sheepy928 commited on
Commit
856dc4c
1 Parent(s): c7d0d13

Training in progress, step 2000, checkpoint

Browse files
checkpoint-2000/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-13b-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": 5120,
122
+ "id2label": {
123
+ "0": "LABEL_0",
124
+ "1": "LABEL_1"
125
+ },
126
+ "initializer_range": 0.02,
127
+ "intermediate_size": 13824,
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": 40,
141
+ "num_beam_groups": 1,
142
+ "num_beams": 1,
143
+ "num_hidden_layers": 40,
144
+ "num_key_value_heads": 40,
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-2000/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.39.3"
4
+ }
checkpoint-2000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:194c4b9a628819b57d1ff9561a184e6bfc983ddb772fe27786871672ae873841
3
+ size 620848256
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1da4f7bb4181cbc5ebc55bdaaeabe6155d425688a8341278b7f0aa73a2aff09
3
+ size 297976223
checkpoint-2000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e0ec3b7d44001bf209fb32953161c20466242cbca9a088a44c72ce2d0449ec2
3
+ size 14960
checkpoint-2000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1076f1ec1401d9b1c571ffa9494af94e2c740818ed470d3580e3926a53fe94f
3
+ size 14960
checkpoint-2000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73cc21a16e4d7e149b7153f671558d7ff21dd8ecdb00a7e50459f9094aacc0a5
3
+ size 14960
checkpoint-2000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0207ebeb1802c02f45822b0c4a45f64e0595aa912b05c784adc6d4b0e029b4b
3
+ size 14960
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e745c3ddfd7ed98232b07fc65826852915442a7b9331732efedfcdb2448cf8e
3
+ size 1064
checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,1453 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.4833252779120348,
5
+ "eval_steps": 500,
6
+ "global_step": 2000,
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": NaN,
14
+ "learning_rate": 2.0000000000000002e-07,
15
+ "loss": 2.0414,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.0,
20
+ "grad_norm": 53.515953063964844,
21
+ "learning_rate": 6.5e-07,
22
+ "loss": 1.8355,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.01,
27
+ "grad_norm": 27.920907974243164,
28
+ "learning_rate": 1.15e-06,
29
+ "loss": 1.5172,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.01,
34
+ "grad_norm": 20.47172737121582,
35
+ "learning_rate": 1.65e-06,
36
+ "loss": 1.2654,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.01,
41
+ "grad_norm": 8.903529167175293,
42
+ "learning_rate": 2.1499999999999997e-06,
43
+ "loss": 1.1095,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.01,
48
+ "grad_norm": 5.902841567993164,
49
+ "learning_rate": 2.65e-06,
50
+ "loss": 1.0017,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.02,
55
+ "grad_norm": 2.9340908527374268,
56
+ "learning_rate": 3.1500000000000003e-06,
57
+ "loss": 0.8846,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.02,
62
+ "grad_norm": 8.701367378234863,
63
+ "learning_rate": 3.6499999999999998e-06,
64
+ "loss": 0.8572,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.02,
69
+ "grad_norm": 1.3205362558364868,
70
+ "learning_rate": 4.15e-06,
71
+ "loss": 0.839,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.02,
76
+ "grad_norm": 3.0334558486938477,
77
+ "learning_rate": 4.65e-06,
78
+ "loss": 0.7838,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.03,
83
+ "grad_norm": 1.8086520433425903,
84
+ "learning_rate": 5.15e-06,
85
+ "loss": 0.7782,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.03,
90
+ "grad_norm": 29.50135040283203,
91
+ "learning_rate": 5.65e-06,
92
+ "loss": 0.7883,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.03,
97
+ "grad_norm": 3.2708683013916016,
98
+ "learning_rate": 6.15e-06,
99
+ "loss": 0.7961,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.03,
104
+ "grad_norm": 33.43790817260742,
105
+ "learning_rate": 6.650000000000001e-06,
106
+ "loss": 0.7626,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.04,
111
+ "grad_norm": 1.2334959506988525,
112
+ "learning_rate": 7.15e-06,
113
+ "loss": 0.7732,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.04,
118
+ "grad_norm": 1.7489686012268066,
119
+ "learning_rate": 7.65e-06,
120
+ "loss": 0.7715,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.04,
125
+ "grad_norm": 3.358823776245117,
126
+ "learning_rate": 8.15e-06,
127
+ "loss": 0.7778,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.04,
