Upload 13 files
Browse files- config.json +186 -0
- generation_config.json +8 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer_config.json +27 -0
- trainer_state.json +1839 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
@@ -0,0 +1,186 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "nlpconnect/vit-gpt2-image-captioning",
|
3 |
+
"architectures": [
|
4 |
+
"VisionEncoderDecoderModel"
|
5 |
+
],
|
6 |
+
"bos_token_id": 50256,
|
7 |
+
"decoder": {
|
8 |
+
"_name_or_path": "",
|
9 |
+
"activation_function": "gelu_new",
|
10 |
+
"add_cross_attention": true,
|
11 |
+
"architectures": [
|
12 |
+
"GPT2LMHeadModel"
|
13 |
+
],
|
14 |
+
"attn_pdrop": 0.1,
|
15 |
+
"bad_words_ids": null,
|
16 |
+
"begin_suppress_tokens": null,
|
17 |
+
"bos_token_id": 50256,
|
18 |
+
"chunk_size_feed_forward": 0,
|
19 |
+
"cross_attention_hidden_size": null,
|
20 |
+
"decoder_start_token_id": 50256,
|
21 |
+
"diversity_penalty": 0.0,
|
22 |
+
"do_sample": false,
|
23 |
+
"early_stopping": false,
|
24 |
+
"embd_pdrop": 0.1,
|
25 |
+
"encoder_no_repeat_ngram_size": 0,
|
26 |
+
"eos_token_id": 50256,
|
27 |
+
"exponential_decay_length_penalty": null,
|
28 |
+
"finetuning_task": null,
|
29 |
+
"forced_bos_token_id": null,
|
30 |
+
"forced_eos_token_id": null,
|
31 |
+
"id2label": {
|
32 |
+
"0": "LABEL_0",
|
33 |
+
"1": "LABEL_1"
|
34 |
+
},
|
35 |
+
"initializer_range": 0.02,
|
36 |
+
"is_decoder": true,
|
37 |
+
"is_encoder_decoder": false,
|
38 |
+
"label2id": {
|
39 |
+
"LABEL_0": 0,
|
40 |
+
"LABEL_1": 1
|
41 |
+
},
|
42 |
+
"layer_norm_epsilon": 1e-05,
|
43 |
+
"length_penalty": 1.0,
|
44 |
+
"max_length": 20,
|
45 |
+
"min_length": 0,
|
46 |
+
"model_type": "gpt2",
|
47 |
+
"n_ctx": 1024,
|
48 |
+
"n_embd": 768,
|
49 |
+
"n_head": 12,
|
50 |
+
"n_inner": null,
|
51 |
+
"n_layer": 12,
|
52 |
+
"n_positions": 1024,
|
53 |
+
"no_repeat_ngram_size": 0,
|
54 |
+
"num_beam_groups": 1,
|
55 |
+
"num_beams": 1,
|
56 |
+
"num_return_sequences": 1,
|
57 |
+
"output_attentions": false,
|
58 |
+
"output_hidden_states": false,
|
59 |
+
"output_scores": false,
|
60 |
+
"pad_token_id": 50256,
|
61 |
+
"prefix": null,
|
62 |
+
"problem_type": null,
|
63 |
+
"pruned_heads": {},
|
64 |
+
"remove_invalid_values": false,
|
65 |
+
"reorder_and_upcast_attn": false,
|
66 |
+
"repetition_penalty": 1.0,
|
67 |
+
"resid_pdrop": 0.1,
|
68 |
+
"return_dict": true,
|
69 |
+
"return_dict_in_generate": false,
|
70 |
+
"scale_attn_by_inverse_layer_idx": false,
|
71 |
+
"scale_attn_weights": true,
|
72 |
+
"sep_token_id": null,
|
73 |
+
"summary_activation": null,
|
74 |
+
"summary_first_dropout": 0.1,
|
75 |
+
"summary_proj_to_labels": true,
|
76 |
+
"summary_type": "cls_index",
|
77 |
+
"summary_use_proj": true,
|
78 |
+
"suppress_tokens": null,
|
79 |
+
"task_specific_params": {
|
80 |
+
"text-generation": {
|
81 |
+
"do_sample": true,
|
82 |
+
"max_length": 50
|
83 |
+
}
|
84 |
+
},
|
85 |
+
"temperature": 1.0,
|
86 |
+
"tf_legacy_loss": false,
|
87 |
+
"tie_encoder_decoder": false,
|
88 |
+
"tie_word_embeddings": true,
|
89 |
+
"tokenizer_class": null,
|
90 |
+
"top_k": 50,
|
91 |
+
"top_p": 1.0,
|
92 |
+
"torch_dtype": null,
|
93 |
+
"torchscript": false,
|
94 |
+
"typical_p": 1.0,
|
95 |
+
"use_bfloat16": false,
|
96 |
+
"use_cache": true,
|
97 |
+
"vocab_size": 50257
|
98 |
+
},
|
99 |
+
"decoder_start_token_id": 50256,
|
100 |
+
"encoder": {
|
101 |
+
"_name_or_path": "",
|
102 |
+
"add_cross_attention": false,
|
103 |
+
"architectures": [
|
104 |
+
"ViTModel"
|
105 |
+
],
|
106 |
+
"attention_probs_dropout_prob": 0.0,
|
107 |
+
"bad_words_ids": null,
|
108 |
+
"begin_suppress_tokens": null,
|
109 |
+
"bos_token_id": null,
|
110 |
+
"chunk_size_feed_forward": 0,
|
111 |
+
"cross_attention_hidden_size": null,
|
112 |
+
"decoder_start_token_id": null,
|
113 |
+
"diversity_penalty": 0.0,
|
114 |
+
"do_sample": false,
|
115 |
+
"early_stopping": false,
|
116 |
+
"encoder_no_repeat_ngram_size": 0,
|
117 |
+
"encoder_stride": 16,
|
118 |
+
"eos_token_id": null,
|
119 |
+
"exponential_decay_length_penalty": null,
|
120 |
+
"finetuning_task": null,
|
121 |
+
"forced_bos_token_id": null,
|
122 |
+
"forced_eos_token_id": null,
|
123 |
+
"hidden_act": "gelu",
|
124 |
+
"hidden_dropout_prob": 0.0,
|
125 |
+
"hidden_size": 768,
|
126 |
+
"id2label": {
|
127 |
+
"0": "LABEL_0",
|
128 |
+
"1": "LABEL_1"
|
129 |
+
},
|
130 |
+
"image_size": 224,
|
131 |
+
"initializer_range": 0.02,
|
132 |
+
"intermediate_size": 3072,
|
133 |
+
"is_decoder": false,
|
134 |
+
"is_encoder_decoder": false,
|
135 |
+
"label2id": {
|
136 |
+
"LABEL_0": 0,
|
137 |
+
"LABEL_1": 1
|
138 |
+
},
|
139 |
+
"layer_norm_eps": 1e-12,
|
140 |
+
"length_penalty": 1.0,
|
141 |
+
"max_length": 20,
|
142 |
+
"min_length": 0,
|
143 |
+
"model_type": "vit",
|
144 |
+
"no_repeat_ngram_size": 0,
|
145 |
+
"num_attention_heads": 12,
|
146 |
+
"num_beam_groups": 1,
|
147 |
+
"num_beams": 1,
|
148 |
+
"num_channels": 3,
|
149 |
+
"num_hidden_layers": 12,
|
150 |
+
"num_return_sequences": 1,
|
151 |
+
"output_attentions": false,
|
152 |
+
"output_hidden_states": false,
|
153 |
+
"output_scores": false,
|
154 |
+
"pad_token_id": null,
|
155 |
+
"patch_size": 16,
|
156 |
+
"prefix": null,
|
157 |
+
"problem_type": null,
|
158 |
+
"pruned_heads": {},
|
159 |
+
"qkv_bias": true,
|
160 |
+
"remove_invalid_values": false,
|
161 |
+
"repetition_penalty": 1.0,
|
162 |
+
"return_dict": true,
|
163 |
+
"return_dict_in_generate": false,
|
164 |
+
"sep_token_id": null,
|
165 |
+
"suppress_tokens": null,
|
166 |
+
"task_specific_params": null,
|
167 |
+
"temperature": 1.0,
|
168 |
+
"tf_legacy_loss": false,
|
169 |
+
"tie_encoder_decoder": false,
|
170 |
+
"tie_word_embeddings": true,
|
171 |
+
"tokenizer_class": null,
|
172 |
+
"top_k": 50,
|
173 |
+
"top_p": 1.0,
|
174 |
+
"torch_dtype": null,
|
175 |
+
"torchscript": false,
|
176 |
+
"typical_p": 1.0,
|
177 |
+
"use_bfloat16": false
|
178 |
+
},
|
179 |
+
"eos_token_id": 50256,
|
180 |
+
"is_encoder_decoder": true,
|
181 |
+
"model_type": "vision-encoder-decoder",
|
182 |
+
"pad_token_id": 50256,
|
183 |
+
"tie_word_embeddings": false,
|
184 |
+
"torch_dtype": "float32",
|
185 |
+
"transformers_version": "4.45.1"
|
186 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 50256,
|
4 |
+
"decoder_start_token_id": 50256,
|
5 |
+
"eos_token_id": 50256,
|
6 |
+
"pad_token_id": 50256,
|
7 |
+
"transformers_version": "4.45.1"
|
8 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d6ac1fdab7481c560cc103a2387a7ab45581b8f3330b0731eff4167cdff27ab7
|
3 |
+
size 956835520
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:898c146266f7b58588095005f527a85021a005020405e0e9f3e8d5340eaac3a8
|
3 |
+
size 1913939014
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
|
3 |
+
size 14244
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:805533a14f5a55598ff5670adf6657b2f351dd94c3b83d445bd2e97c92fa7088
|
3 |
+
size 1064
|
special_tokens_map.json
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<|endoftext|>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "<|endoftext|>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "<|endoftext|>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"unk_token": {
|
24 |
+
"content": "<|endoftext|>",
|
25 |
+
"lstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
}
|
30 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"50256": {
|
5 |
+
