Hugging Face's logo

HydraLM
/
Nous-Hermes-llama-2-7b_7b_cluster00_partitioned_v3_standardized_00

PEFT
Model card Files Community
1
prateeky2806 commited on
Commit
85cbf92
•
1 Parent(s): 5fa6163

Training in progress, step 2400

Browse files
Files changed (15) hide show
  1. adapter_model.bin +1 -1
  2. checkpoint-1800/adapter_model/adapter_model/README.md +12 -0
  3. checkpoint-1800/adapter_model/adapter_model/adapter_model.bin +1 -1
  4. checkpoint-2400/README.md +20 -0
  5. checkpoint-2400/adapter_config.json +26 -0
  6. checkpoint-2400/adapter_model.bin +3 -0
  7. checkpoint-2400/added_tokens.json +3 -0
  8. checkpoint-2400/optimizer.pt +3 -0
  9. checkpoint-2400/rng_state.pth +3 -0
  10. checkpoint-2400/scheduler.pt +3 -0
  11. checkpoint-2400/special_tokens_map.json +6 -0
  12. checkpoint-2400/tokenizer.model +3 -0
  13. checkpoint-2400/tokenizer_config.json +35 -0
  14. checkpoint-2400/trainer_state.json +1552 -0
  15. checkpoint-2400/training_args.bin +3 -0
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:92a47b38995045c235c141bcaac2c871875999c7e490f58537e58f07bc23d0c2
3
  size 319977229
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e7f504a19b35b14f25776bcb6de5985bef84cd6530e91e3af5b5ba0ce65c7ab
3
  size 319977229
checkpoint-1800/adapter_model/adapter_model/README.md CHANGED
@@ -15,6 +15,17 @@ The following `bitsandbytes` quantization config was used during training:
15
  - bnb_4bit_use_double_quant: True
16
  - bnb_4bit_compute_dtype: bfloat16
17
 
 
 
 
 
 
 
 
 
 
 
 
18
  The following `bitsandbytes` quantization config was used during training:
19
  - load_in_8bit: False
20
  - load_in_4bit: True
@@ -27,6 +38,7 @@ The following `bitsandbytes` quantization config was used during training:
27
  - bnb_4bit_compute_dtype: bfloat16
28
  ### Framework versions
29
 
 
30
  - PEFT 0.4.0
31
 
32
  - PEFT 0.4.0
 
15
  - bnb_4bit_use_double_quant: True
16
  - bnb_4bit_compute_dtype: bfloat16
17
 
18
+ The following `bitsandbytes` quantization config was used during training:
19
+ - load_in_8bit: False
20
+ - load_in_4bit: True
21
+ - llm_int8_threshold: 6.0
22
+ - llm_int8_skip_modules: None
23
+ - llm_int8_enable_fp32_cpu_offload: False
24
+ - llm_int8_has_fp16_weight: False
25
+ - bnb_4bit_quant_type: nf4
26
+ - bnb_4bit_use_double_quant: True
27
+ - bnb_4bit_compute_dtype: bfloat16
28
+
29
  The following `bitsandbytes` quantization config was used during training:
30
  - load_in_8bit: False
31
  - load_in_4bit: True
 
