Training in progress, step 2500, checkpoint
Browse files- checkpoint-2500/config.json +195 -0
- checkpoint-2500/generation_config.json +4 -0
- checkpoint-2500/model.safetensors +3 -0
- checkpoint-2500/optimizer.pt +3 -0
- checkpoint-2500/rng_state_0.pth +3 -0
- checkpoint-2500/rng_state_1.pth +3 -0
- checkpoint-2500/rng_state_2.pth +3 -0
- checkpoint-2500/rng_state_3.pth +3 -0
- checkpoint-2500/scheduler.pt +3 -0
- checkpoint-2500/trainer_state.json +1811 -0
- checkpoint-2500/training_args.bin +3 -0
checkpoint-2500/config.json
ADDED
@@ -0,0 +1,195 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"SrcProberForConditionalGeneration"
|
4 |
+
],
|
5 |
+
"asm_encoder_config": {
|
6 |
+
"_name_or_path": "",
|
7 |
+
"add_cross_attention": false,
|
8 |
+
"architectures": [
|
9 |
+
"LongelmForMaskedLMWithEdgePrediction"
|
10 |
+
],
|
11 |
+
"attention_probs_dropout_prob": 0.1,
|
12 |
+
"bad_words_ids": null,
|
13 |
+
"begin_suppress_tokens": null,
|
14 |
+
"block_size": 8,
|
15 |
+
"bos_token_id": 1,
|
16 |
+
"chunk_size_feed_forward": 0,
|
17 |
+
"classifier_dropout": null,
|
18 |
+
"cross_attention_hidden_size": null,
|
19 |
+
"decoder_start_token_id": null,
|
20 |
+
"diversity_penalty": 0.0,
|
21 |
+
"do_sample": false,
|
22 |
+
"early_stopping": false,
|
23 |
+
"encoder_no_repeat_ngram_size": 0,
|
24 |
+
"eos_token_id": 2,
|
25 |
+
"ep_add_linear_projection": true,
|
26 |
+
"exponential_decay_length_penalty": null,
|
27 |
+
"finetuning_task": null,
|
28 |
+
"forced_bos_token_id": null,
|
29 |
+
"forced_eos_token_id": null,
|
30 |
+
"global_memory_size": 1,
|
31 |
+
"hidden_act": "gelu",
|
32 |
+
"hidden_dropout_prob": 0.1,
|
33 |
+
"hidden_size": 768,
|
34 |
+
"id2label": {
|
35 |
+
"0": "LABEL_0",
|
36 |
+
"1": "LABEL_1"
|
37 |
+
},
|
38 |
+
"initializer_range": 0.02,
|
39 |
+
"intermediate_size": 3072,
|
40 |
+
"is_decoder": false,
|
41 |
+
"is_encoder_decoder": false,
|
42 |
+
"label2id": {
|
43 |
+
"LABEL_0": 0,
|
44 |
+
"LABEL_1": 1
|
45 |
+
},
|
46 |
+
"layer_norm_eps": 1e-12,
|
47 |
+
"length_penalty": 1.0,
|
48 |
+
"max_blocks": 200,
|
49 |
+
"max_length": 20,
|
50 |
+
"max_position_embeddings": 1805,
|
51 |
+
"max_relative_position_embeddings": 8,
|
52 |
+
"min_length": 0,
|
53 |
+
"model_type": "longelm",
|
54 |
+
"no_repeat_ngram_size": 0,
|
55 |
+
"node_size": 1,
|
56 |
+
"num_attention_heads": 12,
|
57 |
+
"num_beam_groups": 1,
|
58 |
+
"num_beams": 1,
|
59 |
+
"num_hidden_layers": 12,
|
60 |
+
"num_return_sequences": 1,
|
61 |
+
"output_attentions": false,
|
62 |
+
"output_hidden_states": false,
|
63 |
+
"output_scores": false,
|
64 |
+
"pad_token_id": 3,
|
65 |
+
"position_embedding_type": "mixed",
|
66 |
+
"prefix": null,
|
67 |
+
"problem_type": null,
|
68 |
+
"pruned_heads": {},
|
69 |
+
"remove_invalid_values": false,
|
70 |
+
"repetition_penalty": 1.0,
|
71 |
+
"return_dict": true,
|
72 |
+
"return_dict_in_generate": false,
|
73 |
+
"sep_token_id": null,
|
74 |
+
"suppress_tokens": null,
|
75 |
+
"task_specific_params": null,
|
76 |
+
"temperature": 1.0,
|
77 |
+
"tf_legacy_loss": false,
|
78 |
+
"tie_encoder_decoder": false,
|
79 |
+
"tie_word_embeddings": true,
|
80 |
+
"tokenizer_class": null,
|
81 |
+
"top_k": 50,
|
82 |
+
"top_p": 1.0,
|
83 |
+
"torch_dtype": "float32",
|
84 |
+
"torchscript": false,
|
85 |
+
"type_vocab_size": 2,
|
86 |
+
"typical_p": 1.0,
|
87 |
+
"use_bfloat16": false,
|
88 |
+
"use_cache": true,
|
89 |
+
"vocab_size": 50265
|
90 |
+
},
|
91 |
+
"asm_feature_select_strategy": "nodes",
|
92 |
+
"asm_token_index": 32016,
|
93 |
+
"ignore_index": -100,
|
94 |
+
"model_type": "rad-prober",
|
95 |
+
"pad_token_id": 2,
|
96 |
+
"projector_hidden_act": "gelu",
|
97 |
+
"src_lm_config": {
|
98 |
+
"_name_or_path": "codellama/CodeLlama-7b-hf",
|
99 |
+
"add_cross_attention": false,
|
100 |
+
"architectures": [
|
101 |
+
"LlamaForCausalLM"
|
102 |
+
],
|
103 |
+
"attention_bias": false,
|
104 |
+
"attention_dropout": 0.0,
|
105 |
+
"bad_words_ids": null,
|
106 |
+
"begin_suppress_tokens": null,
|
107 |
+
"bos_token_id": 1,
|
108 |
+
"chunk_size_feed_forward": 0,
|
109 |
+
"cross_attention_hidden_size": null,
|
110 |
+
"decoder_start_token_id": null,
|
111 |
+
"diversity_penalty": 0.0,
|
112 |
+
"do_sample": false,
|
113 |
+
"early_stopping": false,
|
114 |
+
"encoder_no_repeat_ngram_size": 0,
|
115 |
+
"eos_token_id": 2,
|
116 |
+
"exponential_decay_length_penalty": null,
|
117 |
+
"finetuning_task": null,
|
118 |
+
"forced_bos_token_id": null,
|
119 |
+
"forced_eos_token_id": null,
|
120 |
+
"hidden_act": "silu",
|
121 |
+
"hidden_size": 4096,
|
122 |
+
"id2label": {
|
123 |
+
"0": "LABEL_0",
|
124 |
+
"1": "LABEL_1"
|
125 |
+
},
|
126 |
+
"initializer_range": 0.02,
|
127 |
+
"intermediate_size": 11008,
|
128 |
+
"is_decoder": false,
|
129 |
+
"is_encoder_decoder": false,
|
130 |
+
"label2id": {
|
131 |
+
"LABEL_0": 0,
|
132 |
+
"LABEL_1": 1
|
133 |
+
},
|
134 |
+
"length_penalty": 1.0,
|
135 |
+
"max_length": 20,
|
136 |
+
"max_position_embeddings": 16384,
|
137 |
+
"min_length": 0,
|
138 |
+
"model_type": "llama",
|
139 |
+
"no_repeat_ngram_size": 0,
|
140 |
+
"num_attention_heads": 32,
|
141 |
+
"num_beam_groups": 1,
|
142 |
+
"num_beams": 1,
|
143 |
+
"num_hidden_layers": 32,
|
144 |
+
"num_key_value_heads": 32,
|
145 |
+
"num_return_sequences": 1,
|
146 |
+
"output_attentions": false,
|
147 |
+
"output_hidden_states": false,
|
148 |
+
"output_scores": false,
|
149 |
+
"pad_token_id": null,
|
150 |
+
"prefix": null,
|
151 |
+
"pretraining_tp": 1,
|
152 |
+
"problem_type": null,
|
153 |
+
"pruned_heads": {},
|
154 |
+
"quantization_config": {
|
155 |
+
"_load_in_4bit": false,
|
156 |
+
"_load_in_8bit": true,
|
157 |
+
"bnb_4bit_compute_dtype": "float32",
|
158 |
+
"bnb_4bit_quant_storage": "uint8",
|
159 |
+
"bnb_4bit_quant_type": "fp4",
|
160 |
+
"bnb_4bit_use_double_quant": false,
|
161 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
162 |
+
"llm_int8_has_fp16_weight": false,
|
163 |
+
"llm_int8_skip_modules": null,
|
164 |
+
"llm_int8_threshold": 6.0,
|
165 |
+
"load_in_4bit": false,
|
166 |
+
"load_in_8bit": true,
|
167 |
+
"quant_method": "bitsandbytes"
|
168 |
+
},
|
169 |
+
"remove_invalid_values": false,
|
170 |
+
"repetition_penalty": 1.0,
|
171 |
+
"return_dict": true,
|
172 |
+
"return_dict_in_generate": false,
|
173 |
+
"rms_norm_eps": 1e-05,
|
174 |
+
"rope_scaling": null,
|
175 |
+
"rope_theta": 1000000,
|
176 |
+
"sep_token_id": null,
|
177 |
+
"suppress_tokens": null,
|
178 |
+
"task_specific_params": null,
|
179 |
+
"temperature": 1.0,
|
180 |
+
"tf_legacy_loss": false,
|
181 |
+
"tie_encoder_decoder": false,
|
182 |
+
"tie_word_embeddings": false,
|
183 |
+
"tokenizer_class": null,
|
184 |
+
"top_k": 50,
|
185 |
+
"top_p": 1.0,
|
186 |
+
"torch_dtype": "bfloat16",
|
187 |
+
"torchscript": false,
|
188 |
+
"typical_p": 1.0,
|
189 |
+
"use_bfloat16": false,
|
190 |
+
"use_cache": true,
|
191 |
+
"vocab_size": 32017
|
192 |
+
},
|
193 |
+
"torch_dtype": "float32",
|
194 |
+
"transformers_version": "4.39.3"
|
195 |
+
}
|
checkpoint-2500/generation_config.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"transformers_version": "4.39.3"
|
4 |
+
}
|
checkpoint-2500/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8d79411108c29ac7abb7fe79b0258c18de005d62107f2e8059d33bf8ea35cb17
|
3 |
+
size 579945600
|