"content": "<|endoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": false,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
}
|
12 |
+
},
|
13 |
+
"bos_token": "<|endoftext|>",
|
14 |
+
"clean_up_tokenization_spaces": false,
|
15 |
+
"eos_token": "<|endoftext|>",
|
16 |
+
"max_length": 32,
|
17 |
+
"model_max_length": 1024,
|
18 |
+
"pad_to_multiple_of": null,
|
19 |
+
"pad_token": "<|endoftext|>",
|
20 |
+
"pad_token_type_id": 0,
|
21 |
+
"padding_side": "right",
|
22 |
+
"stride": 0,
|
23 |
+
"tokenizer_class": "GPT2Tokenizer",
|
24 |
+
"truncation_side": "right",
|
25 |
+
"truncation_strategy": "longest_first",
|
26 |
+
"unk_token": "<|endoftext|>"
|
27 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1839 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9972941631233089,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 2580,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.003865481252415926,
|
13 |
+
"grad_norm": 0.0,
|
14 |
+
"learning_rate": 4.9935575312459736e-05,
|
15 |
+
"loss": 0.0,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.007730962504831852,
|
20 |
+
"grad_norm": 0.0,
|
21 |
+
"learning_rate": 4.987115062491947e-05,
|
22 |
+
"loss": 0.0,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.011596443757247778,
|
27 |
+
"grad_norm": 0.0,
|
28 |
+
"learning_rate": 4.98067259373792e-05,
|
29 |
+
"loss": 0.0,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.015461925009663703,
|
34 |
+
"grad_norm": 0.0,
|
35 |
+
"learning_rate": 4.974230124983894e-05,
|
36 |
+
"loss": 0.0,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.019327406262079627,
|
41 |
+
"grad_norm": 0.0,
|
42 |
+
"learning_rate": 4.967787656229868e-05,
|
43 |
+
"loss": 0.0,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.023192887514495556,
|
48 |
+
"grad_norm": 0.0,
|
49 |
+
"learning_rate": 4.961345187475841e-05,
|
50 |
+
"loss": 0.0,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.02705836876691148,
|
55 |
+
"grad_norm": 0.0,
|
56 |
+
"learning_rate": 4.9549027187218144e-05,
|
57 |
+
"loss": 0.0,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.030923850019327407,
|
62 |
+
"grad_norm": 0.0,
|
63 |
+
"learning_rate": 4.948460249967788e-05,
|
64 |
+
"loss": 0.0,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.03478933127174333,
|
69 |
+
"grad_norm": 0.0,
|
70 |
+
"learning_rate": 4.942017781213761e-05,
|
71 |
+
"loss": 0.0,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.038654812524159254,
|
76 |
+
"grad_norm": 0.0,
|
77 |
+
"learning_rate": 4.935575312459735e-05,
|
78 |
+
"loss": 0.0,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.04252029377657519,
|
83 |
+
"grad_norm": 0.0,
|
84 |
+
"learning_rate": 4.9291328437057085e-05,
|
85 |
+
"loss": 0.0,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.04638577502899111,
|
90 |
+
"grad_norm": 0.0,
|
91 |
+
"learning_rate": 4.922690374951682e-05,
|
92 |
+
"loss": 0.0,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.05025125628140704,
|
97 |
+
"grad_norm": 0.0,
|
98 |
+
"learning_rate": 4.916247906197655e-05,
|
99 |
+
"loss": 0.0,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.05411673753382296,
|
104 |
+
"grad_norm": 0.0,
|
105 |
+
"learning_rate": 4.9098054374436285e-05,
|
106 |
+
"loss": 0.0,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.05798221878623889,
|
111 |
+
"grad_norm": 0.0,
|
112 |
+
"learning_rate": 4.903362968689602e-05,
|
113 |
+
"loss": 0.0,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.061847700038654814,
|
118 |
+
"grad_norm": 0.0,
|
119 |
+
"learning_rate": 4.896920499935575e-05,
|
120 |
+
"loss": 0.0,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.06571318129107073,
|
125 |
+
"grad_norm": 0.0,
|
126 |
+
"learning_rate": 4.890478031181549e-05,
|
127 |
+
"loss": 0.0,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.06957866254348666,
|
132 |
+
"grad_norm": 0.0,
|
133 |
+
"learning_rate": 4.8840355624275226e-05,
|
134 |
+
"loss": 0.0,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.07344414379590258,
|
139 |
+
"grad_norm": 0.0,
|
140 |
+
"learning_rate": 4.877593093673496e-05,
|
141 |
+
"loss": 0.0,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.07730962504831851,
|
146 |
+
"grad_norm": 0.0,
|
147 |
+
"learning_rate": 4.8711506249194693e-05,
|
148 |
+
"loss": 0.0,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.08117510630073445,
|
153 |
+
"grad_norm": 0.0,
|
154 |
+
"learning_rate": 4.864708156165443e-05,
|
155 |
+
"loss": 0.0,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.08504058755315037,
|
160 |
+
"grad_norm": 0.0,
|
161 |
+
"learning_rate": 4.858265687411416e-05,
|
162 |
+
"loss": 0.0,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.0889060688055663,
|
167 |
+
"grad_norm": 0.0,
|
168 |
+
"learning_rate": 4.85182321865739e-05,
|
169 |
+
"loss": 0.0,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.09277155005798222,
|
174 |
+
"grad_norm": 0.0,
|
175 |
+
"learning_rate": 4.8453807499033634e-05,
|
176 |
+
"loss": 0.0,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.09663703131039815,
|
181 |
+
"grad_norm": 0.0,
|
182 |
+
"learning_rate": 4.838938281149337e-05,
|
183 |
+
"loss": 0.0,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.10050251256281408,
|
188 |
+
"grad_norm": 0.0,
|
189 |
+
"learning_rate": 4.83249581239531e-05,
|
190 |
+
"loss": 0.0,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.10436799381523,
|
195 |
+
"grad_norm": 0.0,
|
196 |
+
"learning_rate": 4.8260533436412835e-05,
|
197 |
+
"loss": 0.0,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.10823347506764593,
|
202 |
+
"grad_norm": 0.0,
|
203 |
+
"learning_rate": 4.819610874887257e-05,
|
204 |
+
"loss": 0.0,
|
205 |
+
"step": 280
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.11209895632006185,
|
209 |
+
"grad_norm": 0.0,
|
210 |
+
"learning_rate": 4.81316840613323e-05,
|
211 |
+
"loss": 0.0,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.11596443757247778,
|
216 |
+
"grad_norm": 0.0,
|
217 |
+
"learning_rate": 4.806725937379204e-05,
|
218 |
+
"loss": 0.0,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.1198299188248937,
|
223 |
+
"grad_norm": 0.0,
|
224 |
+
"learning_rate": 4.8002834686251776e-05,
|
225 |
+
"loss": 0.0,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.12369540007730963,
|
230 |
+
"grad_norm": 0.0,
|
231 |
+
"learning_rate": 4.793840999871151e-05,
|
232 |
+
"loss": 0.0,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.12756088132972554,
|
237 |
+
"grad_norm": 0.0,
|
238 |
+
"learning_rate": 4.787398531117124e-05,
|
239 |
+
"loss": 0.0,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.13142636258214146,
|
244 |
+
"grad_norm": 0.0,
|
245 |
+
"learning_rate": 4.7809560623630976e-05,
|
246 |
+
"loss": 0.0,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.1352918438345574,
|
251 |
+
"grad_norm": 0.0,
|
252 |
+
"learning_rate": 4.774513593609071e-05,
|
253 |
+
"loss": 0.0,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.13915732508697332,
|
258 |
+
"grad_norm": 0.0,
|
259 |
+
"learning_rate": 4.768071124855045e-05,
|
260 |
+
"loss": 0.0,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.14302280633938924,
|
265 |
+
"grad_norm": 0.0,
|
266 |
+
"learning_rate": 4.7616286561010184e-05,
|
267 |
+
"loss": 0.0,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.14688828759180517,
|
272 |
+
"grad_norm": 0.0,
|
273 |
+
"learning_rate": 4.755186187346992e-05,
|
274 |
+
