470588235296e-06,
2158
+ "loss": 1.9475,
2159
+ "step": 307
2160
+ },
2161
+ {
2162
+ "epoch": 3.5410071942446044,
2163
+ "grad_norm": 0.4364980459213257,
2164
+ "learning_rate": 1.5058823529411764e-06,
2165
+ "loss": 1.9514,
2166
+ "step": 308
2167
+ },
2168
+ {
2169
+ "epoch": 3.5525179856115106,
2170
+ "grad_norm": 0.4755602180957794,
2171
+ "learning_rate": 1.4941176470588237e-06,
2172
+ "loss": 1.9057,
2173
+ "step": 309
2174
+ },
2175
+ {
2176
+ "epoch": 3.564028776978417,
2177
+ "grad_norm": 0.39024367928504944,
2178
+ "learning_rate": 1.4823529411764707e-06,
2179
+ "loss": 1.9041,
2180
+ "step": 310
2181
+ },
2182
+ {
2183
+ "epoch": 3.5755395683453237,
2184
+ "grad_norm": 0.4508678615093231,
2185
+ "learning_rate": 1.4705882352941177e-06,
2186
+ "loss": 2.0103,
2187
+ "step": 311
2188
+ },
2189
+ {
2190
+ "epoch": 3.5870503597122303,
2191
+ "grad_norm": 0.4054012894630432,
2192
+ "learning_rate": 1.4588235294117648e-06,
2193
+ "loss": 2.0046,
2194
+ "step": 312
2195
+ },
2196
+ {
2197
+ "epoch": 3.598561151079137,
2198
+ "grad_norm": 0.40608781576156616,
2199
+ "learning_rate": 1.447058823529412e-06,
2200
+ "loss": 1.9209,
2201
+ "step": 313
2202
+ },
2203
+ {
2204
+ "epoch": 3.610071942446043,
2205
+ "grad_norm": 0.4027564525604248,
2206
+ "learning_rate": 1.4352941176470588e-06,
2207
+ "loss": 1.9453,
2208
+ "step": 314
2209
+ },
2210
+ {
2211
+ "epoch": 3.6215827338129496,
2212
+ "grad_norm": 0.37097108364105225,
2213
+ "learning_rate": 1.423529411764706e-06,
2214
+ "loss": 1.8778,
2215
+ "step": 315
2216
+ },
2217
+ {
2218
+ "epoch": 3.633093525179856,
2219
+ "grad_norm": 0.3786408007144928,
2220
+ "learning_rate": 1.4117647058823531e-06,
2221
+ "loss": 2.0226,
2222
+ "step": 316
2223
+ },
2224
+ {
2225
+ "epoch": 3.6446043165467623,
2226
+ "grad_norm": 0.5143160223960876,
2227
+ "learning_rate": 1.4000000000000001e-06,
2228
+ "loss": 1.8967,
2229
+ "step": 317
2230
+ },
2231
+ {
2232
+ "epoch": 3.656115107913669,
2233
+ "grad_norm": 0.4117061197757721,
2234
+ "learning_rate": 1.3882352941176472e-06,
2235
+ "loss": 1.9639,
2236
+ "step": 318
2237
+ },
2238
+ {
2239
+ "epoch": 3.6676258992805755,
2240
+ "grad_norm": 0.38462352752685547,
2241
+ "learning_rate": 1.3764705882352942e-06,
2242
+ "loss": 1.9823,
2243
+ "step": 319
2244
+ },
2245
+ {
2246
+ "epoch": 3.679136690647482,
2247
+ "grad_norm": 0.4092719852924347,
2248
+ "learning_rate": 1.3647058823529413e-06,
2249
+ "loss": 1.9542,
2250
+ "step": 320
2251
+ },
2252
+ {
2253
+ "epoch": 3.6906474820143886,
2254
+ "grad_norm": 0.3653268814086914,
2255
+ "learning_rate": 1.3529411764705883e-06,
2256
+ "loss": 1.9841,
2257
+ "step": 321
2258
+ },
2259
+ {
2260
+ "epoch": 3.702158273381295,
2261
+ "grad_norm": 0.46952006220817566,
2262
+ "learning_rate": 1.3411764705882355e-06,
2263
+ "loss": 1.8227,
2264
+ "step": 322
2265
+ },
2266
+ {
2267
+ "epoch": 3.7136690647482014,
2268
+ "grad_norm": 0.3843960464000702,
2269
+ "learning_rate": 1.3294117647058824e-06,
2270
+ "loss": 1.9256,
2271
+ "step": 323
2272
+ },
2273
+ {
2274