38
  - bnb_4bit_compute_dtype: bfloat16
39
  ### Framework versions
40
 
41
+ - PEFT 0.4.0
42
  - PEFT 0.4.0
43
 
44
  - PEFT 0.4.0
checkpoint-1800/adapter_model/adapter_model/adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5fd7d4274435a21ad815f03e386697a403de379162ca13f2401fa609f98dc802
3
  size 319977229
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92a47b38995045c235c141bcaac2c871875999c7e490f58537e58f07bc23d0c2
3
  size 319977229
checkpoint-2400/README.md ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: False
9
+ - load_in_4bit: True
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: nf4
15
+ - bnb_4bit_use_double_quant: True
16
+ - bnb_4bit_compute_dtype: bfloat16
17
+ ### Framework versions
18
+
19
+
20
+ - PEFT 0.4.0
checkpoint-2400/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "NousResearch/Nous-Hermes-llama-2-7b",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 64,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "gate_proj",
18
+ "o_proj",
19
+ "k_proj",
20
+ "up_proj",
21
+ "v_proj",
22
+ "q_proj",
23
+ "down_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
checkpoint-2400/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e7f504a19b35b14f25776bcb6de5985bef84cd6530e91e3af5b5ba0ce65c7ab
3
+ size 319977229
checkpoint-2400/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<pad>": 32000
3
+ }
checkpoint-2400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:48f37d55fc0af2364983d934d9278ba4598541601996ae86f6445f35bc3c22c3
3
+ size 1279539973
checkpoint-2400/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d58b849dffce73559cd997ac106283c21c22cce88cad7c274ea20f91eb1bc29a
3
+ size 14511
checkpoint-2400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a78e728336b77d3305afe2734acb1241c76f8f2c33b35460bc07397abc49ba2
3
+ size 627
checkpoint-2400/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "<unk>",
5
+ "unk_token": "<unk>"
6
+ }
checkpoint-2400/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-2400/tokenizer_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "bos_token": {
5
+ "__type": "AddedToken",
6
+ "content": "<s>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "clean_up_tokenization_spaces": false,
13
+ "eos_token": {
14
+ "__type": "AddedToken",
15
+ "content": "</s>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "legacy": false,
22
+ "model_max_length": 1000000000000000019884624838656,
23
+ "pad_token": null,
24
+ "padding_side": "right",
25
+ "sp_model_kwargs": {},
26
+ "tokenizer_class": "LlamaTokenizer",
27
+ "unk_token": {
28
+ "__type": "AddedToken",
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false
34
+ }
35
+ }
checkpoint-2400/trainer_state.json ADDED
@@ -0,0 +1,1552 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.39367392659187317,
3
+ "best_model_checkpoint": "./output_v2/7b_cluster00_Nous-Hermes-llama-2-7b_partitioned_v3_standardized_00/checkpoint-1800",
4
+ "epoch": 2.545068928950159,
5
+ "global_step": 2400,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 0.0002,
13
+ "loss": 0.5637,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 0.0002,
19