"loss": 0.0,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.1507537688442211,
|
279 |
+
"grad_norm": 0.0,
|
280 |
+
"learning_rate": 4.748743718592965e-05,
|
281 |
+
"loss": 0.0,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.15461925009663702,
|
286 |
+
"grad_norm": 0.0,
|
287 |
+
"learning_rate": 4.7423012498389384e-05,
|
288 |
+
"loss": 0.0,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.15848473134905297,
|
293 |
+
"grad_norm": 0.0,
|
294 |
+
"learning_rate": 4.735858781084912e-05,
|
295 |
+
"loss": 0.0,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.1623502126014689,
|
300 |
+
"grad_norm": 0.0,
|
301 |
+
"learning_rate": 4.729416312330885e-05,
|
302 |
+
"loss": 0.0,
|
303 |
+
"step": 420
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.16621569385388482,
|
307 |
+
"grad_norm": 0.0,
|
308 |
+
"learning_rate": 4.722973843576859e-05,
|
309 |
+
"loss": 0.0,
|
310 |
+
"step": 430
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.17008117510630075,
|
314 |
+
"grad_norm": 0.0,
|
315 |
+
"learning_rate": 4.7165313748228325e-05,
|
316 |
+
"loss": 0.0,
|
317 |
+
"step": 440
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.17394665635871667,
|
321 |
+
"grad_norm": 0.0,
|
322 |
+
"learning_rate": 4.710088906068806e-05,
|
323 |
+
"loss": 0.0,
|
324 |
+
"step": 450
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.1778121376111326,
|
328 |
+
"grad_norm": 0.0,
|
329 |
+
"learning_rate": 4.703646437314779e-05,
|
330 |
+
"loss": 0.0,
|
331 |
+
"step": 460
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.18167761886354852,
|
335 |
+
"grad_norm": 0.0,
|
336 |
+
"learning_rate": 4.6972039685607526e-05,
|
337 |
+
"loss": 0.0,
|
338 |
+
"step": 470
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.18554310011596445,
|
342 |
+
"grad_norm": 0.0,
|
343 |
+
"learning_rate": 4.690761499806726e-05,
|
344 |
+
"loss": 0.0,
|
345 |
+
"step": 480
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.18940858136838037,
|
349 |
+
"grad_norm": 0.0,
|
350 |
+
"learning_rate": 4.6843190310527e-05,
|
351 |
+
"loss": 0.0,
|
352 |
+
"step": 490
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.1932740626207963,
|
356 |
+
"grad_norm": 0.0,
|
357 |
+
"learning_rate": 4.677876562298673e-05,
|
358 |
+
"loss": 0.0,
|
359 |
+
"step": 500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.19713954387321223,
|
363 |
+
"grad_norm": 0.0,
|
364 |
+
"learning_rate": 4.671434093544647e-05,
|
365 |
+
"loss": 0.0,
|
366 |
+
"step": 510
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.20100502512562815,
|
370 |
+
"grad_norm": 0.0,
|
371 |
+
"learning_rate": 4.66499162479062e-05,
|
372 |
+
"loss": 0.0,
|
373 |
+
"step": 520
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.20487050637804408,
|
377 |
+
"grad_norm": 0.0,
|
378 |
+
"learning_rate": 4.6585491560365934e-05,
|
379 |
+
"loss": 0.0,
|
380 |
+
"step": 530
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.20873598763046,
|
384 |
+
"grad_norm": 0.0,
|
385 |
+
"learning_rate": 4.652106687282567e-05,
|
386 |
+
"loss": 0.0,
|
387 |
+
"step": 540
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.21260146888287593,
|
391 |
+
"grad_norm": 0.0,
|
392 |
+
"learning_rate": 4.64566421852854e-05,
|
393 |
+
"loss": 0.0,
|
394 |
+
"step": 550
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.21646695013529185,
|
398 |
+
"grad_norm": 0.0,
|
399 |
+
"learning_rate": 4.639221749774514e-05,
|
400 |
+
"loss": 0.0,
|
401 |
+
"step": 560
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.22033243138770778,
|
405 |
+
"grad_norm": 0.0,
|
406 |
+
"learning_rate": 4.6327792810204875e-05,
|
407 |
+
"loss": 0.0,
|
408 |
+
"step": 570
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.2241979126401237,
|
412 |
+
"grad_norm": 0.0,
|
413 |
+
"learning_rate": 4.626336812266461e-05,
|
414 |
+
"loss": 0.0,
|
415 |
+
"step": 580
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.22806339389253963,
|
419 |
+
"grad_norm": 0.0,
|
420 |
+
"learning_rate": 4.619894343512434e-05,
|
421 |
+
"loss": 0.0,
|
422 |
+
"step": 590
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.23192887514495555,
|
426 |
+
"grad_norm": 0.0,
|
427 |
+
"learning_rate": 4.6134518747584075e-05,
|
428 |
+
"loss": 0.0,
|
429 |
+
"step": 600
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.23579435639737148,
|
433 |
+
"grad_norm": 0.0,
|
434 |
+
"learning_rate": 4.607009406004381e-05,
|
435 |
+
"loss": 0.0,
|
436 |
+
"step": 610
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.2396598376497874,
|
440 |
+
"grad_norm": 0.0,
|
441 |
+
"learning_rate": 4.600566937250354e-05,
|
442 |
+
"loss": 0.0,
|
443 |
+
"step": 620
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.24352531890220333,
|
447 |
+
"grad_norm": 0.0,
|
448 |
+
"learning_rate": 4.594124468496328e-05,
|
449 |
+
"loss": 0.0,
|
450 |
+
"step": 630
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.24739080015461926,
|
454 |
+
"grad_norm": 0.0,
|
455 |
+
"learning_rate": 4.5876819997423016e-05,
|
456 |
+
"loss": 0.0,
|
457 |
+
"step": 640
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.25125628140703515,
|
461 |
+
"grad_norm": 0.0,
|
462 |
+
"learning_rate": 4.581239530988275e-05,
|
463 |
+
"loss": 0.0,
|
464 |
+
"step": 650
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.2551217626594511,
|
468 |
+
"grad_norm": 0.0,
|
469 |
+
"learning_rate": 4.5747970622342483e-05,
|
470 |
+
"loss": 0.0,
|
471 |
+
"step": 660
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.258987243911867,
|
475 |
+
"grad_norm": 0.0,
|
476 |
+
"learning_rate": 4.568354593480222e-05,
|
477 |
+
"loss": 0.0,
|
478 |
+
"step": 670
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.26285272516428293,
|
482 |
+
"grad_norm": 0.0,
|
483 |
+
"learning_rate": 4.561912124726195e-05,
|
484 |
+
"loss": 0.0,
|
485 |
+
"step": 680
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.26671820641669886,
|
489 |
+
"grad_norm": 0.0,
|
490 |
+
"learning_rate": 4.555469655972169e-05,
|
491 |
+
"loss": 0.0,
|
492 |
+
"step": 690
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.2705836876691148,
|
496 |
+
"grad_norm": 0.0,
|
497 |
+
"learning_rate": 4.5490271872181424e-05,
|
498 |
+
"loss": 0.0,
|
499 |
+
"step": 700
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.2744491689215307,
|
503 |
+
"grad_norm": 0.0,
|
504 |
+
"learning_rate": 4.542584718464116e-05,
|
505 |
+
"loss": 0.0,
|
506 |
+
"step": 710
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.27831465017394663,
|
510 |
+
"grad_norm": 0.0,
|
511 |
+
"learning_rate": 4.536142249710089e-05,
|
512 |
+
"loss": 0.0,
|
513 |
+
"step": 720
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.28218013142636256,
|
517 |
+
"grad_norm": 0.0,
|
518 |
+
"learning_rate": 4.5296997809560625e-05,
|
519 |
+
"loss": 0.0,
|
520 |
+
"step": 730
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.2860456126787785,
|
524 |
+
"grad_norm": 0.0,
|
525 |
+
"learning_rate": 4.523257312202036e-05,
|
526 |
+
"loss": 0.0,
|
527 |
+
"step": 740
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.2899110939311944,
|
531 |
+
"grad_norm": 0.0,
|
532 |
+
"learning_rate": 4.516814843448009e-05,
|
533 |
+
"loss": 0.0,
|
534 |
+
"step": 750
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.29377657518361033,
|
538 |
+
"grad_norm": 0.0,
|
539 |
+
"learning_rate": 4.510372374693983e-05,
|
540 |
+
"loss": 0.0,
|
541 |
+
"step": 760