checkpoint-2500/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:35e92eff9348b22b78d8f8f543cd794eaa8f4fce579652c3909a116b0f5f0049
|
3 |
+
size 216170911
|
checkpoint-2500/rng_state_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3102857b3c68df63bbf867a076d2c4deccb0db0d5944dfc306f9aea1b6a2b2d6
|
3 |
+
size 14960
|
checkpoint-2500/rng_state_1.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a094be5cfdfe7d4fa8a6b199872c19b470489b54c6a3dadfd6f39820f5f57981
|
3 |
+
size 14960
|
checkpoint-2500/rng_state_2.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bdaf3c546f0cb4d1d97a798d039c6cee56633636efe1412d5cb09b113f8c6ca3
|
3 |
+
size 14960
|
checkpoint-2500/rng_state_3.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:313853de7d0ccc7ddd3484fb15b3651df3f635c5bf57f9df46a4197b9f7450bf
|
3 |
+
size 14960
|
checkpoint-2500/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:217b570de5d0c74d5ca03e67a272484fa3558e51484a48ffe171076cc43984b4
|
3 |
+
size 1064
|
checkpoint-2500/trainer_state.json
ADDED
@@ -0,0 +1,1811 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.208313194780087,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 2500,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"grad_norm": 75.8236083984375,
|
14 |
+
"learning_rate": 4.0000000000000003e-07,
|
15 |
+
"loss": 3.7148,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.01,
|
20 |
+
"grad_norm": 52.23726272583008,
|
21 |
+
"learning_rate": 9e-07,
|
22 |
+
"loss": 3.2721,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.01,
|
27 |
+
"grad_norm": 35.391998291015625,
|
28 |
+
"learning_rate": 1.4000000000000001e-06,
|
29 |
+
"loss": 2.4262,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.02,
|
34 |
+
"grad_norm": 15.225934028625488,
|
35 |
+
"learning_rate": 1.9e-06,
|
36 |
+
"loss": 1.8828,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.02,
|
41 |
+
"grad_norm": 13.488654136657715,
|
42 |
+
"learning_rate": 2.4000000000000003e-06,
|
43 |
+
"loss": 1.5398,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.03,
|
48 |
+
"grad_norm": 8.773133277893066,
|
49 |
+
"learning_rate": 2.8500000000000002e-06,
|
50 |
+
"loss": 1.2772,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.03,
|
55 |
+
"grad_norm": 6.736795902252197,
|
56 |
+
"learning_rate": 3.3500000000000005e-06,
|
57 |
+
"loss": 1.1159,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.04,
|
62 |
+
"grad_norm": 2.9463019371032715,
|
63 |
+
"learning_rate": 3.85e-06,
|
64 |
+
"loss": 0.9883,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.04,
|
69 |
+
"grad_norm": 1.3851969242095947,
|
70 |
+
"learning_rate": 4.35e-06,
|
71 |
+
"loss": 0.902,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.05,
|
76 |
+
"grad_norm": 1.4642181396484375,
|
77 |
+
"learning_rate": 4.85e-06,
|
78 |
+
"loss": 0.8701,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.05,
|
83 |
+
"grad_norm": 0.6514687538146973,
|
84 |
+
"learning_rate": 5.3500000000000004e-06,
|
85 |
+
"loss": 0.8548,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.06,
|
90 |
+
"grad_norm": 0.4448617994785309,
|
91 |
+
"learning_rate": 5.850000000000001e-06,
|
92 |
+
"loss": 0.8038,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.06,
|
97 |
+
"grad_norm": 0.5425733327865601,
|
98 |
+
"learning_rate": 6.35e-06,
|
99 |
+
"loss": 0.825,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.07,
|
104 |
+
"grad_norm": 0.47352057695388794,
|
105 |
+
"learning_rate": 6.8500000000000005e-06,
|
106 |
+
"loss": 0.812,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.07,
|
111 |
+
"grad_norm": 0.40401744842529297,
|
112 |
+
"learning_rate": 7.35e-06,
|
113 |
+
"loss": 0.7912,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.08,
|
118 |
+
"grad_norm": 0.3112718164920807,
|
119 |
+
"learning_rate": 7.850000000000001e-06,
|
120 |
+
"loss": 0.8057,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.08,
|
125 |
+
"grad_norm": 0.45351892709732056,
|
126 |
+
"learning_rate": 8.350000000000001e-06,
|
127 |
+
"loss": 0.7912,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.09,
|
132 |
+
"grad_norm": 0.38080641627311707,
|
133 |
+
"learning_rate": 8.85e-06,
|
134 |
+
"loss": 0.7956,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.09,
|
139 |
+
"grad_norm": 0.22949431836605072,
|
140 |
+
"learning_rate": 9.35e-06,
|
141 |
+
"loss": 0.7785,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.1,
|
146 |
+
"grad_norm": 0.8331744074821472,
|
147 |
+
"learning_rate": 9.85e-06,
|
148 |
+
"loss": 0.7769,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.1,
|
153 |
+
"grad_norm": 0.30213871598243713,
|
154 |
+
"learning_rate": 1.035e-05,
|
155 |
+
"loss": 0.7797,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.11,
|
160 |
+
"grad_norm": 0.32015591859817505,
|
161 |
+
"learning_rate": 1.0850000000000001e-05,
|
162 |
+
"loss": 0.7941,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.11,
|
167 |
+
"grad_norm": 0.22447015345096588,
|
168 |
+
"learning_rate": 1.1350000000000001e-05,
|
169 |
+
"loss": 0.8014,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.12,
|
174 |
+
"grad_norm": 0.2546006143093109,
|
175 |
+
"learning_rate": 1.185e-05,
|
176 |
+
"loss": 0.771,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.12,
|
181 |
+
"grad_norm": 0.23833389580249786,
|
182 |
+
"learning_rate": 1.235e-05,
|
183 |
+
"loss": 0.7956,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.13,
|
188 |
+
"grad_norm": 0.1831669956445694,
|
189 |
+
"learning_rate": 1.285e-05,
|
190 |
+
"loss": 0.7833,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.13,
|
195 |
+
"grad_norm": 0.19860319793224335,
|
196 |
+
"learning_rate": 1.3350000000000001e-05,
|
197 |
+
"loss": 0.7956,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.14,
|
202 |
+
"grad_norm": 0.21220380067825317,
|
203 |
+
"learning_rate": 1.3850000000000001e-05,
|
204 |
+
"loss": 0.7942,
|
205 |
+
"step": 280
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.14,
|
209 |
+
"grad_norm": 0.26025038957595825,
|
210 |
+
"learning_rate": 1.435e-05,
|
211 |
+
"loss": 0.7877,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.14,
|
216 |
+
"grad_norm": 0.2029470056295395,
|
217 |
+
"learning_rate": 1.485e-05,
|
218 |
+
"loss": 0.7637,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.15,
|
223 |
+
"grad_norm": 0.21639183163642883,
|
224 |
+
"learning_rate": 1.535e-05,
|
225 |
+
"loss": 0.7885,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.15,
|
230 |
+
"grad_norm": 0.23670387268066406,
|
231 |
+
"learning_rate": 1.5850000000000002e-05,
|
232 |
+
"loss": 0.7558,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.16,
|
237 |
+
"grad_norm": 0.2855335772037506,
|
238 |
+
"learning_rate": 1.635e-05,
|
239 |
+
"loss": 0.7854,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.16,
|
244 |
+
"grad_norm": 0.21515759825706482,
|
245 |
+
"learning_rate": 1.6850000000000003e-05,
|
246 |
+
"loss": 0.7577,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.17,
|
251 |
+
"grad_norm": 0.18981383740901947,
|
252 |
+
"learning_rate": 1.7349999999999998e-05,
|
253 |
+
"loss": 0.7802,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.17,
|
258 |
+
"grad_norm": 0.2674611508846283,
|
259 |
+
"learning_rate": 1.785e-05,
|
260 |
+
"loss": 0.7809,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.18,
|
265 |
+
"grad_norm": 0.24821774661540985,
|
266 |
+