+ "loss": 0.4775,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.03,
24
+ "learning_rate": 0.0002,
25
+ "loss": 0.5079,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.04,
30
+ "learning_rate": 0.0002,
31
+ "loss": 0.4849,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.05,
36
+ "learning_rate": 0.0002,
37
+ "loss": 0.4587,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.06,
42
+ "learning_rate": 0.0002,
43
+ "loss": 0.4867,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.07,
48
+ "learning_rate": 0.0002,
49
+ "loss": 0.4578,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.08,
54
+ "learning_rate": 0.0002,
55
+ "loss": 0.4665,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.1,
60
+ "learning_rate": 0.0002,
61
+ "loss": 0.5044,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.11,
66
+ "learning_rate": 0.0002,
67
+ "loss": 0.4979,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.12,
72
+ "learning_rate": 0.0002,
73
+ "loss": 0.4715,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.13,
78
+ "learning_rate": 0.0002,
79
+ "loss": 0.4717,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.14,
84
+ "learning_rate": 0.0002,
85
+ "loss": 0.4419,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.15,
90
+ "learning_rate": 0.0002,
91
+ "loss": 0.4671,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.16,
96
+ "learning_rate": 0.0002,
97
+ "loss": 0.4503,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.17,
102
+ "learning_rate": 0.0002,
103
+ "loss": 0.4878,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.18,
108
+ "learning_rate": 0.0002,
109
+ "loss": 0.4648,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.19,
114
+ "learning_rate": 0.0002,
115
+ "loss": 0.4625,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.2,
120
+ "learning_rate": 0.0002,
121
+ "loss": 0.4473,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.21,
126
+ "learning_rate": 0.0002,
127
+ "loss": 0.4585,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.21,
132
+ "eval_loss": 0.4498298466205597,
133
+ "eval_runtime": 253.5551,
134
+ "eval_samples_per_second": 3.944,
135
+ "eval_steps_per_second": 0.986,
136
+ "step": 200
137
+ },
138
+ {
139
+ "epoch": 0.22,
140
+ "learning_rate": 0.0002,
141
+ "loss": 0.4396,
142
+ "step": 210
143
+ },
144
+ {
145
+ "epoch": 0.23,
146
+ "learning_rate": 0.0002,
147
+ "loss": 0.4545,
148
+ "step": 220
149
+ },
150
+ {
151
+ "epoch": 0.24,
152
+ "learning_rate": 0.0002,
153
+ "loss": 0.4632,
154
+ "step": 230
155
+ },
156
+ {
157
+ "epoch": 0.25,
158
+ "learning_rate": 0.0002,
159
+ "loss": 0.4458,
160
+ "step": 240
161
+ },
162
+ {
163
+ "epoch": 0.27,
164
+ "learning_rate": 0.0002,
165
+ "loss": 0.4612,
166
+ "step": 250
167
+ },
168
+ {
169
+ "epoch": 0.28,
170
+ "learning_rate": 0.0002,
171
+ "loss": 0.4493,
172
+ "step": 260
173
+ },
174
+ {
175
+ "epoch": 0.29,
176
+ "learning_rate": 0.0002,
177
+ "loss": 0.4652,
178
+ "step": 270
179
+ },
180
+ {
181
+ "epoch": 0.3,
182
+ "learning_rate": 0.0002,
183
+ "loss": 0.4596,
184
+ "step": 280
185
+ },
186
+ {
187
+ "epoch": 0.31,
188
+ "learning_rate": 0.0002,
189
+ "loss": 0.4483,