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.29764205643602626,
|
545 |
+
"grad_norm": 0.0,
|
546 |
+
"learning_rate": 4.5039299059399566e-05,
|
547 |
+
"loss": 0.0,
|
548 |
+
"step": 770
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.3015075376884422,
|
552 |
+
"grad_norm": 0.0,
|
553 |
+
"learning_rate": 4.49748743718593e-05,
|
554 |
+
"loss": 0.0,
|
555 |
+
"step": 780
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.3053730189408581,
|
559 |
+
"grad_norm": 0.0,
|
560 |
+
"learning_rate": 4.491044968431903e-05,
|
561 |
+
"loss": 0.0,
|
562 |
+
"step": 790
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.30923850019327404,
|
566 |
+
"grad_norm": 0.0,
|
567 |
+
"learning_rate": 4.4846024996778767e-05,
|
568 |
+
"loss": 0.0,
|
569 |
+
"step": 800
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.31310398144569,
|
573 |
+
"grad_norm": 0.0,
|
574 |
+
"learning_rate": 4.47816003092385e-05,
|
575 |
+
"loss": 0.0,
|
576 |
+
"step": 810
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.31696946269810594,
|
580 |
+
"grad_norm": 0.0,
|
581 |
+
"learning_rate": 4.471717562169824e-05,
|
582 |
+
"loss": 0.0,
|
583 |
+
"step": 820
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.32083494395052187,
|
587 |
+
"grad_norm": 0.0,
|
588 |
+
"learning_rate": 4.4652750934157974e-05,
|
589 |
+
"loss": 0.0,
|
590 |
+
"step": 830
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.3247004252029378,
|
594 |
+
"grad_norm": 0.0,
|
595 |
+
"learning_rate": 4.458832624661771e-05,
|
596 |
+
"loss": 0.0,
|
597 |
+
"step": 840
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.3285659064553537,
|
601 |
+
"grad_norm": 0.0,
|
602 |
+
"learning_rate": 4.452390155907744e-05,
|
603 |
+
"loss": 0.0,
|
604 |
+
"step": 850
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.33243138770776964,
|
608 |
+
"grad_norm": 0.0,
|
609 |
+
"learning_rate": 4.4459476871537175e-05,
|
610 |
+
"loss": 0.0,
|
611 |
+
"step": 860
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.33629686896018557,
|
615 |
+
"grad_norm": 0.0,
|
616 |
+
"learning_rate": 4.439505218399691e-05,
|
617 |
+
"loss": 0.0,
|
618 |
+
"step": 870
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.3401623502126015,
|
622 |
+
"grad_norm": 0.0,
|
623 |
+
"learning_rate": 4.433062749645664e-05,
|
624 |
+
"loss": 0.0,
|
625 |
+
"step": 880
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.3440278314650174,
|
629 |
+
"grad_norm": 0.0,
|
630 |
+
"learning_rate": 4.426620280891638e-05,
|
631 |
+
"loss": 0.0,
|
632 |
+
"step": 890
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.34789331271743335,
|
636 |
+
"grad_norm": 0.0,
|
637 |
+
"learning_rate": 4.4201778121376115e-05,
|
638 |
+
"loss": 0.0,
|
639 |
+
"step": 900
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.35175879396984927,
|
643 |
+
"grad_norm": 0.0,
|
644 |
+
"learning_rate": 4.413735343383585e-05,
|
645 |
+
"loss": 0.0,
|
646 |
+
"step": 910
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.3556242752222652,
|
650 |
+
"grad_norm": 0.0,
|
651 |
+
"learning_rate": 4.407292874629558e-05,
|
652 |
+
"loss": 0.0,
|
653 |
+
"step": 920
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.3594897564746811,
|
657 |
+
"grad_norm": 0.0,
|
658 |
+
"learning_rate": 4.4008504058755316e-05,
|
659 |
+
"loss": 0.0,
|
660 |
+
"step": 930
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.36335523772709705,
|
664 |
+
"grad_norm": 0.0,
|
665 |
+
"learning_rate": 4.394407937121505e-05,
|
666 |
+
"loss": 0.0,
|
667 |
+
"step": 940
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.367220718979513,
|
671 |
+
"grad_norm": 0.0,
|
672 |
+
"learning_rate": 4.387965468367479e-05,
|
673 |
+
"loss": 0.0,
|
674 |
+
"step": 950
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.3710862002319289,
|
678 |
+
"grad_norm": 0.0,
|
679 |
+
"learning_rate": 4.3815229996134523e-05,
|
680 |
+
"loss": 0.0,
|
681 |
+
"step": 960
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.3749516814843448,
|
685 |
+
"grad_norm": 0.0,
|
686 |
+
"learning_rate": 4.375080530859426e-05,
|
687 |
+
"loss": 0.0,
|
688 |
+
"step": 970
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.37881716273676075,
|
692 |
+
"grad_norm": 0.0,
|
693 |
+
"learning_rate": 4.368638062105399e-05,
|
694 |
+
"loss": 0.0,
|
695 |
+
"step": 980
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.3826826439891767,
|
699 |
+
"grad_norm": 0.0,
|
700 |
+
"learning_rate": 4.3621955933513724e-05,
|
701 |
+
"loss": 0.0,
|
702 |
+
"step": 990
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.3865481252415926,
|
706 |
+
"grad_norm": 0.0,
|
707 |
+
"learning_rate": 4.355753124597346e-05,
|
708 |
+
"loss": 0.0,
|
709 |
+
"step": 1000
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.3904136064940085,
|
713 |
+
"grad_norm": 0.0,
|
714 |
+
"learning_rate": 4.349310655843319e-05,
|
715 |
+
"loss": 0.0,
|
716 |
+
"step": 1010
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.39427908774642445,
|
720 |
+
"grad_norm": 0.0,
|
721 |
+
"learning_rate": 4.342868187089293e-05,
|
722 |
+
"loss": 0.0,
|
723 |
+
"step": 1020
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.3981445689988404,
|
727 |
+
"grad_norm": 0.0,
|
728 |
+
"learning_rate": 4.3364257183352665e-05,
|
729 |
+
"loss": 0.0,
|
730 |
+
"step": 1030
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.4020100502512563,
|
734 |
+
"grad_norm": 0.0,
|
735 |
+
"learning_rate": 4.32998324958124e-05,
|
736 |
+
"loss": 0.0,
|
737 |
+
"step": 1040
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.4058755315036722,
|
741 |
+
"grad_norm": 0.0,
|
742 |
+
"learning_rate": 4.323540780827213e-05,
|
743 |
+
"loss": 0.0,
|
744 |
+
"step": 1050
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.40974101275608815,
|
748 |
+
"grad_norm": 0.0,
|
749 |
+
"learning_rate": 4.3170983120731866e-05,
|
750 |
+
"loss": 0.0,
|
751 |
+
"step": 1060
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.4136064940085041,
|
755 |
+
"grad_norm": 0.0,
|
756 |
+
"learning_rate": 4.31065584331916e-05,
|
757 |
+
"loss": 0.0,
|
758 |
+
"step": 1070
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.41747197526092,
|
762 |
+
"grad_norm": 0.0,
|
763 |
+
"learning_rate": 4.304213374565134e-05,
|
764 |
+
"loss": 0.0,
|
765 |
+
"step": 1080
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.42133745651333593,
|
769 |
+
"grad_norm": 0.0,
|
770 |
+
"learning_rate": 4.297770905811107e-05,
|
771 |
+
"loss": 0.0,
|
772 |
+
"step": 1090
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.42520293776575185,
|
776 |
+
"grad_norm": 0.0,
|
777 |
+
"learning_rate": 4.2913284370570806e-05,
|
778 |
+
"loss": 0.0,
|
779 |
+
"step": 1100
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.4290684190181678,
|
783 |
+
"grad_norm": 0.0,
|
784 |
+
"learning_rate": 4.284885968303054e-05,
|
785 |
+
"loss": 0.0,
|
786 |
+
"step": 1110
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.4329339002705837,
|
790 |
+
"grad_norm": 0.0,
|
791 |
+
"learning_rate": 4.2784434995490274e-05,
|
792 |
+
"loss": 0.0,
|
793 |
+
"step": 1120
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.43679938152299963,
|
797 |
+
"grad_norm": 0.0,
|
798 |
+
"learning_rate": 4.272001030795001e-05,
|
799 |
+
"loss": 0.0,
|
800 |
+
"step": 1130
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.44066486277541556,
|
804 |
+
"grad_norm": 0.0,
|
805 |
+
"learning_rate": 4.265558562040974e-05,
|
806 |
+
"loss": 0.0,
|