+ "epoch": 3.725179856115108,
2275
+ "grad_norm": 0.4532316327095032,
2276
+ "learning_rate": 1.3176470588235296e-06,
2277
+ "loss": 1.8819,
2278
+ "step": 324
2279
+ },
2280
+ {
2281
+ "epoch": 3.7366906474820145,
2282
+ "grad_norm": 0.4273243248462677,
2283
+ "learning_rate": 1.3058823529411766e-06,
2284
+ "loss": 1.7333,
2285
+ "step": 325
2286
+ },
2287
+ {
2288
+ "epoch": 3.7482014388489207,
2289
+ "grad_norm": 0.3750397264957428,
2290
+ "learning_rate": 1.2941176470588237e-06,
2291
+ "loss": 1.8715,
2292
+ "step": 326
2293
+ },
2294
+ {
2295
+ "epoch": 3.7597122302158272,
2296
+ "grad_norm": 0.37180638313293457,
2297
+ "learning_rate": 1.2823529411764707e-06,
2298
+ "loss": 1.8691,
2299
+ "step": 327
2300
+ },
2301
+ {
2302
+ "epoch": 3.771223021582734,
2303
+ "grad_norm": 0.45777222514152527,
2304
+ "learning_rate": 1.2705882352941175e-06,
2305
+ "loss": 2.0797,
2306
+ "step": 328
2307
+ },
2308
+ {
2309
+ "epoch": 3.7827338129496404,
2310
+ "grad_norm": 0.40992122888565063,
2311
+ "learning_rate": 1.2588235294117648e-06,
2312
+ "loss": 1.8158,
2313
+ "step": 329
2314
+ },
2315
+ {
2316
+ "epoch": 3.794244604316547,
2317
+ "grad_norm": 0.39492303133010864,
2318
+ "learning_rate": 1.2470588235294118e-06,
2319
+ "loss": 1.9402,
2320
+ "step": 330
2321
+ },
2322
+ {
2323
+ "epoch": 3.805755395683453,
2324
+ "grad_norm": 0.4207254946231842,
2325
+ "learning_rate": 1.235294117647059e-06,
2326
+ "loss": 1.9019,
2327
+ "step": 331
2328
+ },
2329
+ {
2330
+ "epoch": 3.8172661870503597,
2331
+ "grad_norm": 0.41245564818382263,
2332
+ "learning_rate": 1.223529411764706e-06,
2333
+ "loss": 1.9962,
2334
+ "step": 332
2335
+ },
2336
+ {
2337
+ "epoch": 3.8287769784172663,
2338
+ "grad_norm": 0.4352625906467438,
2339
+ "learning_rate": 1.211764705882353e-06,
2340
+ "loss": 1.9475,
2341
+ "step": 333
2342
+ },
2343
+ {
2344
+ "epoch": 3.8402877697841724,
2345
+ "grad_norm": 0.44661152362823486,
2346
+ "learning_rate": 1.2000000000000002e-06,
2347
+ "loss": 1.9815,
2348
+ "step": 334
2349
+ },
2350
+ {
2351
+ "epoch": 3.851798561151079,
2352
+ "grad_norm": 0.3469794988632202,
2353
+ "learning_rate": 1.1882352941176472e-06,
2354
+ "loss": 1.9558,
2355
+ "step": 335
2356
+ },
2357
+ {
2358
+ "epoch": 3.8633093525179856,
2359
+ "grad_norm": 0.3915042281150818,
2360
+ "learning_rate": 1.1764705882352942e-06,
2361
+ "loss": 1.9432,
2362
+ "step": 336
2363
+ },
2364
+ {
2365
+ "epoch": 3.874820143884892,
2366
+ "grad_norm": 0.40242835879325867,
2367
+ "learning_rate": 1.1647058823529413e-06,
2368
+ "loss": 2.0347,
2369
+ "step": 337
2370
+ },
2371
+ {
2372
+ "epoch": 3.8863309352517987,
2373
+ "grad_norm": 0.4240435063838959,
2374
+ "learning_rate": 1.1529411764705883e-06,
2375
+ "loss": 1.8546,
2376
+ "step": 338
2377
+ },
2378
+ {
2379
+ "epoch": 3.897841726618705,
2380
+ "grad_norm": 0.3874111771583557,
2381
+ "learning_rate": 1.1411764705882353e-06,
2382
+ "loss": 1.8743,
2383
+ "step": 339
2384
+ },
2385
+ {
2386
+ "epoch": 3.9093525179856115,
2387
+ "grad_norm": 0.4088236391544342,
2388