"learning_rate": 1.8350000000000002e-05,
|
267 |
+
"loss": 0.7965,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.18,
|
272 |
+
"grad_norm": 0.3644929528236389,
|
273 |
+
"learning_rate": 1.885e-05,
|
274 |
+
"loss": 0.7957,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.19,
|
279 |
+
"grad_norm": 0.3568972647190094,
|
280 |
+
"learning_rate": 1.9350000000000003e-05,
|
281 |
+
"loss": 0.7882,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.19,
|
286 |
+
"grad_norm": 0.3236319422721863,
|
287 |
+
"learning_rate": 1.985e-05,
|
288 |
+
"loss": 0.7841,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.2,
|
293 |
+
"grad_norm": 0.24553723633289337,
|
294 |
+
"learning_rate": 2.035e-05,
|
295 |
+
"loss": 0.7824,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.2,
|
300 |
+
"grad_norm": 0.1855400949716568,
|
301 |
+
"learning_rate": 2.085e-05,
|
302 |
+
"loss": 0.7645,
|
303 |
+
"step": 420
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.21,
|
307 |
+
"grad_norm": 0.26212435960769653,
|
308 |
+
"learning_rate": 2.135e-05,
|
309 |
+
"loss": 0.7824,
|
310 |
+
"step": 430
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.21,
|
314 |
+
"grad_norm": 0.3167509138584137,
|
315 |
+
"learning_rate": 2.1850000000000003e-05,
|
316 |
+
"loss": 0.7908,
|
317 |
+
"step": 440
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.22,
|
321 |
+
"grad_norm": 0.21845023334026337,
|
322 |
+
"learning_rate": 2.235e-05,
|
323 |
+
"loss": 0.7888,
|
324 |
+
"step": 450
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.22,
|
328 |
+
"grad_norm": 0.30318892002105713,
|
329 |
+
"learning_rate": 2.2850000000000003e-05,
|
330 |
+
"loss": 0.7682,
|
331 |
+
"step": 460
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.23,
|
335 |
+
"grad_norm": 0.28828638792037964,
|
336 |
+
"learning_rate": 2.3350000000000002e-05,
|
337 |
+
"loss": 0.7588,
|
338 |
+
"step": 470
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.23,
|
342 |
+
"grad_norm": 0.2422240674495697,
|
343 |
+
"learning_rate": 2.385e-05,
|
344 |
+
"loss": 0.7787,
|
345 |
+
"step": 480
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.24,
|
349 |
+
"grad_norm": 0.3579326272010803,
|
350 |
+
"learning_rate": 2.435e-05,
|
351 |
+
"loss": 0.7932,
|
352 |
+
"step": 490
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.24,
|
356 |
+
"grad_norm": 0.3549540042877197,
|
357 |
+
"learning_rate": 2.485e-05,
|
358 |
+
"loss": 0.7634,
|
359 |
+
"step": 500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.24,
|
363 |
+
"eval_loss": 0.7744565606117249,
|
364 |
+
"eval_runtime": 43.9626,
|
365 |
+
"eval_samples_per_second": 45.493,
|
366 |
+
"eval_steps_per_second": 0.364,
|
367 |
+
"step": 500
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 0.25,
|
371 |
+
"grad_norm": 0.2931281328201294,
|
372 |
+
"learning_rate": 2.5350000000000003e-05,
|
373 |
+
"loss": 0.7763,
|
374 |
+
"step": 510
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.25,
|
378 |
+
"grad_norm": 0.21228083968162537,
|
379 |
+
"learning_rate": 2.585e-05,
|
380 |
+
"loss": 0.7748,
|
381 |
+
"step": 520
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 0.26,
|
385 |
+
"grad_norm": 0.26342302560806274,
|
386 |
+
"learning_rate": 2.6350000000000004e-05,
|
387 |
+
"loss": 0.789,
|
388 |
+
"step": 530
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 0.26,
|
392 |
+
"grad_norm": 0.2840302288532257,
|
393 |
+
"learning_rate": 2.6850000000000002e-05,
|
394 |
+
"loss": 0.768,
|
395 |
+
"step": 540
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 0.27,
|
399 |
+
"grad_norm": 0.273771196603775,
|
400 |
+
"learning_rate": 2.7350000000000004e-05,
|
401 |
+
"loss": 0.7707,
|
402 |
+
"step": 550
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 0.27,
|
406 |
+
"grad_norm": 1.6022837162017822,
|
407 |
+
"learning_rate": 2.7850000000000003e-05,
|
408 |
+
"loss": 0.7705,
|
409 |
+
"step": 560
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 0.28,
|
413 |
+
"grad_norm": 0.3082919716835022,
|
414 |
+
"learning_rate": 2.8349999999999998e-05,
|
415 |
+
"loss": 0.7849,
|
416 |
+
"step": 570
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.28,
|
420 |
+
"grad_norm": 0.25341132283210754,
|
421 |
+
"learning_rate": 2.885e-05,
|
422 |
+
"loss": 0.7658,
|
423 |
+
"step": 580
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 0.29,
|
427 |
+
"grad_norm": 0.2250708043575287,
|
428 |
+
"learning_rate": 2.935e-05,
|
429 |
+
"loss": 0.756,
|
430 |
+
"step": 590
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 0.29,
|
434 |
+
"grad_norm": 0.23522531986236572,
|
435 |
+
"learning_rate": 2.985e-05,
|
436 |
+
"loss": 0.7692,
|
437 |
+
"step": 600
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 0.29,
|
441 |
+
"grad_norm": 0.2448827624320984,
|
442 |
+
"learning_rate": 3.035e-05,
|
443 |
+
"loss": 0.7635,
|
444 |
+
"step": 610
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 0.3,
|
448 |
+
"grad_norm": 0.19717586040496826,
|
449 |
+
"learning_rate": 3.0850000000000004e-05,
|
450 |
+
"loss": 0.7548,
|
451 |
+
"step": 620
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 0.3,
|
455 |
+
"grad_norm": 0.24299098551273346,
|
456 |
+
"learning_rate": 3.135e-05,
|
457 |
+
"loss": 0.7673,
|
458 |
+
"step": 630
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.31,
|
462 |
+
"grad_norm": 0.2555953562259674,
|
463 |
+
"learning_rate": 3.185e-05,
|
464 |
+
"loss": 0.7528,
|
465 |
+
"step": 640
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 0.31,
|
469 |
+
"grad_norm": 0.2102127969264984,
|
470 |
+
"learning_rate": 3.235e-05,
|
471 |
+
"loss": 0.7678,
|
472 |
+
"step": 650
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 0.32,
|
476 |
+
"grad_norm": 0.21767646074295044,
|
477 |
+
"learning_rate": 3.2850000000000006e-05,
|
478 |
+
"loss": 0.7564,
|
479 |
+
"step": 660
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 0.32,
|
483 |
+
"grad_norm": 0.3582068085670471,
|
484 |
+
"learning_rate": 3.3350000000000004e-05,
|
485 |
+
"loss": 0.7645,
|
486 |
+
"step": 670
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 0.33,
|
490 |
+
"grad_norm": 0.27804431319236755,
|
491 |
+
"learning_rate": 3.385e-05,
|
492 |
+
"loss": 0.7645,
|
493 |
+
"step": 680
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 0.33,
|
497 |
+
"grad_norm": 0.2464047521352768,
|
498 |
+
"learning_rate": 3.435e-05,
|
499 |
+
"loss": 0.7325,
|
500 |
+
"step": 690
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 0.34,
|
504 |
+
"grad_norm": 0.22323842346668243,
|
505 |
+
"learning_rate": 3.485e-05,
|
506 |
+
"loss": 0.7579,
|
507 |
+
"step": 700
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 0.34,
|
511 |
+
"grad_norm": 0.28329703211784363,
|
512 |
+
"learning_rate": 3.535e-05,
|
513 |
+
"loss": 0.7399,
|
514 |
+
"step": 710
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 0.35,
|
518 |
+
"grad_norm": 0.2471865713596344,
|
519 |
+
"learning_rate": 3.585e-05,
|
520 |
+
"loss": 0.765,
|
521 |
+
"step": 720
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 0.35,
|
525 |
+
"grad_norm": 0.314908504486084,
|
526 |
+
"learning_rate": 3.635e-05,
|
527 |
+
"loss": 0.7529,
|
528 |
+
"step": 730
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 0.36,
|
532 |
+
"grad_norm": 0.21466995775699615,
|
533 |
+
"learning_rate": 3.685e-05,
|
534 |
+
"loss": 0.7365,