807 |
+
"step": 1140
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.4445303440278315,
|
811 |
+
"grad_norm": 0.0,
|
812 |
+
"learning_rate": 4.259116093286948e-05,
|
813 |
+
"loss": 0.0,
|
814 |
+
"step": 1150
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.4483958252802474,
|
818 |
+
"grad_norm": 0.0,
|
819 |
+
"learning_rate": 4.2526736245329214e-05,
|
820 |
+
"loss": 0.0,
|
821 |
+
"step": 1160
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.45226130653266333,
|
825 |
+
"grad_norm": 0.0,
|
826 |
+
"learning_rate": 4.246231155778895e-05,
|
827 |
+
"loss": 0.0,
|
828 |
+
"step": 1170
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.45612678778507926,
|
832 |
+
"grad_norm": 0.0,
|
833 |
+
"learning_rate": 4.239788687024868e-05,
|
834 |
+
"loss": 0.0,
|
835 |
+
"step": 1180
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.4599922690374952,
|
839 |
+
"grad_norm": 0.0,
|
840 |
+
"learning_rate": 4.2333462182708415e-05,
|
841 |
+
"loss": 0.0,
|
842 |
+
"step": 1190
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.4638577502899111,
|
846 |
+
"grad_norm": 0.0,
|
847 |
+
"learning_rate": 4.226903749516815e-05,
|
848 |
+
"loss": 0.0,
|
849 |
+
"step": 1200
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.46772323154232703,
|
853 |
+
"grad_norm": 0.0,
|
854 |
+
"learning_rate": 4.220461280762789e-05,
|
855 |
+
"loss": 0.0,
|
856 |
+
"step": 1210
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.47158871279474296,
|
860 |
+
"grad_norm": 0.0,
|
861 |
+
"learning_rate": 4.214018812008762e-05,
|
862 |
+
"loss": 0.0,
|
863 |
+
"step": 1220
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.4754541940471589,
|
867 |
+
"grad_norm": 0.0,
|
868 |
+
"learning_rate": 4.2075763432547356e-05,
|
869 |
+
"loss": 0.0,
|
870 |
+
"step": 1230
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.4793196752995748,
|
874 |
+
"grad_norm": 0.0,
|
875 |
+
"learning_rate": 4.201133874500709e-05,
|
876 |
+
"loss": 0.0,
|
877 |
+
"step": 1240
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.48318515655199074,
|
881 |
+
"grad_norm": 0.0,
|
882 |
+
"learning_rate": 4.194691405746682e-05,
|
883 |
+
"loss": 0.0,
|
884 |
+
"step": 1250
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.48705063780440666,
|
888 |
+
"grad_norm": 0.0,
|
889 |
+
"learning_rate": 4.1882489369926557e-05,
|
890 |
+
"loss": 0.0,
|
891 |
+
"step": 1260
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.4909161190568226,
|
895 |
+
"grad_norm": 0.0,
|
896 |
+
"learning_rate": 4.181806468238629e-05,
|
897 |
+
"loss": 0.0,
|
898 |
+
"step": 1270
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.4947816003092385,
|
902 |
+
"grad_norm": 0.0,
|
903 |
+
"learning_rate": 4.175363999484603e-05,
|
904 |
+
"loss": 0.0,
|
905 |
+
"step": 1280
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.49864708156165444,
|
909 |
+
"grad_norm": 0.0,
|
910 |
+
"learning_rate": 4.1689215307305764e-05,
|
911 |
+
"loss": 0.0,
|
912 |
+
"step": 1290
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.5025125628140703,
|
916 |
+
"grad_norm": 0.0,
|
917 |
+
"learning_rate": 4.16247906197655e-05,
|
918 |
+
"loss": 0.0,
|
919 |
+
"step": 1300
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.5063780440664862,
|
923 |
+
"grad_norm": 0.0,
|
924 |
+
"learning_rate": 4.156036593222523e-05,
|
925 |
+
"loss": 0.0,
|
926 |
+
"step": 1310
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.5102435253189022,
|
930 |
+
"grad_norm": 0.0,
|
931 |
+
"learning_rate": 4.1495941244684965e-05,
|
932 |
+
"loss": 0.0,
|
933 |
+
"step": 1320
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.5141090065713181,
|
937 |
+
"grad_norm": 0.0,
|
938 |
+
"learning_rate": 4.14315165571447e-05,
|
939 |
+
"loss": 0.0,
|
940 |
+
"step": 1330
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.517974487823734,
|
944 |
+
"grad_norm": 0.0,
|
945 |
+
"learning_rate": 4.136709186960444e-05,
|
946 |
+
"loss": 0.0,
|
947 |
+
"step": 1340
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.5218399690761499,
|
951 |
+
"grad_norm": 0.0,
|
952 |
+
"learning_rate": 4.130266718206417e-05,
|
953 |
+
"loss": 0.0,
|
954 |
+
"step": 1350
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.5257054503285659,
|
958 |
+
"grad_norm": 0.0,
|
959 |
+
"learning_rate": 4.1238242494523905e-05,
|
960 |
+
"loss": 0.0,
|
961 |
+
"step": 1360
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.5295709315809818,
|
965 |
+
"grad_norm": 0.0,
|
966 |
+
"learning_rate": 4.117381780698364e-05,
|
967 |
+
"loss": 0.0,
|
968 |
+
"step": 1370
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.5334364128333977,
|
972 |
+
"grad_norm": 0.0,
|
973 |
+
"learning_rate": 4.110939311944337e-05,
|
974 |
+
"loss": 0.0,
|
975 |
+
"step": 1380
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.5373018940858136,
|
979 |
+
"grad_norm": 0.0,
|
980 |
+
"learning_rate": 4.1044968431903106e-05,
|
981 |
+
"loss": 0.0,
|
982 |
+
"step": 1390
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.5411673753382296,
|
986 |
+
"grad_norm": 0.0,
|
987 |
+
"learning_rate": 4.098054374436284e-05,
|
988 |
+
"loss": 0.0,
|
989 |
+
"step": 1400
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.5450328565906455,
|
993 |
+
"grad_norm": 0.0,
|
994 |
+
"learning_rate": 4.091611905682258e-05,
|
995 |
+
"loss": 0.0,
|
996 |
+
"step": 1410
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.5488983378430614,
|
1000 |
+
"grad_norm": 0.0,
|
1001 |
+
"learning_rate": 4.0851694369282313e-05,
|
1002 |
+
"loss": 0.0,
|
1003 |
+
"step": 1420
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.5527638190954773,
|
1007 |
+
"grad_norm": 0.0,
|
1008 |
+
"learning_rate": 4.078726968174205e-05,
|
1009 |
+
"loss": 0.0,
|
1010 |
+
"step": 1430
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.5566293003478933,
|
1014 |
+
"grad_norm": 0.0,
|
1015 |
+
"learning_rate": 4.072284499420178e-05,
|
1016 |
+
"loss": 0.0,
|
1017 |
+
"step": 1440
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.5604947816003092,
|
1021 |
+
"grad_norm": 0.0,
|
1022 |
+
"learning_rate": 4.0658420306661514e-05,
|
1023 |
+
"loss": 0.0,
|
1024 |
+
"step": 1450
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.5643602628527251,
|
1028 |
+
"grad_norm": 0.0,
|
1029 |
+
"learning_rate": 4.059399561912125e-05,
|
1030 |
+
"loss": 0.0,
|
1031 |
+
"step": 1460
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.568225744105141,
|
1035 |
+
"grad_norm": 0.0,
|
1036 |
+
"learning_rate": 4.052957093158099e-05,
|
1037 |
+
"loss": 0.0,
|
1038 |
+
"step": 1470
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.572091225357557,
|
1042 |
+
"grad_norm": 0.0,
|
1043 |
+
"learning_rate": 4.046514624404072e-05,
|
1044 |
+
"loss": 0.0,
|
1045 |
+
"step": 1480
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.5759567066099729,
|
1049 |
+
"grad_norm": 0.0,
|
1050 |
+
"learning_rate": 4.0400721556500455e-05,
|
1051 |
+
"loss": 0.0,
|
1052 |
+
"step": 1490
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.5798221878623888,
|
1056 |
+
"grad_norm": 0.0,
|
1057 |
+
"learning_rate": 4.033629686896019e-05,
|
1058 |
+
"loss": 0.0,
|
1059 |
+
"step": 1500
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.5836876691148047,
|
1063 |
+
"grad_norm": 0.0,
|
1064 |
+
"learning_rate": 4.027187218141992e-05,
|
1065 |
+
"loss": 0.0,
|
1066 |
+
"step": 1510
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.5875531503672207,
|