190
+ "step": 290
191
+ },
192
+ {
193
+ "epoch": 0.32,
194
+ "learning_rate": 0.0002,
195
+ "loss": 0.4386,
196
+ "step": 300
197
+ },
198
+ {
199
+ "epoch": 0.33,
200
+ "learning_rate": 0.0002,
201
+ "loss": 0.4746,
202
+ "step": 310
203
+ },
204
+ {
205
+ "epoch": 0.34,
206
+ "learning_rate": 0.0002,
207
+ "loss": 0.4299,
208
+ "step": 320
209
+ },
210
+ {
211
+ "epoch": 0.35,
212
+ "learning_rate": 0.0002,
213
+ "loss": 0.4462,
214
+ "step": 330
215
+ },
216
+ {
217
+ "epoch": 0.36,
218
+ "learning_rate": 0.0002,
219
+ "loss": 0.4346,
220
+ "step": 340
221
+ },
222
+ {
223
+ "epoch": 0.37,
224
+ "learning_rate": 0.0002,
225
+ "loss": 0.4425,
226
+ "step": 350
227
+ },
228
+ {
229
+ "epoch": 0.38,
230
+ "learning_rate": 0.0002,
231
+ "loss": 0.4356,
232
+ "step": 360
233
+ },
234
+ {
235
+ "epoch": 0.39,
236
+ "learning_rate": 0.0002,
237
+ "loss": 0.4647,
238
+ "step": 370
239
+ },
240
+ {
241
+ "epoch": 0.4,
242
+ "learning_rate": 0.0002,
243
+ "loss": 0.4448,
244
+ "step": 380
245
+ },
246
+ {
247
+ "epoch": 0.41,
248
+ "learning_rate": 0.0002,
249
+ "loss": 0.4208,
250
+ "step": 390
251
+ },
252
+ {
253
+ "epoch": 0.42,
254
+ "learning_rate": 0.0002,
255
+ "loss": 0.4095,
256
+ "step": 400
257
+ },
258
+ {
259
+ "epoch": 0.42,
260
+ "eval_loss": 0.43381235003471375,
261
+ "eval_runtime": 253.5754,
262
+ "eval_samples_per_second": 3.944,
263
+ "eval_steps_per_second": 0.986,
264
+ "step": 400
265
+ },
266
+ {
267
+ "epoch": 0.43,
268
+ "learning_rate": 0.0002,
269
+ "loss": 0.4181,
270
+ "step": 410
271
+ },
272
+ {
273
+ "epoch": 0.45,
274
+ "learning_rate": 0.0002,
275
+ "loss": 0.445,
276
+ "step": 420
277
+ },
278
+ {
279
+ "epoch": 0.46,
280
+ "learning_rate": 0.0002,
281
+ "loss": 0.417,
282
+ "step": 430
283
+ },
284
+ {
285
+ "epoch": 0.47,
286
+ "learning_rate": 0.0002,
287
+ "loss": 0.438,
288
+ "step": 440
289
+ },
290
+ {
291
+ "epoch": 0.48,
292
+ "learning_rate": 0.0002,
293
+ "loss": 0.4376,
294
+ "step": 450
295
+ },
296
+ {
297
+ "epoch": 0.49,
298
+ "learning_rate": 0.0002,
299
+ "loss": 0.4249,
300
+ "step": 460
301
+ },
302
+ {
303
+ "epoch": 0.5,
304
+ "learning_rate": 0.0002,
305
+ "loss": 0.4407,
306
+ "step": 470
307
+ },
308
+ {
309
+ "epoch": 0.51,
310
+ "learning_rate": 0.0002,
311
+ "loss": 0.4506,
312
+ "step": 480
313
+ },
314
+ {
315
+ "epoch": 0.52,
316
+ "learning_rate": 0.0002,
317
+ "loss": 0.4289,
318
+ "step": 490
319
+ },
320
+ {
321
+ "epoch": 0.53,
322
+ "learning_rate": 0.0002,
323
+ "loss": 0.4596,
324
+ "step": 500
325
+ },
326
+ {
327
+ "epoch": 0.54,
328
+ "learning_rate": 0.0002,
329
+ "loss": 0.4174,
330
+ "step": 510
331
+ },
332
+ {
333
+ "epoch": 0.55,
334
+ "learning_rate": 0.0002,
335
+ "loss": 0.4373,
336
+ "step": 520
337
+ },
338
+ {
339
+ "epoch": 0.56,
340
+ "learning_rate": 0.0002,
341
+ "loss": 0.4253,
342
+ "step": 530
343
+ },
344
+ {
345
+ "epoch": 0.57,
346
+ "learning_rate": 0.0002,
347
+ "loss": 0.4243,
348
+ "step": 540
349
+ },
350
+ {
351
+ "epoch": 0.58,
352
+ "learning_rate": 0.0002,
353
+ "loss": 0.4455,
354
+ "step": 550
355
+ },
356
+ {
357
+ "epoch": 0.59,
358
+ "learning_rate": 0.0002,
359