|
535 |
+
"step": 740
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 0.36,
|
539 |
+
"grad_norm": 0.2545580565929413,
|
540 |
+
"learning_rate": 3.735e-05,
|
541 |
+
"loss": 0.7458,
|
542 |
+
"step": 750
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 0.37,
|
546 |
+
"grad_norm": 0.2651059329509735,
|
547 |
+
"learning_rate": 3.7850000000000005e-05,
|
548 |
+
"loss": 0.7578,
|
549 |
+
"step": 760
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 0.37,
|
553 |
+
"grad_norm": 0.2408379763364792,
|
554 |
+
"learning_rate": 3.8350000000000004e-05,
|
555 |
+
"loss": 0.7283,
|
556 |
+
"step": 770
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 0.38,
|
560 |
+
"grad_norm": 0.2575875520706177,
|
561 |
+
"learning_rate": 3.885e-05,
|
562 |
+
"loss": 0.7367,
|
563 |
+
"step": 780
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 0.38,
|
567 |
+
"grad_norm": 0.28182050585746765,
|
568 |
+
"learning_rate": 3.935e-05,
|
569 |
+
"loss": 0.7784,
|
570 |
+
"step": 790
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 0.39,
|
574 |
+
"grad_norm": 0.20579397678375244,
|
575 |
+
"learning_rate": 3.9850000000000006e-05,
|
576 |
+
"loss": 0.7533,
|
577 |
+
"step": 800
|
578 |
+
},
|
579 |
+
{
|
580 |
+
"epoch": 0.39,
|
581 |
+
"grad_norm": 0.1906532198190689,
|
582 |
+
"learning_rate": 4.0350000000000005e-05,
|
583 |
+
"loss": 0.7544,
|
584 |
+
"step": 810
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 0.4,
|
588 |
+
"grad_norm": 0.18311142921447754,
|
589 |
+
"learning_rate": 4.085e-05,
|
590 |
+
"loss": 0.75,
|
591 |
+
"step": 820
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 0.4,
|
595 |
+
"grad_norm": 0.1910839080810547,
|
596 |
+
"learning_rate": 4.135e-05,
|
597 |
+
"loss": 0.7538,
|
598 |
+
"step": 830
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 0.41,
|
602 |
+
"grad_norm": 0.2178005427122116,
|
603 |
+
"learning_rate": 4.185e-05,
|
604 |
+
"loss": 0.7456,
|
605 |
+
"step": 840
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 0.41,
|
609 |
+
"grad_norm": 0.27642378211021423,
|
610 |
+
"learning_rate": 4.235e-05,
|
611 |
+
"loss": 0.744,
|
612 |
+
"step": 850
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 0.42,
|
616 |
+
"grad_norm": 0.26685312390327454,
|
617 |
+
"learning_rate": 4.285e-05,
|
618 |
+
"loss": 0.7611,
|
619 |
+
"step": 860
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 0.42,
|
623 |
+
"grad_norm": 0.2021845579147339,
|
624 |
+
"learning_rate": 4.335e-05,
|
625 |
+
"loss": 0.758,
|
626 |
+
"step": 870
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 0.43,
|
630 |
+
"grad_norm": 0.20516808331012726,
|
631 |
+
"learning_rate": 4.385e-05,
|
632 |
+
"loss": 0.7404,
|
633 |
+
"step": 880
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 0.43,
|
637 |
+
"grad_norm": 0.20619668066501617,
|
638 |
+
"learning_rate": 4.435e-05,
|
639 |
+
"loss": 0.7379,
|
640 |
+
"step": 890
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 0.43,
|
644 |
+
"grad_norm": 0.22439967095851898,
|
645 |
+
"learning_rate": 4.4850000000000006e-05,
|
646 |
+
"loss": 0.7495,
|
647 |
+
"step": 900
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 0.44,
|
651 |
+
"grad_norm": 0.2192295491695404,
|
652 |
+
"learning_rate": 4.5350000000000005e-05,
|
653 |
+
"loss": 0.7377,
|
654 |
+
"step": 910
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 0.44,
|
658 |
+
"grad_norm": 0.26503369212150574,
|
659 |
+
"learning_rate": 4.585e-05,
|
660 |
+
"loss": 0.7481,
|
661 |
+
"step": 920
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 0.45,
|
665 |
+
"grad_norm": 0.2249353975057602,
|
666 |
+
"learning_rate": 4.635e-05,
|
667 |
+
"loss": 0.742,
|
668 |
+
"step": 930
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 0.45,
|
672 |
+
"grad_norm": 0.21413160860538483,
|
673 |
+
"learning_rate": 4.685000000000001e-05,
|
674 |
+
"loss": 0.7485,
|
675 |
+
"step": 940
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 0.46,
|
679 |
+
"grad_norm": 0.20971478521823883,
|
680 |
+
"learning_rate": 4.735e-05,
|
681 |
+
"loss": 0.7339,
|
682 |
+
"step": 950
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 0.46,
|
686 |
+
"grad_norm": 0.20295798778533936,
|
687 |
+
"learning_rate": 4.785e-05,
|
688 |
+
"loss": 0.722,
|
689 |
+
"step": 960
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.47,
|
693 |
+
"grad_norm": 0.1957596242427826,
|
694 |
+
"learning_rate": 4.835e-05,
|
695 |
+
"loss": 0.7538,
|
696 |
+
"step": 970
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 0.47,
|
700 |
+
"grad_norm": 0.18014560639858246,
|
701 |
+
"learning_rate": 4.885e-05,
|
702 |
+
"loss": 0.7251,
|
703 |
+
"step": 980
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 0.48,
|
707 |
+
"grad_norm": 0.19334861636161804,
|
708 |
+
"learning_rate": 4.935e-05,
|
709 |
+
"loss": 0.7444,
|
710 |
+
"step": 990
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 0.48,
|
714 |
+
"grad_norm": 0.1753440499305725,
|
715 |
+
"learning_rate": 4.9850000000000006e-05,
|
716 |
+
"loss": 0.7376,
|
717 |
+
"step": 1000
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 0.48,
|
721 |
+
"eval_loss": 0.73988938331604,
|
722 |
+
"eval_runtime": 42.9535,
|
723 |
+
"eval_samples_per_second": 46.562,
|
724 |
+
"eval_steps_per_second": 0.372,
|
725 |
+
"step": 1000
|
726 |
+
},
|
727 |
+
{
|
728 |
+
"epoch": 0.49,
|
729 |
+
"grad_norm": 0.24455028772354126,
|
730 |
+
"learning_rate": 4.99999307775404e-05,
|
731 |
+
"loss": 0.7404,
|
732 |
+
"step": 1010
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 0.49,
|
736 |
+
"grad_norm": 0.2149570733308792,
|
737 |
+
"learning_rate": 4.999959172968145e-05,
|
738 |
+
"loss": 0.7342,
|
739 |
+
"step": 1020
|
740 |
+
},
|
741 |
+
{
|
742 |
+
"epoch": 0.5,
|
743 |
+
"grad_norm": 0.1931283175945282,
|
744 |
+
"learning_rate": 4.999897014592085e-05,
|
745 |
+
"loss": 0.735,
|
746 |
+
"step": 1030
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 0.5,
|
750 |
+
"grad_norm": 0.2331906110048294,
|
751 |
+
"learning_rate": 4.999806603328352e-05,
|
752 |
+
"loss": 0.7145,
|
753 |
+
"step": 1040
|
754 |
+
},
|
755 |
+
{
|
756 |
+
"epoch": 0.51,
|
757 |
+
"grad_norm": 0.21177324652671814,
|
758 |
+
"learning_rate": 4.999687940198738e-05,
|
759 |
+
"loss": 0.7308,
|
760 |
+
"step": 1050
|
761 |
+
},
|
762 |
+
{
|
763 |
+
"epoch": 0.51,
|
764 |
+
"grad_norm": 0.2026621401309967,
|
765 |
+
"learning_rate": 4.9995410265443286e-05,
|
766 |
+
"loss": 0.7445,
|
767 |
+
"step": 1060
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 0.52,
|
771 |
+
"grad_norm": 0.19811251759529114,
|
772 |
+
"learning_rate": 4.999365864025487e-05,
|
773 |
+
"loss": 0.738,
|
774 |
+
"step": 1070
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 0.52,
|
778 |
+
"grad_norm": 0.1981566846370697,
|
779 |
+
"learning_rate": 4.9991624546218334e-05,
|
780 |
+
"loss": 0.7286,
|
781 |
+
"step": 1080
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 0.53,
|
785 |
+
"grad_norm": 0.17122291028499603,
|
786 |
+
"learning_rate": 4.9989308006322235e-05,
|
787 |
+
"loss": 0.737,
|
788 |
+
"step": 1090
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 0.53,
|
792 |
+
"grad_norm": 0.2182038575410843,
|
793 |
+
"learning_rate": 4.9986709046747225e-05,
|
794 |
+
"loss": 0.7354,
|
795 |
+
"step": 1100
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 0.54,
|
799 |
+