1070 |
+
"grad_norm": 0.0,
|
1071 |
+
"learning_rate": 4.0207447493879656e-05,
|
1072 |
+
"loss": 0.0,
|
1073 |
+
"step": 1520
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.5914186316196366,
|
1077 |
+
"grad_norm": 0.0,
|
1078 |
+
"learning_rate": 4.014302280633939e-05,
|
1079 |
+
"loss": 0.0,
|
1080 |
+
"step": 1530
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.5952841128720525,
|
1084 |
+
"grad_norm": 0.0,
|
1085 |
+
"learning_rate": 4.007859811879913e-05,
|
1086 |
+
"loss": 0.0,
|
1087 |
+
"step": 1540
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.5991495941244684,
|
1091 |
+
"grad_norm": 0.0,
|
1092 |
+
"learning_rate": 4.001417343125886e-05,
|
1093 |
+
"loss": 0.0,
|
1094 |
+
"step": 1550
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.6030150753768844,
|
1098 |
+
"grad_norm": 0.0,
|
1099 |
+
"learning_rate": 3.9949748743718597e-05,
|
1100 |
+
"loss": 0.0,
|
1101 |
+
"step": 1560
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.6068805566293003,
|
1105 |
+
"grad_norm": 0.0,
|
1106 |
+
"learning_rate": 3.988532405617833e-05,
|
1107 |
+
"loss": 0.0,
|
1108 |
+
"step": 1570
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.6107460378817162,
|
1112 |
+
"grad_norm": 0.0,
|
1113 |
+
"learning_rate": 3.9820899368638064e-05,
|
1114 |
+
"loss": 0.0,
|
1115 |
+
"step": 1580
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.6146115191341321,
|
1119 |
+
"grad_norm": 0.0,
|
1120 |
+
"learning_rate": 3.97564746810978e-05,
|
1121 |
+
"loss": 0.0,
|
1122 |
+
"step": 1590
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.6184770003865481,
|
1126 |
+
"grad_norm": 0.0,
|
1127 |
+
"learning_rate": 3.969204999355754e-05,
|
1128 |
+
"loss": 0.0,
|
1129 |
+
"step": 1600
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.622342481638964,
|
1133 |
+
"grad_norm": 0.0,
|
1134 |
+
"learning_rate": 3.962762530601727e-05,
|
1135 |
+
"loss": 0.0,
|
1136 |
+
"step": 1610
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.62620796289138,
|
1140 |
+
"grad_norm": 0.0,
|
1141 |
+
"learning_rate": 3.9563200618477005e-05,
|
1142 |
+
"loss": 0.0,
|
1143 |
+
"step": 1620
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.630073444143796,
|
1147 |
+
"grad_norm": 0.0,
|
1148 |
+
"learning_rate": 3.949877593093674e-05,
|
1149 |
+
"loss": 0.0,
|
1150 |
+
"step": 1630
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.6339389253962119,
|
1154 |
+
"grad_norm": 0.0,
|
1155 |
+
"learning_rate": 3.943435124339647e-05,
|
1156 |
+
"loss": 0.0,
|
1157 |
+
"step": 1640
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.6378044066486278,
|
1161 |
+
"grad_norm": 0.0,
|
1162 |
+
"learning_rate": 3.9369926555856205e-05,
|
1163 |
+
"loss": 0.0,
|
1164 |
+
"step": 1650
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.6416698879010437,
|
1168 |
+
"grad_norm": 0.0,
|
1169 |
+
"learning_rate": 3.930550186831594e-05,
|
1170 |
+
"loss": 0.0,
|
1171 |
+
"step": 1660
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.6455353691534597,
|
1175 |
+
"grad_norm": 0.0,
|
1176 |
+
"learning_rate": 3.924107718077568e-05,
|
1177 |
+
"loss": 0.0,
|
1178 |
+
"step": 1670
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.6494008504058756,
|
1182 |
+
"grad_norm": 0.0,
|
1183 |
+
"learning_rate": 3.917665249323541e-05,
|
1184 |
+
"loss": 0.0,
|
1185 |
+
"step": 1680
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.6532663316582915,
|
1189 |
+
"grad_norm": 0.0,
|
1190 |
+
"learning_rate": 3.9112227805695146e-05,
|
1191 |
+
"loss": 0.0,
|
1192 |
+
"step": 1690
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.6571318129107074,
|
1196 |
+
"grad_norm": 0.0,
|
1197 |
+
"learning_rate": 3.904780311815488e-05,
|
1198 |
+
"loss": 0.0,
|
1199 |
+
"step": 1700
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.6609972941631234,
|
1203 |
+
"grad_norm": 0.0,
|
1204 |
+
"learning_rate": 3.898337843061461e-05,
|
1205 |
+
"loss": 0.0,
|
1206 |
+
"step": 1710
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.6648627754155393,
|
1210 |
+
"grad_norm": 0.0,
|
1211 |
+
"learning_rate": 3.891895374307435e-05,
|
1212 |
+
"loss": 0.0,
|
1213 |
+
"step": 1720
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.6687282566679552,
|
1217 |
+
"grad_norm": 0.0,
|
1218 |
+
"learning_rate": 3.885452905553408e-05,
|
1219 |
+
"loss": 0.0,
|
1220 |
+
"step": 1730
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.6725937379203711,
|
1224 |
+
"grad_norm": 0.0,
|
1225 |
+
"learning_rate": 3.879010436799382e-05,
|
1226 |
+
"loss": 0.0,
|
1227 |
+
"step": 1740
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.6764592191727871,
|
1231 |
+
"grad_norm": 0.0,
|
1232 |
+
"learning_rate": 3.8725679680453554e-05,
|
1233 |
+
"loss": 0.0,
|
1234 |
+
"step": 1750
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.680324700425203,
|
1238 |
+
"grad_norm": 0.0,
|
1239 |
+
"learning_rate": 3.866125499291329e-05,
|
1240 |
+
"loss": 0.0,
|
1241 |
+
"step": 1760
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.6841901816776189,
|
1245 |
+
"grad_norm": 0.0,
|
1246 |
+
"learning_rate": 3.859683030537302e-05,
|
1247 |
+
"loss": 0.0,
|
1248 |
+
"step": 1770
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.6880556629300348,
|
1252 |
+
"grad_norm": 0.0,
|
1253 |
+
"learning_rate": 3.8532405617832755e-05,
|
1254 |
+
"loss": 0.0,
|
1255 |
+
"step": 1780
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.6919211441824508,
|
1259 |
+
"grad_norm": 0.0,
|
1260 |
+
"learning_rate": 3.846798093029249e-05,
|
1261 |
+
"loss": 0.0,
|
1262 |
+
"step": 1790
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.6957866254348667,
|
1266 |
+
"grad_norm": 0.0,
|
1267 |
+
"learning_rate": 3.840355624275223e-05,
|
1268 |
+
"loss": 0.0,
|
1269 |
+
"step": 1800
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.6996521066872826,
|
1273 |
+
"grad_norm": 0.0,
|
1274 |
+
"learning_rate": 3.833913155521196e-05,
|
1275 |
+
"loss": 0.0,
|
1276 |
+
"step": 1810
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.7035175879396985,
|
1280 |
+
"grad_norm": 0.0,
|
1281 |
+
"learning_rate": 3.8274706867671696e-05,
|
1282 |
+
"loss": 0.0,
|
1283 |
+
"step": 1820
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.7073830691921145,
|
1287 |
+
"grad_norm": 0.0,
|
1288 |
+
"learning_rate": 3.821028218013143e-05,
|
1289 |
+
"loss": 0.0,
|
1290 |
+
"step": 1830
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.7112485504445304,
|
1294 |
+
"grad_norm": 0.0,
|
1295 |
+
"learning_rate": 3.814585749259116e-05,
|
1296 |
+
"loss": 0.0,
|
1297 |
+
"step": 1840
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.7151140316969463,
|
1301 |
+
"grad_norm": 0.0,
|
1302 |
+
"learning_rate": 3.8081432805050896e-05,
|
1303 |
+
"loss": 0.0,
|
1304 |
+
"step": 1850
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.7189795129493622,
|
1308 |
+
"grad_norm": 0.0,
|
1309 |
+
"learning_rate": 3.801700811751063e-05,
|
1310 |
+
"loss": 0.0,
|
1311 |
+
"step": 1860
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.7228449942017782,
|
1315 |
+
"grad_norm": 0.0,
|
1316 |
+
"learning_rate": 3.795258342997037e-05,
|
1317 |
+
"loss": 0.0,
|
1318 |
+
"step": 1870
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.7267104754541941,
|
1322 |
+
"grad_norm": 0.0,
|
1323 |
+
"learning_rate": 3.7888158742430104e-05,
|
1324 |
+
"loss": 0.0,
|
1325 |
+
"step": 1880
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.73057595670661,