+ "learning_rate": 1.1294117647058826e-06,
2389
+ "loss": 1.9511,
2390
+ "step": 340
2391
+ },
2392
+ {
2393
+ "epoch": 3.920863309352518,
2394
+ "grad_norm": 0.4264145493507385,
2395
+ "learning_rate": 1.1176470588235296e-06,
2396
+ "loss": 1.8457,
2397
+ "step": 341
2398
+ },
2399
+ {
2400
+ "epoch": 3.9323741007194246,
2401
+ "grad_norm": 0.4625447392463684,
2402
+ "learning_rate": 1.1058823529411766e-06,
2403
+ "loss": 2.0004,
2404
+ "step": 342
2405
+ },
2406
+ {
2407
+ "epoch": 3.9438848920863308,
2408
+ "grad_norm": 0.3717285692691803,
2409
+ "learning_rate": 1.0941176470588235e-06,
2410
+ "loss": 2.018,
2411
+ "step": 343
2412
+ },
2413
+ {
2414
+ "epoch": 3.9553956834532373,
2415
+ "grad_norm": 0.38859859108924866,
2416
+ "learning_rate": 1.0823529411764707e-06,
2417
+ "loss": 2.0653,
2418
+ "step": 344
2419
+ },
2420
+ {
2421
+ "epoch": 3.966906474820144,
2422
+ "grad_norm": 0.4414234161376953,
2423
+ "learning_rate": 1.0705882352941177e-06,
2424
+ "loss": 1.9468,
2425
+ "step": 345
2426
+ },
2427
+ {
2428
+ "epoch": 3.9784172661870505,
2429
+ "grad_norm": 0.44937196373939514,
2430
+ "learning_rate": 1.0588235294117648e-06,
2431
+ "loss": 1.9817,
2432
+ "step": 346
2433
+ },
2434
+ {
2435
+ "epoch": 3.989928057553957,
2436
+ "grad_norm": 0.4258635938167572,
2437
+ "learning_rate": 1.0470588235294118e-06,
2438
+ "loss": 1.8073,
2439
+ "step": 347
2440
+ },
2441
+ {
2442
+ "epoch": 4.0,
2443
+ "grad_norm": 0.43822839856147766,
2444
+ "learning_rate": 1.0352941176470589e-06,
2445
+ "loss": 1.8991,
2446
+ "step": 348
2447
+ },
2448
+ {
2449
+ "epoch": 4.011510791366907,
2450
+ "grad_norm": 0.46056368947029114,
2451
+ "learning_rate": 1.023529411764706e-06,
2452
+ "loss": 1.9822,
2453
+ "step": 349
2454
+ },
2455
+ {
2456
+ "epoch": 4.023021582733813,
2457
+ "grad_norm": 0.37518858909606934,
2458
+ "learning_rate": 1.0117647058823531e-06,
2459
+ "loss": 1.813,
2460
+ "step": 350
2461
+ },
2462
+ {
2463
+ "epoch": 4.03453237410072,
2464
+ "grad_norm": 0.3921293020248413,
2465
+ "learning_rate": 1.0000000000000002e-06,
2466
+ "loss": 2.0461,
2467
+ "step": 351
2468
+ },
2469
+ {
2470
+ "epoch": 4.046043165467626,
2471
+ "grad_norm": 0.4025590419769287,
2472
+ "learning_rate": 9.88235294117647e-07,
2473
+ "loss": 1.9639,
2474
+ "step": 352
2475
+ },
2476
+ {
2477
+ "epoch": 4.057553956834532,
2478
+ "grad_norm": 0.40453991293907166,
2479
+ "learning_rate": 9.764705882352942e-07,
2480
+ "loss": 1.9503,
2481
+ "step": 353
2482
+ },
2483
+ {
2484
+ "epoch": 4.069064748201439,
2485
+ "grad_norm": 0.35269466042518616,
2486
+ "learning_rate": 9.647058823529413e-07,
2487
+ "loss": 1.9183,
2488
+ "step": 354
2489
+ },
2490
+ {
2491
+ "epoch": 4.080575539568345,
2492
+ "grad_norm": 0.4840276837348938,
2493
+ "learning_rate": 9.529411764705883e-07,
2494
+ "loss": 1.8452,
2495
+ "step": 355
2496
+ },
2497
+ {
2498
+ "epoch": 4.092086330935252,
2499
+ "grad_norm": 0.39707890152931213,
2500
+ "learning_rate": 9.411764705882353e-07,
2501
+ "loss": 1.9937,
2502
+ "step": 356
2503
+ },
2504
+ {
2505
+ "epoch": 4.103597122302158,