132
+ "grad_norm": 4.348245620727539,
133
+ "learning_rate": 8.65e-06,
134
+ "loss": 0.7619,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.05,
139
+ "grad_norm": 0.9023125767707825,
140
+ "learning_rate": 9.15e-06,
141
+ "loss": 0.7746,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.05,
146
+ "grad_norm": 0.9158114790916443,
147
+ "learning_rate": 9.65e-06,
148
+ "loss": 0.7592,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.05,
153
+ "grad_norm": 0.6256226301193237,
154
+ "learning_rate": 1.0150000000000001e-05,
155
+ "loss": 0.7787,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.05,
160
+ "grad_norm": 0.3805778920650482,
161
+ "learning_rate": 1.065e-05,
162
+ "loss": 0.7811,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.06,
167
+ "grad_norm": 0.6968041062355042,
168
+ "learning_rate": 1.115e-05,
169
+ "loss": 0.73,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.06,
174
+ "grad_norm": 0.9035410284996033,
175
+ "learning_rate": 1.1650000000000002e-05,
176
+ "loss": 0.7518,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.06,
181
+ "grad_norm": 1.6727488040924072,
182
+ "learning_rate": 1.215e-05,
183
+ "loss": 0.7805,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.06,
188
+ "grad_norm": 0.4153461158275604,
189
+ "learning_rate": 1.2650000000000001e-05,
190
+ "loss": 0.7706,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.07,
195
+ "grad_norm": 1.4846651554107666,
196
+ "learning_rate": 1.3150000000000001e-05,
197
+ "loss": 0.7492,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.07,
202
+ "grad_norm": 2.6631388664245605,
203
+ "learning_rate": 1.3650000000000001e-05,
204
+ "loss": 0.7681,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.07,
209
+ "grad_norm": 0.6325013041496277,
210
+ "learning_rate": 1.415e-05,
211
+ "loss": 0.7753,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.07,
216
+ "grad_norm": 0.877907395362854,
217
+ "learning_rate": 1.465e-05,
218
+ "loss": 0.7188,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.07,
223
+ "grad_norm": 0.37142279744148254,
224
+ "learning_rate": 1.515e-05,
225
+ "loss": 0.7204,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.08,
230
+ "grad_norm": 0.7337246537208557,
231
+ "learning_rate": 1.565e-05,
232
+ "loss": 0.778,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.08,
237
+ "grad_norm": 0.5847220420837402,
238
+ "learning_rate": 1.6150000000000003e-05,
239
+ "loss": 0.7288,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.08,
244
+ "grad_norm": 3.745180606842041,
245
+ "learning_rate": 1.665e-05,
246
+ "loss": 0.7531,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.08,
251
+ "grad_norm": 0.357301265001297,
252
+ "learning_rate": 1.7150000000000004e-05,
253
+ "loss": 0.7448,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.09,
258
+ "grad_norm": 0.9032486081123352,
259
+ "learning_rate": 1.765e-05,
260
+ "loss": 0.7651,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.09,
265
+ "grad_norm": 0.2864232361316681,
266
+ "learning_rate": 1.815e-05,
267
+ "loss": 0.7193,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.09,
272
+ "grad_norm": 1.8560261726379395,
273
+ "learning_rate": 1.865e-05,
274
+ "loss": 0.7421,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.09,
279
+ "grad_norm": 0.2971792221069336,
280
+ "learning_rate": 1.915e-05,
281
+ "loss": 0.7171,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.1,
286
+ "grad_norm": 0.40850627422332764,
287
+ "learning_rate": 1.9650000000000003e-05,
288
+ "loss": 0.7459,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.1,
293
+ "grad_norm": 0.3934139311313629,
294
+ "learning_rate": 2.0150000000000002e-05,
295
+ "loss": 0.7205,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.1,
300
+ "grad_norm": 1.8674131631851196,
301
+ "learning_rate": 2.065e-05,
302
+ "loss": 0.752,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.1,
307
+ "grad_norm": 0.36707818508148193,
308
+ "learning_rate": 2.115e-05,
309
+ "loss": 0.7494,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.11,
314
+ "grad_norm": 0.5044310092926025,
315
+ "learning_rate": 2.165e-05,
316
+ "loss": 0.7595,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.11,
321
+ "grad_norm": 1.9025150537490845,
322
+ "learning_rate": 2.215e-05,
323
+ "loss": 0.7379,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.11,
328
+ "grad_norm": 0.5004140138626099,
329
+ "learning_rate": 2.265e-05,
330