"grad_norm": 0.1823730617761612,
|
800 |
+
"learning_rate": 4.9983827696865764e-05,
|
801 |
+
"loss": 0.725,
|
802 |
+
"step": 1110
|
803 |
+
},
|
804 |
+
{
|
805 |
+
"epoch": 0.54,
|
806 |
+
"grad_norm": 0.2195768803358078,
|
807 |
+
"learning_rate": 4.998066398924177e-05,
|
808 |
+
"loss": 0.7173,
|
809 |
+
"step": 1120
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 0.55,
|
813 |
+
"grad_norm": 0.18349726498126984,
|
814 |
+
"learning_rate": 4.997721795963028e-05,
|
815 |
+
"loss": 0.7159,
|
816 |
+
"step": 1130
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 0.55,
|
820 |
+
"grad_norm": 0.2440386265516281,
|
821 |
+
"learning_rate": 4.997348964697699e-05,
|
822 |
+
"loss": 0.7171,
|
823 |
+
"step": 1140
|
824 |
+
},
|
825 |
+
{
|
826 |
+
"epoch": 0.56,
|
827 |
+
"grad_norm": 0.20371629297733307,
|
828 |
+
"learning_rate": 4.9969479093417894e-05,
|
829 |
+
"loss": 0.7179,
|
830 |
+
"step": 1150
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 0.56,
|
834 |
+
"grad_norm": 0.2982828617095947,
|
835 |
+
"learning_rate": 4.9965186344278746e-05,
|
836 |
+
"loss": 0.7222,
|
837 |
+
"step": 1160
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 0.57,
|
841 |
+
"grad_norm": 0.2097097784280777,
|
842 |
+
"learning_rate": 4.9960611448074555e-05,
|
843 |
+
"loss": 0.7174,
|
844 |
+
"step": 1170
|
845 |
+
},
|
846 |
+
{
|
847 |
+
"epoch": 0.57,
|
848 |
+
"grad_norm": 0.2264368236064911,
|
849 |
+
"learning_rate": 4.995575445650907e-05,
|
850 |
+
"loss": 0.7079,
|
851 |
+
"step": 1180
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 0.58,
|
855 |
+
"grad_norm": 0.24737605452537537,
|
856 |
+
"learning_rate": 4.995061542447418e-05,
|
857 |
+
"loss": 0.7272,
|
858 |
+
"step": 1190
|
859 |
+
},
|
860 |
+
{
|
861 |
+
"epoch": 0.58,
|
862 |
+
"grad_norm": 0.19530241191387177,
|
863 |
+
"learning_rate": 4.9945194410049254e-05,
|
864 |
+
"loss": 0.7216,
|
865 |
+
"step": 1200
|
866 |
+
},
|
867 |
+
{
|
868 |
+
"epoch": 0.58,
|
869 |
+
"grad_norm": 0.2791181802749634,
|
870 |
+
"learning_rate": 4.993949147450055e-05,
|
871 |
+
"loss": 0.712,
|
872 |
+
"step": 1210
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 0.59,
|
876 |
+
"grad_norm": 0.2095661610364914,
|
877 |
+
"learning_rate": 4.993350668228049e-05,
|
878 |
+
"loss": 0.7067,
|
879 |
+
"step": 1220
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 0.59,
|
883 |
+
"grad_norm": 0.17129307985305786,
|
884 |
+
"learning_rate": 4.9927240101026915e-05,
|
885 |
+
"loss": 0.7087,
|
886 |
+
"step": 1230
|
887 |
+
},
|
888 |
+
{
|
889 |
+
"epoch": 0.6,
|
890 |
+
"grad_norm": 0.2154540866613388,
|
891 |
+
"learning_rate": 4.992069180156235e-05,
|
892 |
+
"loss": 0.7411,
|
893 |
+
"step": 1240
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 0.6,
|
897 |
+
"grad_norm": 0.22642803192138672,
|
898 |
+
"learning_rate": 4.991386185789319e-05,
|
899 |
+
"loss": 0.7266,
|
900 |
+
"step": 1250
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 0.61,
|
904 |
+
"grad_norm": 0.19676081836223602,
|
905 |
+
"learning_rate": 4.9906750347208866e-05,
|
906 |
+
"loss": 0.7012,
|
907 |
+
"step": 1260
|
908 |
+
},
|
909 |
+
{
|
910 |
+
"epoch": 0.61,
|
911 |
+
"grad_norm": 0.2095811814069748,
|
912 |
+
"learning_rate": 4.989935734988098e-05,
|
913 |
+
"loss": 0.7162,
|
914 |
+
"step": 1270
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 0.62,
|
918 |
+
"grad_norm": 0.21455398201942444,
|
919 |
+
"learning_rate": 4.9891682949462385e-05,
|
920 |
+
"loss": 0.7177,
|
921 |
+
"step": 1280
|
922 |
+
},
|
923 |
+
{
|
924 |
+
"epoch": 0.62,
|
925 |
+
"grad_norm": 0.20205985009670258,
|
926 |
+
"learning_rate": 4.988372723268623e-05,
|
927 |
+
"loss": 0.718,
|
928 |
+
"step": 1290
|
929 |
+
},
|
930 |
+
{
|
931 |
+
"epoch": 0.63,
|
932 |
+
"grad_norm": 0.18780440092086792,
|
933 |
+
"learning_rate": 4.987549028946502e-05,
|
934 |
+
"loss": 0.7103,
|
935 |
+
"step": 1300
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 0.63,
|
939 |
+
"grad_norm": 0.20362792909145355,
|
940 |
+
"learning_rate": 4.986697221288956e-05,
|
941 |
+
"loss": 0.7304,
|
942 |
+
"step": 1310
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 0.64,
|
946 |
+
"grad_norm": 0.2820493280887604,
|
947 |
+
"learning_rate": 4.985817309922792e-05,
|
948 |
+
"loss": 0.7182,
|
949 |
+
"step": 1320
|
950 |
+
},
|
951 |
+
{
|
952 |
+
"epoch": 0.64,
|
953 |
+
"grad_norm": 0.20787598192691803,
|
954 |
+
"learning_rate": 4.984909304792435e-05,
|
955 |
+
"loss": 0.7142,
|
956 |
+
"step": 1330
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 0.65,
|
960 |
+
"grad_norm": 0.21624509990215302,
|
961 |
+
"learning_rate": 4.983973216159813e-05,
|
962 |
+
"loss": 0.7208,
|
963 |
+
"step": 1340
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"epoch": 0.65,
|
967 |
+
"grad_norm": 0.19343602657318115,
|
968 |
+
"learning_rate": 4.983009054604246e-05,
|
969 |
+
"loss": 0.7097,
|
970 |
+
"step": 1350
|
971 |
+
},
|
972 |
+
{
|
973 |
+
"epoch": 0.66,
|
974 |
+
"grad_norm": 0.254402756690979,
|
975 |
+
"learning_rate": 4.9820168310223215e-05,
|
976 |
+
"loss": 0.7193,
|
977 |
+
"step": 1360
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 0.66,
|
981 |
+
"grad_norm": 0.1940561830997467,
|
982 |
+
"learning_rate": 4.980996556627774e-05,
|
983 |
+
"loss": 0.6988,
|
984 |
+
"step": 1370
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 0.67,
|
988 |
+
"grad_norm": 0.22593462467193604,
|
989 |
+
"learning_rate": 4.9799482429513576e-05,
|
990 |
+
"loss": 0.7054,
|
991 |
+
"step": 1380
|
992 |
+
},
|
993 |
+
{
|
994 |
+
"epoch": 0.67,
|
995 |
+
"grad_norm": 0.22020135819911957,
|
996 |
+
"learning_rate": 4.978871901840716e-05,
|
997 |
+
"loss": 0.7221,
|
998 |
+
"step": 1390
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 0.68,
|
1002 |
+
"grad_norm": 0.22807644307613373,
|
1003 |
+
"learning_rate": 4.977767545460248e-05,
|
1004 |
+
"loss": 0.7213,
|
1005 |
+
"step": 1400
|
1006 |
+
},
|
1007 |
+
{
|
1008 |
+
"epoch": 0.68,
|
1009 |
+
"grad_norm": 0.1925562173128128,
|
1010 |
+
"learning_rate": 4.9766351862909703e-05,
|
1011 |
+
"loss": 0.7127,
|
1012 |
+
"step": 1410
|
1013 |
+
},
|
1014 |
+
{
|
1015 |
+
"epoch": 0.69,
|
1016 |
+
"grad_norm": 0.20845136046409607,
|
1017 |
+
"learning_rate": 4.9754748371303775e-05,
|
1018 |
+
"loss": 0.7207,
|
1019 |
+
"step": 1420
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 0.69,
|
1023 |
+
"grad_norm": 0.18583528697490692,
|
1024 |
+
"learning_rate": 4.974286511092294e-05,
|
1025 |
+
"loss": 0.7098,
|
1026 |
+
"step": 1430
|
1027 |
+
},
|
1028 |
+
{
|
1029 |
+
"epoch": 0.7,
|
1030 |
+
"grad_norm": 0.2019319236278534,
|
1031 |
+
"learning_rate": 4.973070221606732e-05,
|
1032 |
+
"loss": 0.7029,
|
1033 |
+
"step": 1440
|
1034 |
+
},
|
1035 |
+
{
|
1036 |
+
"epoch": 0.7,
|
1037 |
+
"grad_norm": 0.1966458261013031,
|
1038 |
+
"learning_rate": 4.971825982419731e-05,
|
1039 |
+
"loss": 0.7241,
|
1040 |
+
"step": 1450
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 0.71,
|
1044 |
+
"grad_norm": 0.20924623310565948,
|
1045 |
+
"learning_rate": 4.970553807593214e-05,
|
1046 |
+
"loss": 0.7126,
|
1047 |
+
"step": 1460
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 0.71,
|
1051 |
+
"grad_norm": 0.20655803382396698,
|
1052 |
+
"learning_rate": 4.969253711504814e-05,
|
1053 |
+
"loss": 0.6903,
|
1054 |
+
"step": 1470
|
1055 |
+
},