+ "loss": 0.4362,
360
+ "step": 560
361
+ },
362
+ {
363
+ "epoch": 0.6,
364
+ "learning_rate": 0.0002,
365
+ "loss": 0.4364,
366
+ "step": 570
367
+ },
368
+ {
369
+ "epoch": 0.62,
370
+ "learning_rate": 0.0002,
371
+ "loss": 0.4257,
372
+ "step": 580
373
+ },
374
+ {
375
+ "epoch": 0.63,
376
+ "learning_rate": 0.0002,
377
+ "loss": 0.4193,
378
+ "step": 590
379
+ },
380
+ {
381
+ "epoch": 0.64,
382
+ "learning_rate": 0.0002,
383
+ "loss": 0.4233,
384
+ "step": 600
385
+ },
386
+ {
387
+ "epoch": 0.64,
388
+ "eval_loss": 0.42340007424354553,
389
+ "eval_runtime": 253.6896,
390
+ "eval_samples_per_second": 3.942,
391
+ "eval_steps_per_second": 0.985,
392
+ "step": 600
393
+ },
394
+ {
395
+ "epoch": 0.65,
396
+ "learning_rate": 0.0002,
397
+ "loss": 0.4047,
398
+ "step": 610
399
+ },
400
+ {
401
+ "epoch": 0.66,
402
+ "learning_rate": 0.0002,
403
+ "loss": 0.4329,
404
+ "step": 620
405
+ },
406
+ {
407
+ "epoch": 0.67,
408
+ "learning_rate": 0.0002,
409
+ "loss": 0.4423,
410
+ "step": 630
411
+ },
412
+ {
413
+ "epoch": 0.68,
414
+ "learning_rate": 0.0002,
415
+ "loss": 0.4192,
416
+ "step": 640
417
+ },
418
+ {
419
+ "epoch": 0.69,
420
+ "learning_rate": 0.0002,
421
+ "loss": 0.4183,
422
+ "step": 650
423
+ },
424
+ {
425
+ "epoch": 0.7,
426
+ "learning_rate": 0.0002,
427
+ "loss": 0.4268,
428
+ "step": 660
429
+ },
430
+ {
431
+ "epoch": 0.71,
432
+ "learning_rate": 0.0002,
433
+ "loss": 0.425,
434
+ "step": 670
435
+ },
436
+ {
437
+ "epoch": 0.72,
438
+ "learning_rate": 0.0002,
439
+ "loss": 0.4353,
440
+ "step": 680
441
+ },
442
+ {
443
+ "epoch": 0.73,
444
+ "learning_rate": 0.0002,
445
+ "loss": 0.411,
446
+ "step": 690
447
+ },
448
+ {
449
+ "epoch": 0.74,
450
+ "learning_rate": 0.0002,
451
+ "loss": 0.4155,
452
+ "step": 700
453
+ },
454
+ {
455
+ "epoch": 0.75,
456
+ "learning_rate": 0.0002,
457
+ "loss": 0.4066,
458
+ "step": 710
459
+ },
460
+ {
461
+ "epoch": 0.76,
462
+ "learning_rate": 0.0002,
463
+ "loss": 0.4216,
464
+ "step": 720
465
+ },
466
+ {
467
+ "epoch": 0.77,
468
+ "learning_rate": 0.0002,
469
+ "loss": 0.4294,
470
+ "step": 730
471
+ },
472
+ {
473
+ "epoch": 0.78,
474
+ "learning_rate": 0.0002,
475
+ "loss": 0.427,
476
+ "step": 740
477
+ },
478
+ {
479
+ "epoch": 0.8,
480
+ "learning_rate": 0.0002,
481
+ "loss": 0.4276,
482
+ "step": 750
483
+ },
484
+ {
485
+ "epoch": 0.81,
486
+ "learning_rate": 0.0002,
487
+ "loss": 0.4397,
488
+ "step": 760
489
+ },
490
+ {
491
+ "epoch": 0.82,
492
+ "learning_rate": 0.0002,
493
+ "loss": 0.4264,
494
+ "step": 770
495
+ },
496
+ {
497
+ "epoch": 0.83,
498
+ "learning_rate": 0.0002,
499
+ "loss": 0.4021,
500
+ "step": 780
501
+ },
502
+ {
503
+ "epoch": 0.84,
504
+ "learning_rate": 0.0002,
505
+ "loss": 0.4132,
506
+ "step": 790
507
+ },
508
+ {
509
+ "epoch": 0.85,
510
+ "learning_rate": 0.0002,
511
+ "loss": 0.4167,
512
+ "step": 800
513
+ },
514
+ {
515
+ "epoch": 0.85,
516
+ "eval_loss": 0.4146576523780823,
517
+ "eval_runtime": 253.792,
518
+ "eval_samples_per_second": 3.94,
519
+ "eval_steps_per_second": 0.985,
520
+ "step": 800
521
+ },
522
+ {
523
+ "epoch": 0.86,
524
+ "learning_rate": 0.0002,
525
+ "loss": 0.4234,