+ "loss": 0.7886,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.11,
335
+ "grad_norm": 3.544482707977295,
336
+ "learning_rate": 2.3150000000000004e-05,
337
+ "loss": 0.7259,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.12,
342
+ "grad_norm": 0.33023321628570557,
343
+ "learning_rate": 2.365e-05,
344
+ "loss": 0.7333,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.12,
349
+ "grad_norm": 0.3548080325126648,
350
+ "learning_rate": 2.415e-05,
351
+ "loss": 0.7527,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.12,
356
+ "grad_norm": 0.42450150847435,
357
+ "learning_rate": 2.465e-05,
358
+ "loss": 0.7443,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.12,
363
+ "eval_loss": 0.7428915500640869,
364
+ "eval_runtime": 68.3143,
365
+ "eval_samples_per_second": 29.276,
366
+ "eval_steps_per_second": 0.922,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 0.12,
371
+ "grad_norm": 0.3299656808376312,
372
+ "learning_rate": 2.515e-05,
373
+ "loss": 0.7447,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 0.13,
378
+ "grad_norm": 0.27258285880088806,
379
+ "learning_rate": 2.5650000000000003e-05,
380
+ "loss": 0.7278,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 0.13,
385
+ "grad_norm": 0.23753009736537933,
386
+ "learning_rate": 2.6150000000000002e-05,
387
+ "loss": 0.7721,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 0.13,
392
+ "grad_norm": 1.1074901819229126,
393
+ "learning_rate": 2.6650000000000004e-05,
394
+ "loss": 0.7383,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 0.13,
399
+ "grad_norm": 0.428067147731781,
400
+ "learning_rate": 2.7150000000000003e-05,
401
+ "loss": 0.7508,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 0.14,
406
+ "grad_norm": 0.35956478118896484,
407
+ "learning_rate": 2.7650000000000005e-05,
408
+ "loss": 0.7461,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 0.14,
413
+ "grad_norm": 0.23017314076423645,
414
+ "learning_rate": 2.815e-05,
415
+ "loss": 0.7463,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 0.14,
420
+ "grad_norm": 0.20370900630950928,
421
+ "learning_rate": 2.865e-05,
422
+ "loss": 0.7405,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 0.14,
427
+ "grad_norm": 0.24586541950702667,
428
+ "learning_rate": 2.915e-05,
429
+ "loss": 0.6964,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 0.14,
434
+ "grad_norm": 0.27857884764671326,
435
+ "learning_rate": 2.965e-05,
436
+ "loss": 0.7376,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 0.15,
441
+ "grad_norm": 0.2655833065509796,
442
+ "learning_rate": 3.015e-05,
443
+ "loss": 0.7496,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 0.15,
448
+ "grad_norm": 0.17174670100212097,
449
+ "learning_rate": 3.065e-05,
450
+ "loss": 0.7219,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 0.15,
455
+ "grad_norm": 0.22892743349075317,
456
+ "learning_rate": 3.115e-05,
457
+ "loss": 0.7111,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 0.15,
462
+ "grad_norm": 0.5913789868354797,
463
+ "learning_rate": 3.1650000000000004e-05,
464
+ "loss": 0.7097,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 0.16,
469
+ "grad_norm": 0.25620004534721375,
470
+ "learning_rate": 3.215e-05,
471
+ "loss": 0.7306,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 0.16,
476
+ "grad_norm": 0.3311476409435272,
477
+ "learning_rate": 3.265e-05,
478
+ "loss": 0.7406,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 0.16,
483
+ "grad_norm": 0.4708893597126007,
484
+ "learning_rate": 3.3150000000000006e-05,
485
+ "loss": 0.7071,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 0.16,
490
+ "grad_norm": 0.22469288110733032,
491
+ "learning_rate": 3.3650000000000005e-05,
492
+ "loss": 0.7116,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 0.17,
497
+ "grad_norm": 0.2909330725669861,
498
+ "learning_rate": 3.415e-05,
499
+ "loss": 0.7592,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 0.17,
504
+ "grad_norm": 0.21274766325950623,
505
+ "learning_rate": 3.465e-05,
506
+ "loss": 0.7144,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 0.17,
511
+ "grad_norm": 0.23929230868816376,
512
+ "learning_rate": 3.515e-05,
513
+ "loss": 0.7382,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 0.17,
518
+ "grad_norm": 0.3021218180656433,
519
+ "learning_rate": 3.565e-05,
520
+ "loss": 0.7287,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 0.18,
525
+ "grad_norm": 0.23557321727275848,
526
+ "learning_rate": 3.615e-05,