|
1056 |
+
{
|
1057 |
+
"epoch": 0.72,
|
1058 |
+
"grad_norm": 0.2179209142923355,
|
1059 |
+
"learning_rate": 4.967925708847727e-05,
|
1060 |
+
"loss": 0.683,
|
1061 |
+
"step": 1480
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 0.72,
|
1065 |
+
"grad_norm": 0.2198410928249359,
|
1066 |
+
"learning_rate": 4.966569814630534e-05,
|
1067 |
+
"loss": 0.7087,
|
1068 |
+
"step": 1490
|
1069 |
+
},
|
1070 |
+
{
|
1071 |
+
"epoch": 0.72,
|
1072 |
+
"grad_norm": 0.17970815300941467,
|
1073 |
+
"learning_rate": 4.9651860441770374e-05,
|
1074 |
+
"loss": 0.7006,
|
1075 |
+
"step": 1500
|
1076 |
+
},
|
1077 |
+
{
|
1078 |
+
"epoch": 0.72,
|
1079 |
+
"eval_loss": 0.7137542366981506,
|
1080 |
+
"eval_runtime": 43.0052,
|
1081 |
+
"eval_samples_per_second": 46.506,
|
1082 |
+
"eval_steps_per_second": 0.372,
|
1083 |
+
"step": 1500
|
1084 |
+
},
|
1085 |
+
{
|
1086 |
+
"epoch": 0.73,
|
1087 |
+
"grad_norm": 0.1931023746728897,
|
1088 |
+
"learning_rate": 4.963774413126086e-05,
|
1089 |
+
"loss": 0.7245,
|
1090 |
+
"step": 1510
|
1091 |
+
},
|
1092 |
+
{
|
1093 |
+
"epoch": 0.73,
|
1094 |
+
"grad_norm": 0.21812495589256287,
|
1095 |
+
"learning_rate": 4.9623349374313973e-05,
|
1096 |
+
"loss": 0.7119,
|
1097 |
+
"step": 1520
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 0.74,
|
1101 |
+
"grad_norm": 0.20784322917461395,
|
1102 |
+
"learning_rate": 4.960867633361381e-05,
|
1103 |
+
"loss": 0.704,
|
1104 |
+
"step": 1530
|
1105 |
+
},
|
1106 |
+
{
|
1107 |
+
"epoch": 0.74,
|
1108 |
+
"grad_norm": 0.18697650730609894,
|
1109 |
+
"learning_rate": 4.9593725174989507e-05,
|
1110 |
+
"loss": 0.7129,
|
1111 |
+
"step": 1540
|
1112 |
+
},
|
1113 |
+
{
|
1114 |
+
"epoch": 0.75,
|
1115 |
+
"grad_norm": 0.25061744451522827,
|
1116 |
+
"learning_rate": 4.957849606741338e-05,
|
1117 |
+
"loss": 0.6816,
|
1118 |
+
"step": 1550
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 0.75,
|
1122 |
+
"grad_norm": 0.1888163685798645,
|
1123 |
+
"learning_rate": 4.956298918299903e-05,
|
1124 |
+
"loss": 0.6905,
|
1125 |
+
"step": 1560
|
1126 |
+
},
|
1127 |
+
{
|
1128 |
+
"epoch": 0.76,
|
1129 |
+
"grad_norm": 0.1991235315799713,
|
1130 |
+
"learning_rate": 4.954720469699938e-05,
|
1131 |
+
"loss": 0.7258,
|
1132 |
+
"step": 1570
|
1133 |
+
},
|
1134 |
+
{
|
1135 |
+
"epoch": 0.76,
|
1136 |
+
"grad_norm": 0.18779252469539642,
|
1137 |
+
"learning_rate": 4.953114278780471e-05,
|
1138 |
+
"loss": 0.6918,
|
1139 |
+
"step": 1580
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 0.77,
|
1143 |
+
"grad_norm": 0.21268677711486816,
|
1144 |
+
"learning_rate": 4.951480363694061e-05,
|
1145 |
+
"loss": 0.6995,
|
1146 |
+
"step": 1590
|
1147 |
+
},
|
1148 |
+
{
|
1149 |
+
"epoch": 0.77,
|
1150 |
+
"grad_norm": 0.20242677628993988,
|
1151 |
+
"learning_rate": 4.949818742906597e-05,
|
1152 |
+
"loss": 0.7132,
|
1153 |
+
"step": 1600
|
1154 |
+
},
|
1155 |
+
{
|
1156 |
+
"epoch": 0.78,
|
1157 |
+
"grad_norm": 0.2314455509185791,
|
1158 |
+
"learning_rate": 4.9481294351970874e-05,
|
1159 |
+
"loss": 0.6866,
|
1160 |
+
"step": 1610
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 0.78,
|
1164 |
+
"grad_norm": 0.19638539850711823,
|
1165 |
+
"learning_rate": 4.9464124596574465e-05,
|
1166 |
+
"loss": 0.6993,
|
1167 |
+
"step": 1620
|
1168 |
+
},
|
1169 |
+
{
|
1170 |
+
"epoch": 0.79,
|
1171 |
+
"grad_norm": 0.185603529214859,
|
1172 |
+
"learning_rate": 4.944667835692281e-05,
|
1173 |
+
"loss": 0.6971,
|
1174 |
+
"step": 1630
|
1175 |
+
},
|
1176 |
+
{
|
1177 |
+
"epoch": 0.79,
|
1178 |
+
"grad_norm": 0.1594298928976059,
|
1179 |
+
"learning_rate": 4.942895583018669e-05,
|
1180 |
+
"loss": 0.7258,
|
1181 |
+
"step": 1640
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 0.8,
|
1185 |
+
"grad_norm": 0.21844029426574707,
|
1186 |
+
"learning_rate": 4.9410957216659374e-05,
|
1187 |
+
"loss": 0.7022,
|
1188 |
+
"step": 1650
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 0.8,
|
1192 |
+
"grad_norm": 0.19146642088890076,
|
1193 |
+
"learning_rate": 4.9392682719754364e-05,
|
1194 |
+
"loss": 0.6954,
|
1195 |
+
"step": 1660
|
1196 |
+
},
|
1197 |
+
{
|
1198 |
+
"epoch": 0.81,
|
1199 |
+
"grad_norm": 0.20112347602844238,
|
1200 |
+
"learning_rate": 4.9374132546003096e-05,
|
1201 |
+
"loss": 0.7017,
|
1202 |
+
"step": 1670
|
1203 |
+
},
|
1204 |
+
{
|
1205 |
+
"epoch": 0.81,
|
1206 |
+
"grad_norm": 0.22593769431114197,
|
1207 |
+
"learning_rate": 4.935530690505259e-05,
|
1208 |
+
"loss": 0.7397,
|
1209 |
+
"step": 1680
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 0.82,
|
1213 |
+
"grad_norm": 0.2611504793167114,
|
1214 |
+
"learning_rate": 4.933620600966311e-05,
|
1215 |
+
"loss": 0.7048,
|
1216 |
+
"step": 1690
|
1217 |
+
},
|
1218 |
+
{
|
1219 |
+
"epoch": 0.82,
|
1220 |
+
"grad_norm": 0.23995471000671387,
|
1221 |
+
"learning_rate": 4.931683007570571e-05,
|
1222 |
+
"loss": 0.709,
|
1223 |
+
"step": 1700
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 0.83,
|
1227 |
+
"grad_norm": 0.22278735041618347,
|
1228 |
+
"learning_rate": 4.9297179322159856e-05,
|
1229 |
+
"loss": 0.7116,
|
1230 |
+
"step": 1710
|
1231 |
+
},
|
1232 |
+
{
|
1233 |
+
"epoch": 0.83,
|
1234 |
+
"grad_norm": 0.2096502035856247,
|
1235 |
+
"learning_rate": 4.927725397111093e-05,
|
1236 |
+
"loss": 0.7061,
|
1237 |
+
"step": 1720
|
1238 |
+
},
|
1239 |
+
{
|
1240 |
+
"epoch": 0.84,
|
1241 |
+
"grad_norm": 0.22335517406463623,
|
1242 |
+
"learning_rate": 4.9257054247747644e-05,
|
1243 |
+
"loss": 0.6982,
|
1244 |
+
"step": 1730
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 0.84,
|
1248 |
+
"grad_norm": 0.18597687780857086,
|
1249 |
+
"learning_rate": 4.923658038035965e-05,
|
1250 |
+
"loss": 0.6927,
|
1251 |
+
"step": 1740
|
1252 |
+
},
|
1253 |
+
{
|
1254 |
+
"epoch": 0.85,
|
1255 |
+
"grad_norm": 0.1971946507692337,
|
1256 |
+
"learning_rate": 4.9215832600334816e-05,
|
1257 |
+
"loss": 0.6768,
|
1258 |
+
"step": 1750
|
1259 |
+
},
|
1260 |
+
{
|
1261 |
+
"epoch": 0.85,
|
1262 |
+
"grad_norm": 0.2101408839225769,
|
1263 |
+
"learning_rate": 4.9194811142156674e-05,
|
1264 |
+
"loss": 0.6933,
|
1265 |
+
"step": 1760
|
1266 |
+
},
|
1267 |
+
{
|
1268 |
+
"epoch": 0.86,
|
1269 |
+
"grad_norm": 0.17352430522441864,
|
1270 |
+
"learning_rate": 4.917351624340178e-05,
|
1271 |
+
"loss": 0.6918,
|
1272 |
+
"step": 1770
|
1273 |
+
},
|
1274 |
+
{
|
1275 |
+
"epoch": 0.86,
|
1276 |
+
"grad_norm": 0.18568973243236542,
|
1277 |
+
"learning_rate": 4.915194814473699e-05,
|
1278 |
+
"loss": 0.6711,
|
1279 |
+
"step": 1780
|
1280 |
+
},
|
1281 |
+
{
|
1282 |
+
"epoch": 0.87,
|
1283 |
+
"grad_norm": 0.22991472482681274,
|
1284 |
+
"learning_rate": 4.913010708991679e-05,
|
1285 |
+
"loss": 0.6938,
|
1286 |
+
"step": 1790
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 0.87,
|
1290 |
+
"grad_norm": 0.201872780919075,
|
1291 |
+
"learning_rate": 4.910799332578048e-05,
|
1292 |
+
"loss": 0.7051,
|
1293 |
+
"step": 1800
|
1294 |
+
},
|
1295 |
+
{
|
1296 |
+
"epoch": 0.87,
|
1297 |
+
"grad_norm": 0.2069864720106125,
|
1298 |
+
"learning_rate": 4.908560710224943e-05,
|
1299 |
+
"loss": 0.6983,
|
1300 |
+
"step": 1810
|
1301 |
+
},
|
1302 |
+
{
|
1303 |
+
"epoch": 0.88,
|
1304 |
+
"grad_norm": 0.2075955718755722,
|
1305 |
+
"learning_rate": 4.906294867232426e-05,
|
1306 |
+
"loss": 0.6996,
|
1307 |
+
"step": 1820
|
1308 |
+
},
|
1309 |
+
{
|
1310 |