526
+ "step": 810
527
+ },
528
+ {
529
+ "epoch": 0.87,
530
+ "learning_rate": 0.0002,
531
+ "loss": 0.4192,
532
+ "step": 820
533
+ },
534
+ {
535
+ "epoch": 0.88,
536
+ "learning_rate": 0.0002,
537
+ "loss": 0.4132,
538
+ "step": 830
539
+ },
540
+ {
541
+ "epoch": 0.89,
542
+ "learning_rate": 0.0002,
543
+ "loss": 0.4091,
544
+ "step": 840
545
+ },
546
+ {
547
+ "epoch": 0.9,
548
+ "learning_rate": 0.0002,
549
+ "loss": 0.4153,
550
+ "step": 850
551
+ },
552
+ {
553
+ "epoch": 0.91,
554
+ "learning_rate": 0.0002,
555
+ "loss": 0.417,
556
+ "step": 860
557
+ },
558
+ {
559
+ "epoch": 0.92,
560
+ "learning_rate": 0.0002,
561
+ "loss": 0.4004,
562
+ "step": 870
563
+ },
564
+ {
565
+ "epoch": 0.93,
566
+ "learning_rate": 0.0002,
567
+ "loss": 0.4282,
568
+ "step": 880
569
+ },
570
+ {
571
+ "epoch": 0.94,
572
+ "learning_rate": 0.0002,
573
+ "loss": 0.4334,
574
+ "step": 890
575
+ },
576
+ {
577
+ "epoch": 0.95,
578
+ "learning_rate": 0.0002,
579
+ "loss": 0.4042,
580
+ "step": 900
581
+ },
582
+ {
583
+ "epoch": 0.97,
584
+ "learning_rate": 0.0002,
585
+ "loss": 0.4159,
586
+ "step": 910
587
+ },
588
+ {
589
+ "epoch": 0.98,
590
+ "learning_rate": 0.0002,
591
+ "loss": 0.3986,
592
+ "step": 920
593
+ },
594
+ {
595
+ "epoch": 0.99,
596
+ "learning_rate": 0.0002,
597
+ "loss": 0.4207,
598
+ "step": 930
599
+ },
600
+ {
601
+ "epoch": 1.0,
602
+ "learning_rate": 0.0002,
603
+ "loss": 0.4144,
604
+ "step": 940
605
+ },
606
+ {
607
+ "epoch": 1.01,
608
+ "learning_rate": 0.0002,
609
+ "loss": 0.3744,
610
+ "step": 950
611
+ },
612
+ {
613
+ "epoch": 1.02,
614
+ "learning_rate": 0.0002,
615
+ "loss": 0.3685,
616
+ "step": 960
617
+ },
618
+ {
619
+ "epoch": 1.03,
620
+ "learning_rate": 0.0002,
621
+ "loss": 0.3812,
622
+ "step": 970
623
+ },
624
+ {
625
+ "epoch": 1.04,
626
+ "learning_rate": 0.0002,
627
+ "loss": 0.3604,
628
+ "step": 980
629
+ },
630
+ {
631
+ "epoch": 1.05,
632
+ "learning_rate": 0.0002,
633
+ "loss": 0.3595,
634
+ "step": 990
635
+ },
636
+ {
637
+ "epoch": 1.06,
638
+ "learning_rate": 0.0002,
639
+ "loss": 0.3659,
640
+ "step": 1000
641
+ },
642
+ {
643
+ "epoch": 1.06,
644
+ "eval_loss": 0.4106157422065735,
645
+ "eval_runtime": 253.7163,
646
+ "eval_samples_per_second": 3.941,
647
+ "eval_steps_per_second": 0.985,
648
+ "step": 1000
649
+ },
650
+ {
651
+ "epoch": 1.07,
652
+ "learning_rate": 0.0002,
653
+ "loss": 0.3586,
654
+ "step": 1010
655
+ },
656
+ {
657
+ "epoch": 1.08,
658
+ "learning_rate": 0.0002,
659
+ "loss": 0.3645,
660
+ "step": 1020
661
+ },
662
+ {
663
+ "epoch": 1.09,
664
+ "learning_rate": 0.0002,
665
+ "loss": 0.3894,
666
+ "step": 1030
667
+ },
668
+ {
669
+ "epoch": 1.1,
670
+ "learning_rate": 0.0002,
671
+ "loss": 0.3591,
672
+ "step": 1040
673
+ },
674
+ {
675
+ "epoch": 1.11,
676
+ "learning_rate": 0.0002,
677
+ "loss": 0.363,
678
+ "step": 1050
679
+ },
680
+ {
681
+ "epoch": 1.12,
682
+ "learning_rate": 0.0002,
683
+ "loss": 0.3654,
684
+ "step": 1060
685
+ },
686
+ {
687
+ "epoch": 1.13,
688
+ "learning_rate": 0.0002,
689
+ "loss": 0.3903,
690
+ "step": 1070
691
+ },
692
+ {
693
+ "epoch": 1.15,
694
+ "learning_rate": 0.0002,