527
+ "loss": 0.7475,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 0.18,
532
+ "grad_norm": 0.22648084163665771,
533
+ "learning_rate": 3.665e-05,
534
+ "loss": 0.7457,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 0.18,
539
+ "grad_norm": 0.2707761824131012,
540
+ "learning_rate": 3.715e-05,
541
+ "loss": 0.7322,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 0.18,
546
+ "grad_norm": 0.2799661457538605,
547
+ "learning_rate": 3.765e-05,
548
+ "loss": 0.7682,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 0.19,
553
+ "grad_norm": 0.326861172914505,
554
+ "learning_rate": 3.8150000000000006e-05,
555
+ "loss": 0.73,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 0.19,
560
+ "grad_norm": 0.31985723972320557,
561
+ "learning_rate": 3.8650000000000004e-05,
562
+ "loss": 0.744,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 0.19,
567
+ "grad_norm": 0.21637533605098724,
568
+ "learning_rate": 3.915e-05,
569
+ "loss": 0.7161,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 0.19,
574
+ "grad_norm": 0.26102888584136963,
575
+ "learning_rate": 3.965e-05,
576
+ "loss": 0.7553,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 0.2,
581
+ "grad_norm": 0.2469174712896347,
582
+ "learning_rate": 4.015000000000001e-05,
583
+ "loss": 0.7444,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 0.2,
588
+ "grad_norm": 0.2022310495376587,
589
+ "learning_rate": 4.065e-05,
590
+ "loss": 0.7089,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 0.2,
595
+ "grad_norm": 0.1729898750782013,
596
+ "learning_rate": 4.115e-05,
597
+ "loss": 0.7122,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 0.2,
602
+ "grad_norm": 0.22450707852840424,
603
+ "learning_rate": 4.165e-05,
604
+ "loss": 0.7099,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 0.21,
609
+ "grad_norm": 0.22389698028564453,
610
+ "learning_rate": 4.215e-05,
611
+ "loss": 0.7445,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 0.21,
616
+ "grad_norm": 0.24962696433067322,
617
+ "learning_rate": 4.265e-05,
618
+ "loss": 0.7052,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 0.21,
623
+ "grad_norm": 0.23780404031276703,
624
+ "learning_rate": 4.315e-05,
625
+ "loss": 0.7423,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 0.21,
630
+ "grad_norm": 0.2434270679950714,
631
+ "learning_rate": 4.3650000000000004e-05,
632
+ "loss": 0.7291,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 0.22,
637
+ "grad_norm": 0.3332350552082062,
638
+ "learning_rate": 4.415e-05,
639
+ "loss": 0.7306,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 0.22,
644
+ "grad_norm": 0.5084495544433594,
645
+ "learning_rate": 4.465e-05,
646
+ "loss": 0.735,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 0.22,
651
+ "grad_norm": 0.27552103996276855,
652
+ "learning_rate": 4.5150000000000006e-05,
653
+ "loss": 0.7082,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 0.22,
658
+ "grad_norm": 0.2184416800737381,
659
+ "learning_rate": 4.5650000000000005e-05,
660
+ "loss": 0.7066,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 0.22,
665
+ "grad_norm": 0.21738290786743164,
666
+ "learning_rate": 4.6150000000000004e-05,
667
+ "loss": 0.6977,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 0.23,
672
+ "grad_norm": 0.21888123452663422,
673
+ "learning_rate": 4.665e-05,
674
+ "loss": 0.7083,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 0.23,
679
+ "grad_norm": 0.2516074776649475,
680
+ "learning_rate": 4.715e-05,
681
+ "loss": 0.7451,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 0.23,
686
+ "grad_norm": 0.39361557364463806,
687
+ "learning_rate": 4.765e-05,
688
+ "loss": 0.7013,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 0.23,
693
+ "grad_norm": 0.2461744099855423,
694
+ "learning_rate": 4.815e-05,
695
+ "loss": 0.7325,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 0.24,
700
+ "grad_norm": 0.2554089426994324,
701
+ "learning_rate": 4.8650000000000003e-05,
702
+ "loss": 0.7384,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 0.24,
707
+ "grad_norm": 0.2717374861240387,
708
+ "learning_rate": 4.915e-05,
709
+ "loss": 0.7172,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 0.24,
714
+ "grad_norm": 0.35919347405433655,
715
+ "learning_rate": 4.965e-05,
716
+ "loss": 0.6851,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 0.24,
721
+ "eval_loss": 0.7169972658157349,
722
+ "eval_runtime": 67.4497,
723
+ "eval_samples_per_second": 29.652,
724