2506
+ "grad_norm": 0.4407122731208801,
2507
+ "learning_rate": 9.294117647058825e-07,
2508
+ "loss": 2.022,
2509
+ "step": 357
2510
+ },
2511
+ {
2512
+ "epoch": 4.115107913669065,
2513
+ "grad_norm": 0.41454342007637024,
2514
+ "learning_rate": 9.176470588235295e-07,
2515
+ "loss": 1.8907,
2516
+ "step": 358
2517
+ },
2518
+ {
2519
+ "epoch": 4.1266187050359715,
2520
+ "grad_norm": 0.45531314611434937,
2521
+ "learning_rate": 9.058823529411765e-07,
2522
+ "loss": 1.8662,
2523
+ "step": 359
2524
+ },
2525
+ {
2526
+ "epoch": 4.138129496402878,
2527
+ "grad_norm": 0.4606649577617645,
2528
+ "learning_rate": 8.941176470588237e-07,
2529
+ "loss": 1.8686,
2530
+ "step": 360
2531
+ },
2532
+ {
2533
+ "epoch": 4.149640287769784,
2534
+ "grad_norm": 0.3887675106525421,
2535
+ "learning_rate": 8.823529411764707e-07,
2536
+ "loss": 1.9282,
2537
+ "step": 361
2538
+ },
2539
+ {
2540
+ "epoch": 4.16115107913669,
2541
+ "grad_norm": 0.40791404247283936,
2542
+ "learning_rate": 8.705882352941177e-07,
2543
+ "loss": 1.8935,
2544
+ "step": 362
2545
+ },
2546
+ {
2547
+ "epoch": 4.172661870503597,
2548
+ "grad_norm": 0.4134286046028137,
2549
+ "learning_rate": 8.588235294117647e-07,
2550
+ "loss": 1.9222,
2551
+ "step": 363
2552
+ },
2553
+ {
2554
+ "epoch": 4.1841726618705035,
2555
+ "grad_norm": 0.45545920729637146,
2556
+ "learning_rate": 8.470588235294118e-07,
2557
+ "loss": 1.7464,
2558
+ "step": 364
2559
+ },
2560
+ {
2561
+ "epoch": 4.19568345323741,
2562
+ "grad_norm": 0.3929649889469147,
2563
+ "learning_rate": 8.352941176470589e-07,
2564
+ "loss": 1.9524,
2565
+ "step": 365
2566
+ },
2567
+ {
2568
+ "epoch": 4.207194244604317,
2569
+ "grad_norm": 0.3917909860610962,
2570
+ "learning_rate": 8.235294117647059e-07,
2571
+ "loss": 1.8608,
2572
+ "step": 366
2573
+ },
2574
+ {
2575
+ "epoch": 4.218705035971223,
2576
+ "grad_norm": 0.3615923821926117,
2577
+ "learning_rate": 8.11764705882353e-07,
2578
+ "loss": 2.0196,
2579
+ "step": 367
2580
+ },
2581
+ {
2582
+ "epoch": 4.23021582733813,
2583
+ "grad_norm": 0.42296525835990906,
2584
+ "learning_rate": 8.000000000000001e-07,
2585
+ "loss": 2.0074,
2586
+ "step": 368
2587
+ },
2588
+ {
2589
+ "epoch": 4.2417266187050355,
2590
+ "grad_norm": 0.38392674922943115,
2591
+ "learning_rate": 7.882352941176471e-07,
2592
+ "loss": 1.8838,
2593
+ "step": 369
2594
+ },
2595
+ {
2596
+ "epoch": 4.253237410071942,
2597
+ "grad_norm": 0.45736461877822876,
2598
+ "learning_rate": 7.764705882352942e-07,
2599
+ "loss": 1.9283,
2600
+ "step": 370
2601
+ },
2602
+ {
2603
+ "epoch": 4.264748201438849,
2604
+ "grad_norm": 0.4002780020236969,
2605
+ "learning_rate": 7.647058823529413e-07,
2606
+ "loss": 1.9358,
2607
+ "step": 371
2608
+ },
2609
+ {
2610
+ "epoch": 4.276258992805755,
2611
+ "grad_norm": 0.4492432475090027,
2612
+ "learning_rate": 7.529411764705882e-07,
2613
+ "loss": 1.9168,
2614
+ "step": 372
2615
+ },
2616
+ {
2617
+ "epoch": 4.287769784172662,
2618
+ "grad_norm": 0.4120420813560486,
2619
+ "learning_rate": 7.411764705882353e-07,
2620
+ "loss": 2.0267,