|
1329 |
+
"grad_norm": 0.0,
|
1330 |
+
"learning_rate": 3.782373405488984e-05,
|
1331 |
+
"loss": 0.0,
|
1332 |
+
"step": 1890
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.734441437959026,
|
1336 |
+
"grad_norm": 0.0,
|
1337 |
+
"learning_rate": 3.775930936734957e-05,
|
1338 |
+
"loss": 0.0,
|
1339 |
+
"step": 1900
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.7383069192114419,
|
1343 |
+
"grad_norm": 0.0,
|
1344 |
+
"learning_rate": 3.7694884679809304e-05,
|
1345 |
+
"loss": 0.0,
|
1346 |
+
"step": 1910
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.7421724004638578,
|
1350 |
+
"grad_norm": 0.0,
|
1351 |
+
"learning_rate": 3.763045999226904e-05,
|
1352 |
+
"loss": 0.0,
|
1353 |
+
"step": 1920
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.7460378817162737,
|
1357 |
+
"grad_norm": 0.0,
|
1358 |
+
"learning_rate": 3.756603530472878e-05,
|
1359 |
+
"loss": 0.0,
|
1360 |
+
"step": 1930
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.7499033629686896,
|
1364 |
+
"grad_norm": 0.0,
|
1365 |
+
"learning_rate": 3.750161061718851e-05,
|
1366 |
+
"loss": 0.0,
|
1367 |
+
"step": 1940
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.7537688442211056,
|
1371 |
+
"grad_norm": 0.0,
|
1372 |
+
"learning_rate": 3.7437185929648245e-05,
|
1373 |
+
"loss": 0.0,
|
1374 |
+
"step": 1950
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.7576343254735215,
|
1378 |
+
"grad_norm": 0.0,
|
1379 |
+
"learning_rate": 3.737276124210798e-05,
|
1380 |
+
"loss": 0.0,
|
1381 |
+
"step": 1960
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.7614998067259374,
|
1385 |
+
"grad_norm": 0.0,
|
1386 |
+
"learning_rate": 3.730833655456771e-05,
|
1387 |
+
"loss": 0.0,
|
1388 |
+
"step": 1970
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.7653652879783533,
|
1392 |
+
"grad_norm": 0.0,
|
1393 |
+
"learning_rate": 3.7243911867027446e-05,
|
1394 |
+
"loss": 0.0,
|
1395 |
+
"step": 1980
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.7692307692307693,
|
1399 |
+
"grad_norm": 0.0,
|
1400 |
+
"learning_rate": 3.717948717948718e-05,
|
1401 |
+
"loss": 0.0,
|
1402 |
+
"step": 1990
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.7730962504831852,
|
1406 |
+
"grad_norm": 0.0,
|
1407 |
+
"learning_rate": 3.711506249194692e-05,
|
1408 |
+
"loss": 0.0,
|
1409 |
+
"step": 2000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.7769617317356011,
|
1413 |
+
"grad_norm": 0.0,
|
1414 |
+
"learning_rate": 3.705063780440665e-05,
|
1415 |
+
"loss": 0.0,
|
1416 |
+
"step": 2010
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.780827212988017,
|
1420 |
+
"grad_norm": 0.0,
|
1421 |
+
"learning_rate": 3.6986213116866387e-05,
|
1422 |
+
"loss": 0.0,
|
1423 |
+
"step": 2020
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.784692694240433,
|
1427 |
+
"grad_norm": 0.0,
|
1428 |
+
"learning_rate": 3.692178842932612e-05,
|
1429 |
+
"loss": 0.0,
|
1430 |
+
"step": 2030
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.7885581754928489,
|
1434 |
+
"grad_norm": 0.0,
|
1435 |
+
"learning_rate": 3.6857363741785854e-05,
|
1436 |
+
"loss": 0.0,
|
1437 |
+
"step": 2040
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.7924236567452648,
|
1441 |
+
"grad_norm": 0.0,
|
1442 |
+
"learning_rate": 3.679293905424559e-05,
|
1443 |
+
"loss": 0.0,
|
1444 |
+
"step": 2050
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.7962891379976808,
|
1448 |
+
"grad_norm": 0.0,
|
1449 |
+
"learning_rate": 3.672851436670533e-05,
|
1450 |
+
"loss": 0.0,
|
1451 |
+
"step": 2060
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.8001546192500967,
|
1455 |
+
"grad_norm": 0.0,
|
1456 |
+
"learning_rate": 3.666408967916506e-05,
|
1457 |
+
"loss": 0.0,
|
1458 |
+
"step": 2070
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.8040201005025126,
|
1462 |
+
"grad_norm": 0.0,
|
1463 |
+
"learning_rate": 3.6599664991624795e-05,
|
1464 |
+
"loss": 0.0,
|
1465 |
+
"step": 2080
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.8078855817549285,
|
1469 |
+
"grad_norm": 0.0,
|
1470 |
+
"learning_rate": 3.653524030408453e-05,
|
1471 |
+
"loss": 0.0,
|
1472 |
+
"step": 2090
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.8117510630073445,
|
1476 |
+
"grad_norm": 0.0,
|
1477 |
+
"learning_rate": 3.647081561654426e-05,
|
1478 |
+
"loss": 0.0,
|
1479 |
+
"step": 2100
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.8156165442597604,
|
1483 |
+
"grad_norm": 0.0,
|
1484 |
+
"learning_rate": 3.6406390929003995e-05,
|
1485 |
+
"loss": 0.0,
|
1486 |
+
"step": 2110
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.8194820255121763,
|
1490 |
+
"grad_norm": 0.0,
|
1491 |
+
"learning_rate": 3.634196624146373e-05,
|
1492 |
+
"loss": 0.0,
|
1493 |
+
"step": 2120
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.8233475067645922,
|
1497 |
+
"grad_norm": 0.0,
|
1498 |
+
"learning_rate": 3.627754155392347e-05,
|
1499 |
+
"loss": 0.0,
|
1500 |
+
"step": 2130
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.8272129880170082,
|
1504 |
+
"grad_norm": 0.0,
|
1505 |
+
"learning_rate": 3.62131168663832e-05,
|
1506 |
+
"loss": 0.0,
|
1507 |
+
"step": 2140
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.8310784692694241,
|
1511 |
+
"grad_norm": 0.0,
|
1512 |
+
"learning_rate": 3.6148692178842936e-05,
|
1513 |
+
"loss": 0.0,
|
1514 |
+
"step": 2150
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.83494395052184,
|
1518 |
+
"grad_norm": 0.0,
|
1519 |
+
"learning_rate": 3.608426749130267e-05,
|
1520 |
+
"loss": 0.0,
|
1521 |
+
"step": 2160
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.8388094317742559,
|
1525 |
+
"grad_norm": 0.0,
|
1526 |
+
"learning_rate": 3.60198428037624e-05,
|
1527 |
+
"loss": 0.0,
|
1528 |
+
"step": 2170
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.8426749130266719,
|
1532 |
+
"grad_norm": 0.0,
|
1533 |
+
"learning_rate": 3.595541811622214e-05,
|
1534 |
+
"loss": 0.0,
|
1535 |
+
"step": 2180
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.8465403942790878,
|
1539 |
+
"grad_norm": 0.0,
|
1540 |
+
"learning_rate": 3.589099342868188e-05,
|
1541 |
+
"loss": 0.0,
|
1542 |
+
"step": 2190
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.8504058755315037,
|
1546 |
+
"grad_norm": 0.0,
|
1547 |
+
"learning_rate": 3.582656874114161e-05,
|
1548 |
+
"loss": 0.0,
|
1549 |
+
"step": 2200
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.8542713567839196,
|
1553 |
+
"grad_norm": 0.0,
|
1554 |
+
"learning_rate": 3.5762144053601344e-05,
|
1555 |
+
"loss": 0.0,
|
1556 |
+
"step": 2210
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.8581368380363356,
|
1560 |
+
"grad_norm": 0.0,
|
1561 |
+
"learning_rate": 3.569771936606108e-05,
|
1562 |
+
"loss": 0.0,
|
1563 |
+
"step": 2220
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.8620023192887515,
|
1567 |
+
"grad_norm": 0.0,
|
1568 |
+
"learning_rate": 3.563329467852081e-05,
|
1569 |
+
"loss": 0.0,
|
1570 |
+
"step": 2230
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.8658678005411674,
|
1574 |
+
"grad_norm": 0.0,
|
1575 |
+
"learning_rate": 3.5568869990980545e-05,
|
1576 |
+
"loss": 0.0,
|
1577 |
+
"step": 2240
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.8697332817935833,
|
1581 |
+
"grad_norm": 0.0,
|
1582 |
+
"learning_rate": 3.550444530344028e-05,
|
1583 |
+
"loss": 0.0,
|
1584 |
+
"step": 2250
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.8735987630459993,