+
"epoch": 0.88,
|
1311 |
+
"grad_norm": 0.19524317979812622,
|
1312 |
+
"learning_rate": 4.904001829208194e-05,
|
1313 |
+
"loss": 0.691,
|
1314 |
+
"step": 1830
|
1315 |
+
},
|
1316 |
+
{
|
1317 |
+
"epoch": 0.89,
|
1318 |
+
"grad_norm": 0.1967206597328186,
|
1319 |
+
"learning_rate": 4.901681622067293e-05,
|
1320 |
+
"loss": 0.6964,
|
1321 |
+
"step": 1840
|
1322 |
+
},
|
1323 |
+
{
|
1324 |
+
"epoch": 0.89,
|
1325 |
+
"grad_norm": 0.2195650041103363,
|
1326 |
+
"learning_rate": 4.899334272031825e-05,
|
1327 |
+
"loss": 0.6982,
|
1328 |
+
"step": 1850
|
1329 |
+
},
|
1330 |
+
{
|
1331 |
+
"epoch": 0.9,
|
1332 |
+
"grad_norm": 0.23072399199008942,
|
1333 |
+
"learning_rate": 4.8969598056306475e-05,
|
1334 |
+
"loss": 0.6838,
|
1335 |
+
"step": 1860
|
1336 |
+
},
|
1337 |
+
{
|
1338 |
+
"epoch": 0.9,
|
1339 |
+
"grad_norm": 0.1950792670249939,
|
1340 |
+
"learning_rate": 4.89455824969908e-05,
|
1341 |
+
"loss": 0.6984,
|
1342 |
+
"step": 1870
|
1343 |
+
},
|
1344 |
+
{
|
1345 |
+
"epoch": 0.91,
|
1346 |
+
"grad_norm": 0.2289438396692276,
|
1347 |
+
"learning_rate": 4.892129631378597e-05,
|
1348 |
+
"loss": 0.7012,
|
1349 |
+
"step": 1880
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 0.91,
|
1353 |
+
"grad_norm": 0.22316494584083557,
|
1354 |
+
"learning_rate": 4.8896739781165215e-05,
|
1355 |
+
"loss": 0.6904,
|
1356 |
+
"step": 1890
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 0.92,
|
1360 |
+
"grad_norm": 0.24066634476184845,
|
1361 |
+
"learning_rate": 4.8871913176657135e-05,
|
1362 |
+
"loss": 0.6974,
|
1363 |
+
"step": 1900
|
1364 |
+
},
|
1365 |
+
{
|
1366 |
+
"epoch": 0.92,
|
1367 |
+
"grad_norm": 0.18922199308872223,
|
1368 |
+
"learning_rate": 4.8846816780842606e-05,
|
1369 |
+
"loss": 0.686,
|
1370 |
+
"step": 1910
|
1371 |
+
},
|
1372 |
+
{
|
1373 |
+
"epoch": 0.93,
|
1374 |
+
"grad_norm": 0.2425553798675537,
|
1375 |
+
"learning_rate": 4.882145087735158e-05,
|
1376 |
+
"loss": 0.715,
|
1377 |
+
"step": 1920
|
1378 |
+
},
|
1379 |
+
{
|
1380 |
+
"epoch": 0.93,
|
1381 |
+
"grad_norm": 0.2267230898141861,
|
1382 |
+
"learning_rate": 4.879581575285988e-05,
|
1383 |
+
"loss": 0.698,
|
1384 |
+
"step": 1930
|
1385 |
+
},
|
1386 |
+
{
|
1387 |
+
"epoch": 0.94,
|
1388 |
+
"grad_norm": 0.22119645774364471,
|
1389 |
+
"learning_rate": 4.876991169708592e-05,
|
1390 |
+
"loss": 0.6942,
|
1391 |
+
"step": 1940
|
1392 |
+
},
|
1393 |
+
{
|
1394 |
+
"epoch": 0.94,
|
1395 |
+
"grad_norm": 0.22854161262512207,
|
1396 |
+
"learning_rate": 4.874373900278756e-05,
|
1397 |
+
"loss": 0.6912,
|
1398 |
+
"step": 1950
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 0.95,
|
1402 |
+
"grad_norm": 0.1927136778831482,
|
1403 |
+
"learning_rate": 4.871729796575863e-05,
|
1404 |
+
"loss": 0.6829,
|
1405 |
+
"step": 1960
|
1406 |
+
},
|
1407 |
+
{
|
1408 |
+
"epoch": 0.95,
|
1409 |
+
"grad_norm": 0.2238394170999527,
|
1410 |
+
"learning_rate": 4.869058888482571e-05,
|
1411 |
+
"loss": 0.6831,
|
1412 |
+
"step": 1970
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 0.96,
|
1416 |
+
"grad_norm": 0.17643775045871735,
|
1417 |
+
"learning_rate": 4.86636120618447e-05,
|
1418 |
+
"loss": 0.6875,
|
1419 |
+
"step": 1980
|
1420 |
+
},
|
1421 |
+
{
|
1422 |
+
"epoch": 0.96,
|
1423 |
+
"grad_norm": 0.2161882072687149,
|
1424 |
+
"learning_rate": 4.863636780169742e-05,
|
1425 |
+
"loss": 0.6893,
|
1426 |
+
"step": 1990
|
1427 |
+
},
|
1428 |
+
{
|
1429 |
+
"epoch": 0.97,
|
1430 |
+
"grad_norm": 0.20689886808395386,
|
1431 |
+
"learning_rate": 4.860885641228815e-05,
|
1432 |
+
"loss": 0.6721,
|
1433 |
+
"step": 2000
|
1434 |
+
},
|
1435 |
+
{
|
1436 |
+
"epoch": 0.97,
|
1437 |
+
"eval_loss": 0.701475203037262,
|
1438 |
+
"eval_runtime": 43.016,
|
1439 |
+
"eval_samples_per_second": 46.494,
|
1440 |
+
"eval_steps_per_second": 0.372,
|
1441 |
+
"step": 2000
|
1442 |
+
},
|
1443 |
+
{
|
1444 |
+
"epoch": 0.97,
|
1445 |
+
"grad_norm": 0.18866655230522156,
|
1446 |
+
"learning_rate": 4.8581078204540196e-05,
|
1447 |
+
"loss": 0.6941,
|
1448 |
+
"step": 2010
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 0.98,
|
1452 |
+
"grad_norm": 0.20633092522621155,
|
1453 |
+
"learning_rate": 4.855303349239231e-05,
|
1454 |
+
"loss": 0.6943,
|
1455 |
+
"step": 2020
|
1456 |
+
},
|
1457 |
+
{
|
1458 |
+
"epoch": 0.98,
|
1459 |
+
"grad_norm": 0.2335944026708603,
|
1460 |
+
"learning_rate": 4.85247225927952e-05,
|
1461 |
+
"loss": 0.6986,
|
1462 |
+
"step": 2030
|
1463 |
+
},
|
1464 |
+
{
|
1465 |
+
"epoch": 0.99,
|
1466 |
+
"grad_norm": 0.18950659036636353,
|
1467 |
+
"learning_rate": 4.8496145825707905e-05,
|
1468 |
+
"loss": 0.6894,
|
1469 |
+
"step": 2040
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 0.99,
|
1473 |
+
"grad_norm": 0.22117875516414642,
|
1474 |
+
"learning_rate": 4.8467303514094204e-05,
|
1475 |
+
"loss": 0.6956,
|
1476 |
+
"step": 2050
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 1.0,
|
1480 |
+
"grad_norm": 0.23478396236896515,
|
1481 |
+
"learning_rate": 4.8438195983918967e-05,
|
1482 |
+
"loss": 0.6919,
|
1483 |
+
"step": 2060
|
1484 |
+
},
|
1485 |
+
{
|
1486 |
+
"epoch": 1.0,
|
1487 |
+
"grad_norm": 0.24754443764686584,
|
1488 |
+
"learning_rate": 4.8408823564144454e-05,
|
1489 |
+
"loss": 0.6904,
|
1490 |
+
"step": 2070
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 1.01,
|
1494 |
+
"grad_norm": 0.19141903519630432,
|
1495 |
+
"learning_rate": 4.837918658672661e-05,
|
1496 |
+
"loss": 0.6914,
|
1497 |
+
"step": 2080
|
1498 |
+
},
|
1499 |
+
{
|
1500 |
+
"epoch": 1.01,
|
1501 |
+
"grad_norm": 0.21387813985347748,
|
1502 |
+
"learning_rate": 4.834928538661131e-05,
|
1503 |
+
"loss": 0.6965,
|
1504 |
+
"step": 2090
|
1505 |
+
},
|
1506 |
+
{
|
1507 |
+
"epoch": 1.01,
|
1508 |
+
"grad_norm": 0.20495550334453583,
|
1509 |
+
"learning_rate": 4.831912030173058e-05,
|
1510 |
+
"loss": 0.6949,
|
1511 |
+
"step": 2100
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 1.02,
|
1515 |
+
"grad_norm": 0.35174649953842163,
|
1516 |
+
"learning_rate": 4.8288691672998765e-05,
|
1517 |
+
"loss": 0.6813,
|
1518 |
+
"step": 2110
|
1519 |
+
},
|
1520 |
+
{
|
1521 |
+
"epoch": 1.02,
|
1522 |
+
"grad_norm": 0.2461618334054947,
|
1523 |
+
"learning_rate": 4.825799984430867e-05,
|
1524 |
+
"loss": 0.6687,
|
1525 |
+
"step": 2120
|
1526 |
+
},
|
1527 |
+
{
|
1528 |
+
"epoch": 1.03,
|
1529 |
+
"grad_norm": 0.20039792358875275,
|
1530 |
+
"learning_rate": 4.8227045162527714e-05,
|
1531 |
+
"loss": 0.6737,
|
1532 |
+
"step": 2130
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 1.03,
|
1536 |
+
"grad_norm": 0.19774436950683594,
|
1537 |
+
"learning_rate": 4.8195827977493955e-05,
|
1538 |
+
"loss": 0.6815,
|
1539 |
+
"step": 2140
|
1540 |
+
},
|
1541 |
+
{
|
1542 |
+
"epoch": 1.04,
|
1543 |
+
"grad_norm": 0.21472251415252686,
|
1544 |
+
"learning_rate": 4.8164348642012194e-05,
|
1545 |
+
"loss": 0.6957,
|
1546 |
+
"step": 2150
|
1547 |
+
},
|
1548 |
+
{
|
1549 |
+
"epoch": 1.04,
|
1550 |
+
"grad_norm": 0.1990274041891098,
|
1551 |
+
"learning_rate": 4.813260751184992e-05,
|
1552 |
+
"loss": 0.6916,
|
1553 |
+
"step": 2160
|
1554 |
+
},
|
1555 |
+
{
|
1556 |
+
"epoch": 1.05,
|
1557 |
+
"grad_norm": 0.20387563109397888,
|
1558 |
+
"learning_rate": 4.810060494573335e-05,
|
1559 |
+
"loss": 0.6814,
|
1560 |
+
"step": 2170
|
1561 |
+
},
|
1562 |
+
{
|
1563 |
+
"epoch": 1.05,
|
1564 |