2621
+ "step": 373
2622
+ },
2623
+ {
2624
+ "epoch": 4.299280575539568,
2625
+ "grad_norm": 0.37060075998306274,
2626
+ "learning_rate": 7.294117647058824e-07,
2627
+ "loss": 1.7903,
2628
+ "step": 374
2629
+ },
2630
+ {
2631
+ "epoch": 4.310791366906475,
2632
+ "grad_norm": 0.41300690174102783,
2633
+ "learning_rate": 7.176470588235294e-07,
2634
+ "loss": 1.9742,
2635
+ "step": 375
2636
+ },
2637
+ {
2638
+ "epoch": 4.322302158273382,
2639
+ "grad_norm": 0.39079055190086365,
2640
+ "learning_rate": 7.058823529411766e-07,
2641
+ "loss": 1.9725,
2642
+ "step": 376
2643
+ },
2644
+ {
2645
+ "epoch": 4.333812949640288,
2646
+ "grad_norm": 0.37425902485847473,
2647
+ "learning_rate": 6.941176470588236e-07,
2648
+ "loss": 1.9327,
2649
+ "step": 377
2650
+ },
2651
+ {
2652
+ "epoch": 4.345323741007194,
2653
+ "grad_norm": 0.39660215377807617,
2654
+ "learning_rate": 6.823529411764706e-07,
2655
+ "loss": 1.9264,
2656
+ "step": 378
2657
+ },
2658
+ {
2659
+ "epoch": 4.3568345323741005,
2660
+ "grad_norm": 0.3530445098876953,
2661
+ "learning_rate": 6.705882352941178e-07,
2662
+ "loss": 2.0301,
2663
+ "step": 379
2664
+ },
2665
+ {
2666
+ "epoch": 4.368345323741007,
2667
+ "grad_norm": 0.3451977074146271,
2668
+ "learning_rate": 6.588235294117648e-07,
2669
+ "loss": 1.9497,
2670
+ "step": 380
2671
+ },
2672
+ {
2673
+ "epoch": 4.379856115107914,
2674
+ "grad_norm": 0.39704129099845886,
2675
+ "learning_rate": 6.470588235294118e-07,
2676
+ "loss": 2.0197,
2677
+ "step": 381
2678
+ },
2679
+ {
2680
+ "epoch": 4.39136690647482,
2681
+ "grad_norm": 0.4510452151298523,
2682
+ "learning_rate": 6.352941176470588e-07,
2683
+ "loss": 2.0232,
2684
+ "step": 382
2685
+ },
2686
+ {
2687
+ "epoch": 4.402877697841727,
2688
+ "grad_norm": 0.40996772050857544,
2689
+ "learning_rate": 6.235294117647059e-07,
2690
+ "loss": 1.9544,
2691
+ "step": 383
2692
+ },
2693
+ {
2694
+ "epoch": 4.414388489208633,
2695
+ "grad_norm": 0.39487242698669434,
2696
+ "learning_rate": 6.11764705882353e-07,
2697
+ "loss": 1.906,
2698
+ "step": 384
2699
+ },
2700
+ {
2701
+ "epoch": 4.42589928057554,
2702
+ "grad_norm": 0.38208380341529846,
2703
+ "learning_rate": 6.000000000000001e-07,
2704
+ "loss": 1.9622,
2705
+ "step": 385
2706
+ },
2707
+ {
2708
+ "epoch": 4.437410071942446,
2709
+ "grad_norm": 0.4923550486564636,
2710
+ "learning_rate": 5.882352941176471e-07,
2711
+ "loss": 1.8734,
2712
+ "step": 386
2713
+ },
2714
+ {
2715
+ "epoch": 4.448920863309352,
2716
+ "grad_norm": 0.3934495151042938,
2717
+ "learning_rate": 5.764705882352941e-07,
2718
+ "loss": 1.9016,
2719
+ "step": 387
2720
+ },
2721
+ {
2722
+ "epoch": 4.460431654676259,
2723
+ "grad_norm": 0.3597968518733978,
2724
+ "learning_rate": 5.647058823529413e-07,
2725
+ "loss": 1.8471,
2726
+ "step": 388
2727
+ },
2728
+ {
2729
+ "epoch": 4.471942446043165,
2730
+ "grad_norm": 0.3755582273006439,
2731
+ "learning_rate": 5.529411764705883e-07,
2732
+ "loss": 1.9416,
2733
+ "step": 389
2734
+ },
2735
+ {
2736
+ "epoch": 4.483453237410072,
2737
+ "grad_norm": 0.41233447194099426,