+ "eval_steps_per_second": 0.934,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 0.24,
729
+ "grad_norm": 0.19895273447036743,
730
+ "learning_rate": 4.999999713608037e-05,
731
+ "loss": 0.7104,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 0.25,
736
+ "grad_norm": 0.21741873025894165,
737
+ "learning_rate": 4.999994622197174e-05,
738
+ "loss": 0.7217,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 0.25,
743
+ "grad_norm": 0.3415576219558716,
744
+ "learning_rate": 4.999983166535371e-05,
745
+ "loss": 0.716,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 0.25,
750
+ "grad_norm": 0.22503221035003662,
751
+ "learning_rate": 4.99996534665179e-05,
752
+ "loss": 0.7166,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 0.25,
757
+ "grad_norm": 0.22469323873519897,
758
+ "learning_rate": 4.999941162591795e-05,
759
+ "loss": 0.7173,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 0.26,
764
+ "grad_norm": 0.24848666787147522,
765
+ "learning_rate": 4.999910614416952e-05,
766
+ "loss": 0.7442,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 0.26,
771
+ "grad_norm": 0.23361551761627197,
772
+ "learning_rate": 4.999873702205027e-05,
773
+ "loss": 0.7142,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 0.26,
778
+ "grad_norm": 0.2513304054737091,
779
+ "learning_rate": 4.999830426049987e-05,
780
+ "loss": 0.6998,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 0.26,
785
+ "grad_norm": 0.2846020758152008,
786
+ "learning_rate": 4.999780786062003e-05,
787
+ "loss": 0.6967,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 0.27,
792
+ "grad_norm": 0.2192545235157013,
793
+ "learning_rate": 4.999724782367441e-05,
794
+ "loss": 0.7259,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 0.27,
799
+ "grad_norm": 0.37123605608940125,
800
+ "learning_rate": 4.999662415108872e-05,
801
+ "loss": 0.7228,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 0.27,
806
+ "grad_norm": 0.3963713049888611,
807
+ "learning_rate": 4.999593684445063e-05,
808
+ "loss": 0.6848,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 0.27,
813
+ "grad_norm": 0.24653227627277374,
814
+ "learning_rate": 4.9995185905509836e-05,
815
+ "loss": 0.7181,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 0.28,
820
+ "grad_norm": 0.25218069553375244,
821
+ "learning_rate": 4.999437133617799e-05,
822
+ "loss": 0.7263,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 0.28,
827
+ "grad_norm": 0.20412948727607727,
828
+ "learning_rate": 4.9993493138528765e-05,
829
+ "loss": 0.6965,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 0.28,
834
+ "grad_norm": 0.21635311841964722,
835
+ "learning_rate": 4.9992551314797775e-05,
836
+ "loss": 0.7042,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 0.28,
841
+ "grad_norm": 0.21901027858257294,
842
+ "learning_rate": 4.999154586738264e-05,
843
+ "loss": 0.6967,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 0.29,
848
+ "grad_norm": 0.24104245007038116,
849
+ "learning_rate": 4.9990476798842935e-05,
850
+ "loss": 0.7009,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 0.29,
855
+ "grad_norm": 0.2433982491493225,
856
+ "learning_rate": 4.998934411190018e-05,
857
+ "loss": 0.7052,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 0.29,
862
+ "grad_norm": 0.5261262655258179,
863
+ "learning_rate": 4.9988147809437876e-05,
864
+ "loss": 0.7103,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 0.29,
869
+ "grad_norm": 0.36266064643859863,
870
+ "learning_rate": 4.998688789450146e-05,
871
+ "loss": 0.7028,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 0.29,
876
+ "grad_norm": 0.2509474754333496,
877
+ "learning_rate": 4.9985564370298274e-05,
878
+ "loss": 0.7047,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 0.3,
883
+ "grad_norm": 0.1925598382949829,
884
+ "learning_rate": 4.9984177240197665e-05,
885
+ "loss": 0.6943,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 0.3,
890
+ "grad_norm": 0.22494031488895416,
891
+ "learning_rate": 4.998272650773083e-05,
892
+ "loss": 0.6953,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 0.3,
897
+ "grad_norm": 0.24152888357639313,
898
+ "learning_rate": 4.998121217659092e-05,
899
+ "loss": 0.7181,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 0.3,
904
+ "grad_norm": 0.24296504259109497,
905
+ "learning_rate": 4.997963425063297e-05,
906
+ "loss": 0.6934,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 0.31,
911
+ "grad_norm": 0.2438274323940277,
912