695
+ "loss": 0.3521,
696
+ "step": 1080
697
+ },
698
+ {
699
+ "epoch": 1.16,
700
+ "learning_rate": 0.0002,
701
+ "loss": 0.3698,
702
+ "step": 1090
703
+ },
704
+ {
705
+ "epoch": 1.17,
706
+ "learning_rate": 0.0002,
707
+ "loss": 0.3525,
708
+ "step": 1100
709
+ },
710
+ {
711
+ "epoch": 1.18,
712
+ "learning_rate": 0.0002,
713
+ "loss": 0.3749,
714
+ "step": 1110
715
+ },
716
+ {
717
+ "epoch": 1.19,
718
+ "learning_rate": 0.0002,
719
+ "loss": 0.357,
720
+ "step": 1120
721
+ },
722
+ {
723
+ "epoch": 1.2,
724
+ "learning_rate": 0.0002,
725
+ "loss": 0.3733,
726
+ "step": 1130
727
+ },
728
+ {
729
+ "epoch": 1.21,
730
+ "learning_rate": 0.0002,
731
+ "loss": 0.3956,
732
+ "step": 1140
733
+ },
734
+ {
735
+ "epoch": 1.22,
736
+ "learning_rate": 0.0002,
737
+ "loss": 0.3666,
738
+ "step": 1150
739
+ },
740
+ {
741
+ "epoch": 1.23,
742
+ "learning_rate": 0.0002,
743
+ "loss": 0.3694,
744
+ "step": 1160
745
+ },
746
+ {
747
+ "epoch": 1.24,
748
+ "learning_rate": 0.0002,
749
+ "loss": 0.3737,
750
+ "step": 1170
751
+ },
752
+ {
753
+ "epoch": 1.25,
754
+ "learning_rate": 0.0002,
755
+ "loss": 0.3623,
756
+ "step": 1180
757
+ },
758
+ {
759
+ "epoch": 1.26,
760
+ "learning_rate": 0.0002,
761
+ "loss": 0.3588,
762
+ "step": 1190
763
+ },
764
+ {
765
+ "epoch": 1.27,
766
+ "learning_rate": 0.0002,
767
+ "loss": 0.3494,
768
+ "step": 1200
769
+ },
770
+ {
771
+ "epoch": 1.27,
772
+ "eval_loss": 0.40632152557373047,
773
+ "eval_runtime": 253.8741,
774
+ "eval_samples_per_second": 3.939,
775
+ "eval_steps_per_second": 0.985,
776
+ "step": 1200
777
+ },
778
+ {
779
+ "epoch": 1.28,
780
+ "learning_rate": 0.0002,
781
+ "loss": 0.3791,
782
+ "step": 1210
783
+ },
784
+ {
785
+ "epoch": 1.29,
786
+ "learning_rate": 0.0002,
787
+ "loss": 0.3709,
788
+ "step": 1220
789
+ },
790
+ {
791
+ "epoch": 1.3,
792
+ "learning_rate": 0.0002,
793
+ "loss": 0.3647,
794
+ "step": 1230
795
+ },
796
+ {
797
+ "epoch": 1.31,
798
+ "learning_rate": 0.0002,
799
+ "loss": 0.3702,
800
+ "step": 1240
801
+ },
802
+ {
803
+ "epoch": 1.33,
804
+ "learning_rate": 0.0002,
805
+ "loss": 0.3774,
806
+ "step": 1250
807
+ },
808
+ {
809
+ "epoch": 1.34,
810
+ "learning_rate": 0.0002,
811
+ "loss": 0.3608,
812
+ "step": 1260
813
+ },
814
+ {
815
+ "epoch": 1.35,
816
+ "learning_rate": 0.0002,
817
+ "loss": 0.358,
818
+ "step": 1270
819
+ },
820
+ {
821
+ "epoch": 1.36,
822
+ "learning_rate": 0.0002,
823
+ "loss": 0.3576,
824
+ "step": 1280
825
+ },
826
+ {
827
+ "epoch": 1.37,
828
+ "learning_rate": 0.0002,
829
+ "loss": 0.3781,
830
+ "step": 1290
831
+ },
832
+ {
833
+ "epoch": 1.38,
834
+ "learning_rate": 0.0002,
835
+ "loss": 0.3658,
836
+ "step": 1300
837
+ },
838
+ {
839
+ "epoch": 1.39,
840
+ "learning_rate": 0.0002,
841
+ "loss": 0.3473,
842
+ "step": 1310
843
+ },
844
+ {
845
+ "epoch": 1.4,
846
+ "learning_rate": 0.0002,
847
+ "loss": 0.3546,
848
+ "step": 1320
849
+ },
850
+ {
851
+ "epoch": 1.41,
852
+ "learning_rate": 0.0002,
853
+ "loss": 0.3681,
854
+ "step": 1330
855
+ },
856
+ {
857
+ "epoch": 1.42,
858
+ "learning_rate": 0.0002,
859
+ "loss": 0.3491,
860
+ "step": 1340
861
+ },
862
+ {
863
+ "epoch": 1.43,