2738
+ "learning_rate": 5.411764705882354e-07,
2739
+ "loss": 1.8175,
2740
+ "step": 390
2741
+ },
2742
+ {
2743
+ "epoch": 4.4949640287769785,
2744
+ "grad_norm": 0.43287456035614014,
2745
+ "learning_rate": 5.294117647058824e-07,
2746
+ "loss": 2.0551,
2747
+ "step": 391
2748
+ },
2749
+ {
2750
+ "epoch": 4.506474820143885,
2751
+ "grad_norm": 0.3722149133682251,
2752
+ "learning_rate": 5.176470588235294e-07,
2753
+ "loss": 1.8064,
2754
+ "step": 392
2755
+ },
2756
+ {
2757
+ "epoch": 4.517985611510792,
2758
+ "grad_norm": 0.36640405654907227,
2759
+ "learning_rate": 5.058823529411766e-07,
2760
+ "loss": 1.8697,
2761
+ "step": 393
2762
+ },
2763
+ {
2764
+ "epoch": 4.529496402877697,
2765
+ "grad_norm": 0.42729923129081726,
2766
+ "learning_rate": 4.941176470588235e-07,
2767
+ "loss": 1.9544,
2768
+ "step": 394
2769
+ },
2770
+ {
2771
+ "epoch": 4.541007194244604,
2772
+ "grad_norm": 0.4456847012042999,
2773
+ "learning_rate": 4.823529411764706e-07,
2774
+ "loss": 1.9493,
2775
+ "step": 395
2776
+ },
2777
+ {
2778
+ "epoch": 4.5525179856115106,
2779
+ "grad_norm": 0.37357428669929504,
2780
+ "learning_rate": 4.7058823529411767e-07,
2781
+ "loss": 1.8166,
2782
+ "step": 396
2783
+ },
2784
+ {
2785
+ "epoch": 4.564028776978417,
2786
+ "grad_norm": 0.36978697776794434,
2787
+ "learning_rate": 4.5882352941176476e-07,
2788
+ "loss": 1.9415,
2789
+ "step": 397
2790
+ },
2791
+ {
2792
+ "epoch": 4.575539568345324,
2793
+ "grad_norm": 0.4100985825061798,
2794
+ "learning_rate": 4.4705882352941184e-07,
2795
+ "loss": 1.9396,
2796
+ "step": 398
2797
+ },
2798
+ {
2799
+ "epoch": 4.58705035971223,
2800
+ "grad_norm": 0.4530591368675232,
2801
+ "learning_rate": 4.352941176470588e-07,
2802
+ "loss": 2.0031,
2803
+ "step": 399
2804
+ },
2805
+ {
2806
+ "epoch": 4.598561151079137,
2807
+ "grad_norm": 0.43112170696258545,
2808
+ "learning_rate": 4.235294117647059e-07,
2809
+ "loss": 1.8297,
2810
+ "step": 400
2811
+ },
2812
+ {
2813
+ "epoch": 4.6100719424460435,
2814
+ "grad_norm": 0.4256054162979126,
2815
+ "learning_rate": 4.1176470588235295e-07,
2816
+ "loss": 1.9466,
2817
+ "step": 401
2818
+ },
2819
+ {
2820
+ "epoch": 4.62158273381295,
2821
+ "grad_norm": 0.41112977266311646,
2822
+ "learning_rate": 4.0000000000000003e-07,
2823
+ "loss": 1.9521,
2824
+ "step": 402
2825
+ },
2826
+ {
2827
+ "epoch": 4.633093525179856,
2828
+ "grad_norm": 0.4099273681640625,
2829
+ "learning_rate": 3.882352941176471e-07,
2830
+ "loss": 1.8238,
2831
+ "step": 403
2832
+ },
2833
+ {
2834
+ "epoch": 4.644604316546762,
2835
+ "grad_norm": 0.3838717043399811,
2836
+ "learning_rate": 3.764705882352941e-07,
2837
+ "loss": 2.0297,
2838
+ "step": 404
2839
+ },
2840
+ {
2841
+ "epoch": 4.656115107913669,
2842
+ "grad_norm": 0.40768367052078247,
2843
+ "learning_rate": 3.647058823529412e-07,
2844
+ "loss": 1.763,
2845
+ "step": 405
2846
+ },
2847
+ {
2848
+ "epoch": 4.6676258992805755,
2849
+ "grad_norm": 0.4218824803829193,
2850
+ "learning_rate": 3.529411764705883e-07,
2851
+ "loss": 1.9831,
2852
+ "step": 406
2853