+ "learning_rate": 4.9977992733873906e-05,
913
+ "loss": 0.7088,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 0.31,
918
+ "grad_norm": 0.21081580221652985,
919
+ "learning_rate": 4.997628763049257e-05,
920
+ "loss": 0.6833,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 0.31,
925
+ "grad_norm": 0.2111068218946457,
926
+ "learning_rate": 4.9974518944829626e-05,
927
+ "loss": 0.6922,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 0.31,
932
+ "grad_norm": 0.309772253036499,
933
+ "learning_rate": 4.997268668138766e-05,
934
+ "loss": 0.7195,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 0.32,
939
+ "grad_norm": 0.2199670672416687,
940
+ "learning_rate": 4.997079084483105e-05,
941
+ "loss": 0.6924,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 0.32,
946
+ "grad_norm": 0.21781674027442932,
947
+ "learning_rate": 4.996883143998605e-05,
948
+ "loss": 0.6873,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 0.32,
953
+ "grad_norm": 0.2969052791595459,
954
+ "learning_rate": 4.996680847184072e-05,
955
+ "loss": 0.7135,
956
+ "step": 1330
957
+ },
958
+ {
959
+ "epoch": 0.32,
960
+ "grad_norm": 0.23092682659626007,
961
+ "learning_rate": 4.996472194554495e-05,
962
+ "loss": 0.6827,
963
+ "step": 1340
964
+ },
965
+ {
966
+ "epoch": 0.33,
967
+ "grad_norm": 0.29615628719329834,
968
+ "learning_rate": 4.996257186641042e-05,
969
+ "loss": 0.6738,
970
+ "step": 1350
971
+ },
972
+ {
973
+ "epoch": 0.33,
974
+ "grad_norm": 0.23491479456424713,
975
+ "learning_rate": 4.99603582399106e-05,
976
+ "loss": 0.7396,
977
+ "step": 1360
978
+ },
979
+ {
980
+ "epoch": 0.33,
981
+ "grad_norm": 0.1986425668001175,
982
+ "learning_rate": 4.9958081071680726e-05,
983
+ "loss": 0.6787,
984
+ "step": 1370
985
+ },
986
+ {
987
+ "epoch": 0.33,
988
+ "grad_norm": 0.19658410549163818,
989
+ "learning_rate": 4.99557403675178e-05,
990
+ "loss": 0.66,
991
+ "step": 1380
992
+ },
993
+ {
994
+ "epoch": 0.34,
995
+ "grad_norm": 0.29404616355895996,
996
+ "learning_rate": 4.995333613338057e-05,
997
+ "loss": 0.6824,
998
+ "step": 1390
999
+ },
1000
+ {
1001
+ "epoch": 0.34,
1002
+ "grad_norm": 0.23689547181129456,
1003
+ "learning_rate": 4.9950868375389514e-05,
1004
+ "loss": 0.7008,
1005
+ "step": 1400
1006
+ },
1007
+ {
1008
+ "epoch": 0.34,
1009
+ "grad_norm": 0.22758638858795166,
1010
+ "learning_rate": 4.99483370998268e-05,
1011
+ "loss": 0.6946,
1012
+ "step": 1410
1013
+ },
1014
+ {
1015
+ "epoch": 0.34,
1016
+ "grad_norm": 0.20259805023670197,
1017
+ "learning_rate": 4.994574231313634e-05,
1018
+ "loss": 0.6734,
1019
+ "step": 1420
1020
+ },
1021
+ {
1022
+ "epoch": 0.35,
1023
+ "grad_norm": 0.21935436129570007,
1024
+ "learning_rate": 4.994308402192366e-05,
1025
+ "loss": 0.7094,
1026
+ "step": 1430
1027
+ },
1028
+ {
1029
+ "epoch": 0.35,
1030
+ "grad_norm": 0.23414015769958496,
1031
+ "learning_rate": 4.9940362232956026e-05,
1032
+ "loss": 0.7019,
1033
+ "step": 1440
1034
+ },
1035
+ {
1036
+ "epoch": 0.35,
1037
+ "grad_norm": 0.2623121440410614,
1038
+ "learning_rate": 4.993757695316228e-05,
1039
+ "loss": 0.6793,
1040
+ "step": 1450
1041
+ },
1042
+ {
1043
+ "epoch": 0.35,
1044
+ "grad_norm": 0.31137487292289734,
1045
+ "learning_rate": 4.993472818963295e-05,
1046
+ "loss": 0.6902,
1047
+ "step": 1460
1048
+ },
1049
+ {
1050
+ "epoch": 0.36,
1051
+ "grad_norm": 0.22092680633068085,
1052
+ "learning_rate": 4.993181594962013e-05,
1053
+ "loss": 0.6912,
1054
+ "step": 1470
1055
+ },
1056
+ {
1057
+ "epoch": 0.36,
1058
+ "grad_norm": 0.23723623156547546,
1059
+ "learning_rate": 4.992884024053754e-05,
1060
+ "loss": 0.652,
1061
+ "step": 1480
1062
+ },
1063
+ {
1064
+ "epoch": 0.36,
1065
+ "grad_norm": 0.26028233766555786,
1066
+ "learning_rate": 4.9925801069960454e-05,
1067
+ "loss": 0.6969,
1068
+ "step": 1490
1069
+ },
1070
+ {
1071
+ "epoch": 0.36,
1072
+ "grad_norm": 0.23856200277805328,
1073
+ "learning_rate": 4.992269844562572e-05,
1074
+ "loss": 0.6723,
1075
+ "step": 1500
1076
+ },
1077
+ {
1078
+ "epoch": 0.36,
1079
+ "eval_loss": 0.6912096738815308,
1080
+ "eval_runtime": 67.244,
1081
+ "eval_samples_per_second": 29.742,
1082
+ "eval_steps_per_second": 0.937,
1083
+ "step": 1500
1084
+ },
1085
+ {
1086
+ "epoch": 0.36,
1087
+ "grad_norm": 0.167326420545578,
1088
+ "learning_rate": 4.9919532375431677e-05,
1089
+ "loss": 0.6897,
1090
+ "step": 1510
1091
+ },
1092
+ {
1093
+ "epoch": 0.37,
1094
+ "grad_norm": 0.26183101534843445,
1095
+ "learning_rate": 4.991630286743823e-05,
1096