|
1588 |
+
"grad_norm": 0.0,
|
1589 |
+
"learning_rate": 3.544002061590002e-05,
|
1590 |
+
"loss": 0.0,
|
1591 |
+
"step": 2260
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.8774642442984152,
|
1595 |
+
"grad_norm": 0.0,
|
1596 |
+
"learning_rate": 3.537559592835975e-05,
|
1597 |
+
"loss": 0.0,
|
1598 |
+
"step": 2270
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.8813297255508311,
|
1602 |
+
"grad_norm": 0.0,
|
1603 |
+
"learning_rate": 3.5311171240819486e-05,
|
1604 |
+
"loss": 0.0,
|
1605 |
+
"step": 2280
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.885195206803247,
|
1609 |
+
"grad_norm": 0.0,
|
1610 |
+
"learning_rate": 3.524674655327922e-05,
|
1611 |
+
"loss": 0.0,
|
1612 |
+
"step": 2290
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.889060688055663,
|
1616 |
+
"grad_norm": 0.0,
|
1617 |
+
"learning_rate": 3.518232186573895e-05,
|
1618 |
+
"loss": 0.0,
|
1619 |
+
"step": 2300
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.8929261693080789,
|
1623 |
+
"grad_norm": 0.0,
|
1624 |
+
"learning_rate": 3.5117897178198686e-05,
|
1625 |
+
"loss": 0.0,
|
1626 |
+
"step": 2310
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.8967916505604948,
|
1630 |
+
"grad_norm": 0.0,
|
1631 |
+
"learning_rate": 3.5053472490658427e-05,
|
1632 |
+
"loss": 0.0,
|
1633 |
+
"step": 2320
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.9006571318129107,
|
1637 |
+
"grad_norm": 0.0,
|
1638 |
+
"learning_rate": 3.498904780311816e-05,
|
1639 |
+
"loss": 0.0,
|
1640 |
+
"step": 2330
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.9045226130653267,
|
1644 |
+
"grad_norm": 0.0,
|
1645 |
+
"learning_rate": 3.4924623115577894e-05,
|
1646 |
+
"loss": 0.0,
|
1647 |
+
"step": 2340
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.9083880943177426,
|
1651 |
+
"grad_norm": 0.0,
|
1652 |
+
"learning_rate": 3.486019842803763e-05,
|
1653 |
+
"loss": 0.0,
|
1654 |
+
"step": 2350
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.9122535755701585,
|
1658 |
+
"grad_norm": 0.0,
|
1659 |
+
"learning_rate": 3.479577374049736e-05,
|
1660 |
+
"loss": 0.0,
|
1661 |
+
"step": 2360
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.9161190568225744,
|
1665 |
+
"grad_norm": 0.0,
|
1666 |
+
"learning_rate": 3.4731349052957094e-05,
|
1667 |
+
"loss": 0.0,
|
1668 |
+
"step": 2370
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.9199845380749904,
|
1672 |
+
"grad_norm": 0.0,
|
1673 |
+
"learning_rate": 3.466692436541683e-05,
|
1674 |
+
"loss": 0.0,
|
1675 |
+
"step": 2380
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.9238500193274063,
|
1679 |
+
"grad_norm": 0.0,
|
1680 |
+
"learning_rate": 3.460249967787657e-05,
|
1681 |
+
"loss": 0.0,
|
1682 |
+
"step": 2390
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.9277155005798222,
|
1686 |
+
"grad_norm": 0.0,
|
1687 |
+
"learning_rate": 3.45380749903363e-05,
|
1688 |
+
"loss": 0.0,
|
1689 |
+
"step": 2400
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.9315809818322381,
|
1693 |
+
"grad_norm": 0.0,
|
1694 |
+
"learning_rate": 3.4473650302796035e-05,
|
1695 |
+
"loss": 0.0,
|
1696 |
+
"step": 2410
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.9354464630846541,
|
1700 |
+
"grad_norm": 0.0,
|
1701 |
+
"learning_rate": 3.440922561525577e-05,
|
1702 |
+
"loss": 0.0,
|
1703 |
+
"step": 2420
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.93931194433707,
|
1707 |
+
"grad_norm": 0.0,
|
1708 |
+
"learning_rate": 3.43448009277155e-05,
|
1709 |
+
"loss": 0.0,
|
1710 |
+
"step": 2430
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.9431774255894859,
|
1714 |
+
"grad_norm": 0.0,
|
1715 |
+
"learning_rate": 3.4280376240175236e-05,
|
1716 |
+
"loss": 0.0,
|
1717 |
+
"step": 2440
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.9470429068419018,
|
1721 |
+
"grad_norm": 0.0,
|
1722 |
+
"learning_rate": 3.4215951552634976e-05,
|
1723 |
+
"loss": 0.0,
|
1724 |
+
"step": 2450
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.9509083880943178,
|
1728 |
+
"grad_norm": 0.0,
|
1729 |
+
"learning_rate": 3.415152686509471e-05,
|
1730 |
+
"loss": 0.0,
|
1731 |
+
"step": 2460
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 0.9547738693467337,
|
1735 |
+
"grad_norm": 0.0,
|
1736 |
+
"learning_rate": 3.408710217755444e-05,
|
1737 |
+
"loss": 0.0,
|
1738 |
+
"step": 2470
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 0.9586393505991496,
|
1742 |
+
"grad_norm": 0.0,
|
1743 |
+
"learning_rate": 3.402267749001418e-05,
|
1744 |
+
"loss": 0.0,
|
1745 |
+
"step": 2480
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.9625048318515655,
|
1749 |
+
"grad_norm": 0.0,
|
1750 |
+
"learning_rate": 3.395825280247391e-05,
|
1751 |
+
"loss": 0.0,
|
1752 |
+
"step": 2490
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 0.9663703131039815,
|
1756 |
+
"grad_norm": 0.0,
|
1757 |
+
"learning_rate": 3.3893828114933644e-05,
|
1758 |
+
"loss": 0.0,
|
1759 |
+
"step": 2500
|
1760 |
+
},
|
1761 |
+
{
|
1762 |
+
"epoch": 0.9702357943563974,
|
1763 |
+
"grad_norm": 0.0,
|
1764 |
+
"learning_rate": 3.382940342739338e-05,
|
1765 |
+
"loss": 0.0,
|
1766 |
+
"step": 2510
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 0.9741012756088133,
|
1770 |
+
"grad_norm": 0.0,
|
1771 |
+
"learning_rate": 3.376497873985312e-05,
|
1772 |
+
"loss": 0.0,
|
1773 |
+
"step": 2520
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 0.9779667568612292,
|
1777 |
+
"grad_norm": 0.0,
|
1778 |
+
"learning_rate": 3.370055405231285e-05,
|
1779 |
+
"loss": 0.0,
|
1780 |
+
"step": 2530
|
1781 |
+
},
|
1782 |
+
{
|
1783 |
+
"epoch": 0.9818322381136452,
|
1784 |
+
"grad_norm": 0.0,
|
1785 |
+
"learning_rate": 3.3636129364772585e-05,
|
1786 |
+
"loss": 0.0,
|
1787 |
+
"step": 2540
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.9856977193660611,
|
1791 |
+
"grad_norm": 0.0,
|
1792 |
+
"learning_rate": 3.357170467723232e-05,
|
1793 |
+
"loss": 0.0,
|
1794 |
+
"step": 2550
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 0.989563200618477,
|
1798 |
+
"grad_norm": 0.0,
|
1799 |
+
"learning_rate": 3.350727998969205e-05,
|
1800 |
+
"loss": 0.0,
|
1801 |
+
"step": 2560
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 0.993428681870893,
|
1805 |
+
"grad_norm": 0.0,
|
1806 |
+
"learning_rate": 3.3442855302151785e-05,
|
1807 |
+
"loss": 0.0,
|
1808 |
+
"step": 2570
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.9972941631233089,
|
1812 |
+
"grad_norm": 0.0,
|
1813 |
+
"learning_rate": 3.3378430614611526e-05,
|
1814 |
+
"loss": 0.0,
|
1815 |
+
"step": 2580
|
1816 |
+
}
|
1817 |
+
],
|
1818 |
+
"logging_steps": 10,
|
1819 |
+
"max_steps": 7761,
|
1820 |
+
"num_input_tokens_seen": 0,
|
1821 |
+
"num_train_epochs": 3,
|
1822 |
+
"save_steps": 10,
|
1823 |
+
"stateful_callbacks": {
|
1824 |
+
"TrainerControl": {
|
1825 |
+
"args": {
|
1826 |
+
"should_epoch_stop": false,
|
1827 |
+
"should_evaluate": false,
|
1828 |
+
"should_log": false,
|
1829 |
+
"should_save": true,
|
1830 |
+
"should_training_stop": false
|
1831 |
+
},
|
1832 |
+
"attributes": {}
|
1833 |
+
}
|
1834 |
+
},
|
1835 |
+
"total_flos": 0.0,
|
1836 |
+
"train_batch_size": 16,
|
1837 |
+
"trial_name": null,
|
1838 |
+
"trial_params": null
|
1839 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:767936e59f3e8337b149383e0c26422ce3573807770aff72f312af16437f53dc
|
3 |
+
size 5368
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|