+ },
2854
+ {
2855
+ "epoch": 4.679136690647482,
2856
+ "grad_norm": 0.4038139581680298,
2857
+ "learning_rate": 3.411764705882353e-07,
2858
+ "loss": 1.9038,
2859
+ "step": 407
2860
+ },
2861
+ {
2862
+ "epoch": 4.690647482014389,
2863
+ "grad_norm": 0.444604754447937,
2864
+ "learning_rate": 3.294117647058824e-07,
2865
+ "loss": 1.8462,
2866
+ "step": 408
2867
+ },
2868
+ {
2869
+ "epoch": 4.702158273381295,
2870
+ "grad_norm": 0.398887038230896,
2871
+ "learning_rate": 3.176470588235294e-07,
2872
+ "loss": 1.8827,
2873
+ "step": 409
2874
+ },
2875
+ {
2876
+ "epoch": 4.713669064748202,
2877
+ "grad_norm": 0.3743375837802887,
2878
+ "learning_rate": 3.058823529411765e-07,
2879
+ "loss": 1.9547,
2880
+ "step": 410
2881
+ },
2882
+ {
2883
+ "epoch": 4.725179856115108,
2884
+ "grad_norm": 0.36487120389938354,
2885
+ "learning_rate": 2.9411764705882356e-07,
2886
+ "loss": 1.9927,
2887
+ "step": 411
2888
+ },
2889
+ {
2890
+ "epoch": 4.736690647482014,
2891
+ "grad_norm": 0.38175249099731445,
2892
+ "learning_rate": 2.8235294117647064e-07,
2893
+ "loss": 1.8559,
2894
+ "step": 412
2895
+ },
2896
+ {
2897
+ "epoch": 4.748201438848921,
2898
+ "grad_norm": 0.45353925228118896,
2899
+ "learning_rate": 2.705882352941177e-07,
2900
+ "loss": 1.8788,
2901
+ "step": 413
2902
+ },
2903
+ {
2904
+ "epoch": 4.759712230215827,
2905
+ "grad_norm": 0.38203585147857666,
2906
+ "learning_rate": 2.588235294117647e-07,
2907
+ "loss": 1.9798,
2908
+ "step": 414
2909
+ },
2910
+ {
2911
+ "epoch": 4.771223021582734,
2912
+ "grad_norm": 0.4031854271888733,
2913
+ "learning_rate": 2.4705882352941175e-07,
2914
+ "loss": 1.924,
2915
+ "step": 415
2916
+ },
2917
+ {
2918
+ "epoch": 4.78273381294964,
2919
+ "grad_norm": 0.39555788040161133,
2920
+ "learning_rate": 2.3529411764705883e-07,
2921
+ "loss": 1.9385,
2922
+ "step": 416
2923
+ },
2924
+ {
2925
+ "epoch": 4.794244604316547,
2926
+ "grad_norm": 0.36034807562828064,
2927
+ "learning_rate": 2.2352941176470592e-07,
2928
+ "loss": 1.8572,
2929
+ "step": 417
2930
+ },
2931
+ {
2932
+ "epoch": 4.805755395683454,
2933
+ "grad_norm": 0.416274756193161,
2934
+ "learning_rate": 2.1176470588235296e-07,
2935
+ "loss": 1.9249,
2936
+ "step": 418
2937
+ },
2938
+ {
2939
+ "epoch": 4.817266187050359,
2940
+ "grad_norm": 0.40581023693084717,
2941
+ "learning_rate": 2.0000000000000002e-07,
2942
+ "loss": 2.0491,
2943
+ "step": 419
2944
+ },
2945
+ {
2946
+ "epoch": 4.828776978417266,
2947
+ "grad_norm": 0.433010995388031,
2948
+ "learning_rate": 1.8823529411764705e-07,
2949
+ "loss": 1.9709,
2950
+ "step": 420
2951
+ },
2952
+ {
2953
+ "epoch": 4.840287769784172,
2954
+ "grad_norm": 0.37076178193092346,
2955
+ "learning_rate": 1.7647058823529414e-07,
2956
+ "loss": 1.8053,
2957
+ "step": 421
2958
+ },
2959
+ {
2960
+ "epoch": 4.851798561151079,
2961
+ "grad_norm": 0.4253133237361908,
2962
+ "learning_rate": 1.647058823529412e-07,
2963
+ "loss": 1.8407,
2964
+ "step": 422
2965
+ },
2966
+ {
2967
+ "epoch": 4.863309352517986,
2968
+ "grad_norm": 0.3607207238674164,
2969