+ "loss": 0.6844,
1097
+ "step": 1520
1098
+ },
1099
+ {
1100
+ "epoch": 0.37,
1101
+ "grad_norm": 0.25918036699295044,
1102
+ "learning_rate": 4.991300992986676e-05,
1103
+ "loss": 0.6645,
1104
+ "step": 1530
1105
+ },
1106
+ {
1107
+ "epoch": 0.37,
1108
+ "grad_norm": 0.2354690134525299,
1109
+ "learning_rate": 4.99096535711001e-05,
1110
+ "loss": 0.6672,
1111
+ "step": 1540
1112
+ },
1113
+ {
1114
+ "epoch": 0.37,
1115
+ "grad_norm": 0.24444998800754547,
1116
+ "learning_rate": 4.990623379968257e-05,
1117
+ "loss": 0.6749,
1118
+ "step": 1550
1119
+ },
1120
+ {
1121
+ "epoch": 0.38,
1122
+ "grad_norm": 0.2562585771083832,
1123
+ "learning_rate": 4.990275062431989e-05,
1124
+ "loss": 0.6652,
1125
+ "step": 1560
1126
+ },
1127
+ {
1128
+ "epoch": 0.38,
1129
+ "grad_norm": 0.22618699073791504,
1130
+ "learning_rate": 4.98992040538792e-05,
1131
+ "loss": 0.7163,
1132
+ "step": 1570
1133
+ },
1134
+ {
1135
+ "epoch": 0.38,
1136
+ "grad_norm": 0.20957091450691223,
1137
+ "learning_rate": 4.9895594097389044e-05,
1138
+ "loss": 0.7053,
1139
+ "step": 1580
1140
+ },
1141
+ {
1142
+ "epoch": 0.38,
1143
+ "grad_norm": 0.26973310112953186,
1144
+ "learning_rate": 4.989192076403928e-05,
1145
+ "loss": 0.6716,
1146
+ "step": 1590
1147
+ },
1148
+ {
1149
+ "epoch": 0.39,
1150
+ "grad_norm": 0.24841250479221344,
1151
+ "learning_rate": 4.9888184063181154e-05,
1152
+ "loss": 0.6884,
1153
+ "step": 1600
1154
+ },
1155
+ {
1156
+ "epoch": 0.39,
1157
+ "grad_norm": 0.25283706188201904,
1158
+ "learning_rate": 4.98843840043272e-05,
1159
+ "loss": 0.6632,
1160
+ "step": 1610
1161
+ },
1162
+ {
1163
+ "epoch": 0.39,
1164
+ "grad_norm": 0.196313738822937,
1165
+ "learning_rate": 4.988052059715126e-05,
1166
+ "loss": 0.712,
1167
+ "step": 1620
1168
+ },
1169
+ {
1170
+ "epoch": 0.39,
1171
+ "grad_norm": 0.21621187031269073,
1172
+ "learning_rate": 4.987659385148842e-05,
1173
+ "loss": 0.6732,
1174
+ "step": 1630
1175
+ },
1176
+ {
1177
+ "epoch": 0.4,
1178
+ "grad_norm": 0.2180820256471634,
1179
+ "learning_rate": 4.987260377733502e-05,
1180
+ "loss": 0.7045,
1181
+ "step": 1640
1182
+ },
1183
+ {
1184
+ "epoch": 0.4,
1185
+ "grad_norm": 0.20110243558883667,
1186
+ "learning_rate": 4.986855038484862e-05,
1187
+ "loss": 0.6604,
1188
+ "step": 1650
1189
+ },
1190
+ {
1191
+ "epoch": 0.4,
1192
+ "grad_norm": 0.20813095569610596,
1193
+ "learning_rate": 4.9864433684347964e-05,
1194
+ "loss": 0.7051,
1195
+ "step": 1660
1196
+ },
1197
+ {
1198
+ "epoch": 0.4,
1199
+ "grad_norm": 0.22026021778583527,
1200
+ "learning_rate": 4.9860253686312964e-05,
1201
+ "loss": 0.6752,
1202
+ "step": 1670
1203
+ },
1204
+ {
1205
+ "epoch": 0.41,
1206
+ "grad_norm": 0.23409809172153473,
1207
+ "learning_rate": 4.9856010401384654e-05,
1208
+ "loss": 0.6805,
1209
+ "step": 1680
1210
+ },
1211
+ {
1212
+ "epoch": 0.41,
1213
+ "grad_norm": 0.1982721984386444,
1214
+ "learning_rate": 4.985170384036521e-05,
1215
+ "loss": 0.6888,
1216
+ "step": 1690
1217
+ },
1218
+ {
1219
+ "epoch": 0.41,
1220
+ "grad_norm": 0.24984286725521088,
1221
+ "learning_rate": 4.984733401421785e-05,
1222
+ "loss": 0.6741,
1223
+ "step": 1700
1224
+ },
1225
+ {
1226
+ "epoch": 0.41,
1227
+ "grad_norm": 0.24465042352676392,
1228
+ "learning_rate": 4.9842900934066874e-05,
1229
+ "loss": 0.6781,
1230
+ "step": 1710
1231
+ },
1232
+ {
1233
+ "epoch": 0.42,
1234
+ "grad_norm": 0.26656293869018555,
1235
+ "learning_rate": 4.98384046111976e-05,
1236
+ "loss": 0.7086,
1237
+ "step": 1720
1238
+ },
1239
+ {
1240
+ "epoch": 0.42,
1241
+ "grad_norm": 0.23073184490203857,
1242
+ "learning_rate": 4.9833845057056336e-05,
1243
+ "loss": 0.6966,
1244
+ "step": 1730
1245
+ },
1246
+ {
1247
+ "epoch": 0.42,
1248
+ "grad_norm": 0.23296333849430084,
1249
+ "learning_rate": 4.982922228325037e-05,
1250
+ "loss": 0.6769,
1251
+ "step": 1740
1252
+ },
1253
+ {
1254
+ "epoch": 0.42,
1255
+ "grad_norm": 0.22282017767429352,
1256
+ "learning_rate": 4.982453630154794e-05,
1257
+ "loss": 0.6581,
1258
+ "step": 1750
1259
+ },
1260
+ {
1261
+ "epoch": 0.43,
1262
+ "grad_norm": 0.22943562269210815,
1263
+ "learning_rate": 4.981978712387815e-05,
1264
+ "loss": 0.6803,
1265
+ "step": 1760
1266
+ },
1267
+ {
1268
+ "epoch": 0.43,
1269
+ "grad_norm": 0.29295608401298523,
1270
+ "learning_rate": 4.9814974762331034e-05,
1271
+ "loss": 0.6514,
1272
+ "step": 1770
1273
+ },
1274
+ {
1275
+ "epoch": 0.43,
1276
+ "grad_norm": 0.21845972537994385,
1277
+ "learning_rate": 4.981009922915743e-05,
1278