+
"grad_norm": 0.1792985051870346,
|
1565 |
+
"learning_rate": 4.806834130534336e-05,
|
1566 |
+
"loss": 0.698,
|
1567 |
+
"step": 2180
|
1568 |
+
},
|
1569 |
+
{
|
1570 |
+
"epoch": 1.06,
|
1571 |
+
"grad_norm": 0.22255714237689972,
|
1572 |
+
"learning_rate": 4.803581695531134e-05,
|
1573 |
+
"loss": 0.6765,
|
1574 |
+
"step": 2190
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 1.06,
|
1578 |
+
"grad_norm": 0.19549421966075897,
|
1579 |
+
"learning_rate": 4.8003032263215185e-05,
|
1580 |
+
"loss": 0.6867,
|
1581 |
+
"step": 2200
|
1582 |
+
},
|
1583 |
+
{
|
1584 |
+
"epoch": 1.07,
|
1585 |
+
"grad_norm": 0.1929677277803421,
|
1586 |
+
"learning_rate": 4.796998759957504e-05,
|
1587 |
+
"loss": 0.6935,
|
1588 |
+
"step": 2210
|
1589 |
+
},
|
1590 |
+
{
|
1591 |
+
"epoch": 1.07,
|
1592 |
+
"grad_norm": 0.22705228626728058,
|
1593 |
+
"learning_rate": 4.793668333784915e-05,
|
1594 |
+
"loss": 0.6897,
|
1595 |
+
"step": 2220
|
1596 |
+
},
|
1597 |
+
{
|
1598 |
+
"epoch": 1.08,
|
1599 |
+
"grad_norm": 0.21798522770404816,
|
1600 |
+
"learning_rate": 4.790311985442966e-05,
|
1601 |
+
"loss": 0.6694,
|
1602 |
+
"step": 2230
|
1603 |
+
},
|
1604 |
+
{
|
1605 |
+
"epoch": 1.08,
|
1606 |
+
"grad_norm": 0.19095788896083832,
|
1607 |
+
"learning_rate": 4.7869297528638315e-05,
|
1608 |
+
"loss": 0.6699,
|
1609 |
+
"step": 2240
|
1610 |
+
},
|
1611 |
+
{
|
1612 |
+
"epoch": 1.09,
|
1613 |
+
"grad_norm": 0.23002374172210693,
|
1614 |
+
"learning_rate": 4.7835216742722225e-05,
|
1615 |
+
"loss": 0.6807,
|
1616 |
+
"step": 2250
|
1617 |
+
},
|
1618 |
+
{
|
1619 |
+
"epoch": 1.09,
|
1620 |
+
"grad_norm": 0.2107260823249817,
|
1621 |
+
"learning_rate": 4.780087788184947e-05,
|
1622 |
+
"loss": 0.6855,
|
1623 |
+
"step": 2260
|
1624 |
+
},
|
1625 |
+
{
|
1626 |
+
"epoch": 1.1,
|
1627 |
+
"grad_norm": 0.23853424191474915,
|
1628 |
+
"learning_rate": 4.776628133410487e-05,
|
1629 |
+
"loss": 0.6946,
|
1630 |
+
"step": 2270
|
1631 |
+
},
|
1632 |
+
{
|
1633 |
+
"epoch": 1.1,
|
1634 |
+
"grad_norm": 0.25372380018234253,
|
1635 |
+
"learning_rate": 4.7731427490485455e-05,
|
1636 |
+
"loss": 0.6724,
|
1637 |
+
"step": 2280
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 1.11,
|
1641 |
+
"grad_norm": 0.182255819439888,
|
1642 |
+
"learning_rate": 4.7696316744896146e-05,
|
1643 |
+
"loss": 0.6976,
|
1644 |
+
"step": 2290
|
1645 |
+
},
|
1646 |
+
{
|
1647 |
+
"epoch": 1.11,
|
1648 |
+
"grad_norm": 0.20137862861156464,
|
1649 |
+
"learning_rate": 4.7660949494145276e-05,
|
1650 |
+
"loss": 0.6853,
|
1651 |
+
"step": 2300
|
1652 |
+
},
|
1653 |
+
{
|
1654 |
+
"epoch": 1.12,
|
1655 |
+
"grad_norm": 0.19899927079677582,
|
1656 |
+
"learning_rate": 4.7625326137940106e-05,
|
1657 |
+
"loss": 0.6696,
|
1658 |
+
"step": 2310
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 1.12,
|
1662 |
+
"grad_norm": 0.24147948622703552,
|
1663 |
+
"learning_rate": 4.758944707888228e-05,
|
1664 |
+
"loss": 0.6615,
|
1665 |
+
"step": 2320
|
1666 |
+
},
|
1667 |
+
{
|
1668 |
+
"epoch": 1.13,
|
1669 |
+
"grad_norm": 0.19519510865211487,
|
1670 |
+
"learning_rate": 4.7553312722463305e-05,
|
1671 |
+
"loss": 0.685,
|
1672 |
+
"step": 2330
|
1673 |
+
},
|
1674 |
+
{
|
1675 |
+
"epoch": 1.13,
|
1676 |
+
"grad_norm": 0.20128916203975677,
|
1677 |
+
"learning_rate": 4.751692347705998e-05,
|
1678 |
+
"loss": 0.6829,
|
1679 |
+
"step": 2340
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 1.14,
|
1683 |
+
"grad_norm": 0.16112156212329865,
|
1684 |
+
"learning_rate": 4.748027975392976e-05,
|
1685 |
+
"loss": 0.6714,
|
1686 |
+
"step": 2350
|
1687 |
+
},
|
1688 |
+
{
|
1689 |
+
"epoch": 1.14,
|
1690 |
+
"grad_norm": 0.2659178674221039,
|
1691 |
+
"learning_rate": 4.744338196720608e-05,
|
1692 |
+
"loss": 0.6834,
|
1693 |
+
"step": 2360
|
1694 |
+
},
|
1695 |
+
{
|
1696 |
+
"epoch": 1.15,
|
1697 |
+
"grad_norm": 0.20477250218391418,
|
1698 |
+
"learning_rate": 4.740623053389374e-05,
|
1699 |
+
"loss": 0.6664,
|
1700 |
+
"step": 2370
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 1.15,
|
1704 |
+
"grad_norm": 0.18069259822368622,
|
1705 |
+
"learning_rate": 4.7368825873864154e-05,
|
1706 |
+
"loss": 0.669,
|
1707 |
+
"step": 2380
|
1708 |
+
},
|
1709 |
+
{
|
1710 |
+
"epoch": 1.16,
|
1711 |
+
"grad_norm": 0.2146090418100357,
|
1712 |
+
"learning_rate": 4.733116840985058e-05,
|
1713 |
+
"loss": 0.6836,
|
1714 |
+
"step": 2390
|
1715 |
+
},
|
1716 |
+
{
|
1717 |
+
"epoch": 1.16,
|
1718 |
+
"grad_norm": 0.20656266808509827,
|
1719 |
+
"learning_rate": 4.729325856744341e-05,
|
1720 |
+
"loss": 0.6818,
|
1721 |
+
"step": 2400
|
1722 |
+
},
|
1723 |
+
{
|
1724 |
+
"epoch": 1.16,
|
1725 |
+
"grad_norm": 0.1764717549085617,
|
1726 |
+
"learning_rate": 4.725509677508528e-05,
|
1727 |
+
"loss": 0.7012,
|
1728 |
+
"step": 2410
|
1729 |
+
},
|
1730 |
+
{
|
1731 |
+
"epoch": 1.17,
|
1732 |
+
"grad_norm": 0.19059453904628754,
|
1733 |
+
"learning_rate": 4.721668346406631e-05,
|
1734 |
+
"loss": 0.6885,
|
1735 |
+
"step": 2420
|
1736 |
+
},
|
1737 |
+
{
|
1738 |
+
"epoch": 1.17,
|
1739 |
+
"grad_norm": 0.1867195963859558,
|
1740 |
+
"learning_rate": 4.7178019068519165e-05,
|
1741 |
+
"loss": 0.6704,
|
1742 |
+
"step": 2430
|
1743 |
+
},
|
1744 |
+
{
|
1745 |
+
"epoch": 1.18,
|
1746 |
+
"grad_norm": 0.19844096899032593,
|
1747 |
+
"learning_rate": 4.713910402541416e-05,
|
1748 |
+
"loss": 0.6746,
|
1749 |
+
"step": 2440
|
1750 |
+
},
|
1751 |
+
{
|
1752 |
+
"epoch": 1.18,
|
1753 |
+
"grad_norm": 0.21426193416118622,
|
1754 |
+
"learning_rate": 4.709993877455436e-05,
|
1755 |
+
"loss": 0.6627,
|
1756 |
+
"step": 2450
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 1.19,
|
1760 |
+
"grad_norm": 0.19326262176036835,
|
1761 |
+
"learning_rate": 4.706052375857058e-05,
|
1762 |
+
"loss": 0.6652,
|
1763 |
+
"step": 2460
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 1.19,
|
1767 |
+
"grad_norm": 0.229305237531662,
|
1768 |
+
"learning_rate": 4.7020859422916365e-05,
|
1769 |
+
"loss": 0.6746,
|
1770 |
+
"step": 2470
|
1771 |
+
},
|
1772 |
+
{
|
1773 |
+
"epoch": 1.2,
|
1774 |
+
"grad_norm": 0.20060321688652039,
|
1775 |
+
"learning_rate": 4.698094621586299e-05,
|
1776 |
+
"loss": 0.6658,
|
1777 |
+
"step": 2480
|
1778 |
+
},
|
1779 |
+
{
|
1780 |
+
"epoch": 1.2,
|
1781 |
+
"grad_norm": 0.21455462276935577,
|
1782 |
+
"learning_rate": 4.694078458849438e-05,
|
1783 |
+
"loss": 0.6966,
|
1784 |
+
"step": 2490
|
1785 |
+
},
|
1786 |
+
{
|
1787 |
+
"epoch": 1.21,
|
1788 |
+
"grad_norm": 0.18379837274551392,
|
1789 |
+
"learning_rate": 4.690037499470202e-05,
|
1790 |
+
"loss": 0.6753,
|
1791 |
+
"step": 2500
|
1792 |
+
},
|
1793 |
+
{
|
1794 |
+
"epoch": 1.21,
|
1795 |
+
"eval_loss": 0.6940956711769104,
|
1796 |
+
"eval_runtime": 42.7889,
|
1797 |
+
"eval_samples_per_second": 46.741,
|
1798 |
+
"eval_steps_per_second": 0.374,
|
1799 |
+
"step": 2500
|
1800 |
+
}
|
1801 |
+
],
|
1802 |
+
"logging_steps": 10,
|
1803 |
+
"max_steps": 10345,
|
1804 |
+
"num_input_tokens_seen": 0,
|
1805 |
+
"num_train_epochs": 5,
|
1806 |
+
"save_steps": 500,
|
1807 |
+
"total_flos": 6.328494039171596e+18,
|
1808 |
+
"train_batch_size": 8,
|
1809 |
+
"trial_name": null,
|
1810 |
+
"trial_params": null
|
1811 |
+
}
|
checkpoint-2500/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3814e1058daec6f70a62fab0452b08b4d1c663bae76d77a07b33001875b2afcd
|
3 |
+
size 5048
|