+ "learning_rate": 1.5294117647058826e-07,
2970
+ "loss": 1.9195,
2971
+ "step": 423
2972
+ },
2973
+ {
2974
+ "epoch": 4.874820143884892,
2975
+ "grad_norm": 0.39863321185112,
2976
+ "learning_rate": 1.4117647058823532e-07,
2977
+ "loss": 1.9297,
2978
+ "step": 424
2979
+ },
2980
+ {
2981
+ "epoch": 4.886330935251799,
2982
+ "grad_norm": 0.36778998374938965,
2983
+ "learning_rate": 1.2941176470588236e-07,
2984
+ "loss": 1.9024,
2985
+ "step": 425
2986
+ },
2987
+ {
2988
+ "epoch": 4.897841726618705,
2989
+ "grad_norm": 0.4625915288925171,
2990
+ "learning_rate": 1.1764705882352942e-07,
2991
+ "loss": 1.969,
2992
+ "step": 426
2993
+ },
2994
+ {
2995
+ "epoch": 4.909352517985612,
2996
+ "grad_norm": 0.39172741770744324,
2997
+ "learning_rate": 1.0588235294117648e-07,
2998
+ "loss": 1.9035,
2999
+ "step": 427
3000
+ },
3001
+ {
3002
+ "epoch": 4.920863309352518,
3003
+ "grad_norm": 0.38108983635902405,
3004
+ "learning_rate": 9.411764705882353e-08,
3005
+ "loss": 1.921,
3006
+ "step": 428
3007
+ },
3008
+ {
3009
+ "epoch": 4.932374100719424,
3010
+ "grad_norm": 0.4520784020423889,
3011
+ "learning_rate": 8.23529411764706e-08,
3012
+ "loss": 2.0255,
3013
+ "step": 429
3014
+ },
3015
+ {
3016
+ "epoch": 4.943884892086331,
3017
+ "grad_norm": 0.4058828055858612,
3018
+ "learning_rate": 7.058823529411766e-08,
3019
+ "loss": 1.9021,
3020
+ "step": 430
3021
+ },
3022
+ {
3023
+ "epoch": 4.955395683453237,
3024
+ "grad_norm": 0.38254988193511963,
3025
+ "learning_rate": 5.882352941176471e-08,
3026
+ "loss": 1.865,
3027
+ "step": 431
3028
+ },
3029
+ {
3030
+ "epoch": 4.966906474820144,
3031
+ "grad_norm": 0.4428682327270508,
3032
+ "learning_rate": 4.705882352941176e-08,
3033
+ "loss": 1.9649,
3034
+ "step": 432
3035
+ },
3036
+ {
3037
+ "epoch": 4.9784172661870505,
3038
+ "grad_norm": 0.41550108790397644,
3039
+ "learning_rate": 3.529411764705883e-08,
3040
+ "loss": 1.9868,
3041
+ "step": 433
3042
+ },
3043
+ {
3044
+ "epoch": 4.989928057553957,
3045
+ "grad_norm": 0.430279016494751,
3046
+ "learning_rate": 2.352941176470588e-08,
3047
+ "loss": 1.913,
3048
+ "step": 434
3049
+ },
3050
+ {
3051
+ "epoch": 5.0,
3052
+ "grad_norm": 0.4352044463157654,
3053
+ "learning_rate": 1.176470588235294e-08,
3054
+ "loss": 1.9378,
3055
+ "step": 435
3056
+ }
3057
+ ],
3058
+ "logging_steps": 1,
3059
+ "max_steps": 435,
3060
+ "num_input_tokens_seen": 0,
3061
+ "num_train_epochs": 5,
3062
+ "save_steps": 500,
3063
+ "stateful_callbacks": {
3064
+ "TrainerControl": {
3065
+ "args": {
3066
+ "should_epoch_stop": false,
3067
+ "should_evaluate": false,
3068
+ "should_log": false,
3069
+ "should_save": true,
3070
+ "should_training_stop": true
3071
+ },
3072
+ "attributes": {}
3073
+ }
3074
+ },
3075
+ "total_flos": 1.1776104868752e+16,
3076
+ "train_batch_size": 2,
3077
+ "trial_name": null,
3078
+ "trial_params": null
3079
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f25f15cd195ede289467fc7cb612aae73a03d781d109be905a8b72788f46ebb3
3
+ size 6289
vocab.json ADDED
The diff for this file is too large to render. See raw diff