+ "loss": 0.691,
1279
+ "step": 1780
1280
+ },
1281
+ {
1282
+ "epoch": 0.43,
1283
+ "grad_norm": 0.24616730213165283,
1284
+ "learning_rate": 4.980516053676903e-05,
1285
+ "loss": 0.6948,
1286
+ "step": 1790
1287
+ },
1288
+ {
1289
+ "epoch": 0.43,
1290
+ "grad_norm": 0.2354935109615326,
1291
+ "learning_rate": 4.9800158697738264e-05,
1292
+ "loss": 0.6778,
1293
+ "step": 1800
1294
+ },
1295
+ {
1296
+ "epoch": 0.44,
1297
+ "grad_norm": 0.3475668430328369,
1298
+ "learning_rate": 4.979509372479837e-05,
1299
+ "loss": 0.6626,
1300
+ "step": 1810
1301
+ },
1302
+ {
1303
+ "epoch": 0.44,
1304
+ "grad_norm": 0.2817700207233429,
1305
+ "learning_rate": 4.9789965630843265e-05,
1306
+ "loss": 0.6724,
1307
+ "step": 1820
1308
+ },
1309
+ {
1310
+ "epoch": 0.44,
1311
+ "grad_norm": 0.28306934237480164,
1312
+ "learning_rate": 4.978477442892758e-05,
1313
+ "loss": 0.6673,
1314
+ "step": 1830
1315
+ },
1316
+ {
1317
+ "epoch": 0.44,
1318
+ "grad_norm": 0.2593139111995697,
1319
+ "learning_rate": 4.9779520132266575e-05,
1320
+ "loss": 0.6866,
1321
+ "step": 1840
1322
+ },
1323
+ {
1324
+ "epoch": 0.45,
1325
+ "grad_norm": 0.2618165612220764,
1326
+ "learning_rate": 4.9774202754236145e-05,
1327
+ "loss": 0.6671,
1328
+ "step": 1850
1329
+ },
1330
+ {
1331
+ "epoch": 0.45,
1332
+ "grad_norm": 0.22875259816646576,
1333
+ "learning_rate": 4.9768822308372784e-05,
1334
+ "loss": 0.6743,
1335
+ "step": 1860
1336
+ },
1337
+ {
1338
+ "epoch": 0.45,
1339
+ "grad_norm": 0.20143887400627136,
1340
+ "learning_rate": 4.976337880837351e-05,
1341
+ "loss": 0.6866,
1342
+ "step": 1870
1343
+ },
1344
+ {
1345
+ "epoch": 0.45,
1346
+ "grad_norm": 0.23163306713104248,
1347
+ "learning_rate": 4.975787226809587e-05,
1348
+ "loss": 0.6772,
1349
+ "step": 1880
1350
+ },
1351
+ {
1352
+ "epoch": 0.46,
1353
+ "grad_norm": 0.21628862619400024,
1354
+ "learning_rate": 4.975230270155791e-05,
1355
+ "loss": 0.6706,
1356
+ "step": 1890
1357
+ },
1358
+ {
1359
+ "epoch": 0.46,
1360
+ "grad_norm": 0.21055322885513306,
1361
+ "learning_rate": 4.9746670122938105e-05,
1362
+ "loss": 0.6536,
1363
+ "step": 1900
1364
+ },
1365
+ {
1366
+ "epoch": 0.46,
1367
+ "grad_norm": 0.20045976340770721,
1368
+ "learning_rate": 4.974097454657534e-05,
1369
+ "loss": 0.6477,
1370
+ "step": 1910
1371
+ },
1372
+ {
1373
+ "epoch": 0.46,
1374
+ "grad_norm": 0.20996223390102386,
1375
+ "learning_rate": 4.9735215986968874e-05,
1376
+ "loss": 0.6526,
1377
+ "step": 1920
1378
+ },
1379
+ {
1380
+ "epoch": 0.47,
1381
+ "grad_norm": 0.1778746396303177,
1382
+ "learning_rate": 4.972939445877831e-05,
1383
+ "loss": 0.6951,
1384
+ "step": 1930
1385
+ },
1386
+ {
1387
+ "epoch": 0.47,
1388
+ "grad_norm": 0.2416403591632843,
1389
+ "learning_rate": 4.972350997682354e-05,
1390
+ "loss": 0.6703,
1391
+ "step": 1940
1392
+ },
1393
+ {
1394
+ "epoch": 0.47,
1395
+ "grad_norm": 0.23652783036231995,
1396
+ "learning_rate": 4.9717562556084735e-05,
1397
+ "loss": 0.6401,
1398
+ "step": 1950
1399
+ },
1400
+ {
1401
+ "epoch": 0.47,
1402
+ "grad_norm": 0.23598232865333557,
1403
+ "learning_rate": 4.9711552211702274e-05,
1404
+ "loss": 0.6668,
1405
+ "step": 1960
1406
+ },
1407
+ {
1408
+ "epoch": 0.48,
1409
+ "grad_norm": 0.22741574048995972,
1410
+ "learning_rate": 4.970547895897672e-05,
1411
+ "loss": 0.6765,
1412
+ "step": 1970
1413
+ },
1414
+ {
1415
+ "epoch": 0.48,
1416
+ "grad_norm": 0.20418991148471832,
1417
+ "learning_rate": 4.96993428133688e-05,
1418
+ "loss": 0.6771,
1419
+ "step": 1980
1420
+ },
1421
+ {
1422
+ "epoch": 0.48,
1423
+ "grad_norm": 0.24983160197734833,
1424
+ "learning_rate": 4.969314379049932e-05,
1425
+ "loss": 0.6656,
1426
+ "step": 1990
1427
+ },
1428
+ {
1429
+ "epoch": 0.48,
1430
+ "grad_norm": 0.2511340379714966,
1431
+ "learning_rate": 4.968688190614919e-05,
1432
+ "loss": 0.6605,
1433
+ "step": 2000
1434
+ },
1435
+ {
1436
+ "epoch": 0.48,
1437
+ "eval_loss": 0.6730201244354248,
1438
+ "eval_runtime": 67.2807,
1439
+ "eval_samples_per_second": 29.726,
1440
+ "eval_steps_per_second": 0.936,
1441
+ "step": 2000
1442
+ }
1443
+ ],
1444
+ "logging_steps": 10,
1445
+ "max_steps": 20690,
1446
+ "num_input_tokens_seen": 0,
1447
+ "num_train_epochs": 5,
1448
+ "save_steps": 500,
1449
+ "total_flos": 4.3244811194046874e+18,
1450
+ "train_batch_size": 4,
1451
+ "trial_name": null,
1452
+ "trial_params": null
1453
+ }
checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd6a31d2da7b4686068d511137b843e541147626d5878a410233f3ef91ae5dc6
3
+ size 5048