864
+ "learning_rate": 0.0002,
865
+ "loss": 0.3454,
866
+ "step": 1350
867
+ },
868
+ {
869
+ "epoch": 1.44,
870
+ "learning_rate": 0.0002,
871
+ "loss": 0.3786,
872
+ "step": 1360
873
+ },
874
+ {
875
+ "epoch": 1.45,
876
+ "learning_rate": 0.0002,
877
+ "loss": 0.3536,
878
+ "step": 1370
879
+ },
880
+ {
881
+ "epoch": 1.46,
882
+ "learning_rate": 0.0002,
883
+ "loss": 0.3607,
884
+ "step": 1380
885
+ },
886
+ {
887
+ "epoch": 1.47,
888
+ "learning_rate": 0.0002,
889
+ "loss": 0.372,
890
+ "step": 1390
891
+ },
892
+ {
893
+ "epoch": 1.48,
894
+ "learning_rate": 0.0002,
895
+ "loss": 0.3621,
896
+ "step": 1400
897
+ },
898
+ {
899
+ "epoch": 1.48,
900
+ "eval_loss": 0.40282392501831055,
901
+ "eval_runtime": 253.8219,
902
+ "eval_samples_per_second": 3.94,
903
+ "eval_steps_per_second": 0.985,
904
+ "step": 1400
905
+ },
906
+ {
907
+ "epoch": 1.5,
908
+ "learning_rate": 0.0002,
909
+ "loss": 0.3644,
910
+ "step": 1410
911
+ },
912
+ {
913
+ "epoch": 1.51,
914
+ "learning_rate": 0.0002,
915
+ "loss": 0.351,
916
+ "step": 1420
917
+ },
918
+ {
919
+ "epoch": 1.52,
920
+ "learning_rate": 0.0002,
921
+ "loss": 0.381,
922
+ "step": 1430
923
+ },
924
+ {
925
+ "epoch": 1.53,
926
+ "learning_rate": 0.0002,
927
+ "loss": 0.3589,
928
+ "step": 1440
929
+ },
930
+ {
931
+ "epoch": 1.54,
932
+ "learning_rate": 0.0002,
933
+ "loss": 0.3528,
934
+ "step": 1450
935
+ },
936
+ {
937
+ "epoch": 1.55,
938
+ "learning_rate": 0.0002,
939
+ "loss": 0.3597,
940
+ "step": 1460
941
+ },
942
+ {
943
+ "epoch": 1.56,
944
+ "learning_rate": 0.0002,
945
+ "loss": 0.3577,
946
+ "step": 1470
947
+ },
948
+ {
949
+ "epoch": 1.57,
950
+ "learning_rate": 0.0002,
951
+ "loss": 0.3595,
952
+ "step": 1480
953
+ },
954
+ {
955
+ "epoch": 1.58,
956
+ "learning_rate": 0.0002,
957
+ "loss": 0.3507,
958
+ "step": 1490
959
+ },
960
+ {
961
+ "epoch": 1.59,
962
+ "learning_rate": 0.0002,
963
+ "loss": 0.3478,
964
+ "step": 1500
965
+ },
966
+ {
967
+ "epoch": 1.6,
968
+ "learning_rate": 0.0002,
969
+ "loss": 0.363,
970
+ "step": 1510
971
+ },
972
+ {
973
+ "epoch": 1.61,
974
+ "learning_rate": 0.0002,
975
+ "loss": 0.3594,
976
+ "step": 1520
977
+ },
978
+ {
979
+ "epoch": 1.62,
980
+ "learning_rate": 0.0002,
981
+ "loss": 0.3531,
982
+ "step": 1530
983
+ },
984
+ {
985
+ "epoch": 1.63,
986
+ "learning_rate": 0.0002,
987
+ "loss": 0.3616,
988
+ "step": 1540
989
+ },
990
+ {
991
+ "epoch": 1.64,
992
+ "learning_rate": 0.0002,
993
+ "loss": 0.3534,
994
+ "step": 1550
995
+ },
996
+ {
997
+ "epoch": 1.65,
998
+ "learning_rate": 0.0002,
999
+ "loss": 0.3657,
1000
+ "step": 1560
1001
+ },
1002
+ {
1003
+ "epoch": 1.66,
1004
+ "learning_rate": 0.0002,
1005
+ "loss": 0.3859,
1006
+ "step": 1570
1007
+ },
1008
+ {
1009
+ "epoch": 1.68,
1010
+ "learning_rate": 0.0002,
1011
+ "loss": 0.3647,
1012
+ "step": 1580
1013
+ },
1014
+ {
1015
+ "epoch": 1.69,
1016
+ "learning_rate": 0.0002,
1017
+ "loss": 0.3659,
1018
+ "step": 1590
1019
+ },
1020
+ {
1021
+ "epoch": 1.7,
1022
+ "learning_rate": 0.0002,
1023
+ "loss": 0.3625,
1024
+ "step": 1600
1025
+ },
1026
+ {
1027
+ "epoch": 1.7,
1028
+ "eval_loss": 0.3978842794895172,
1029
+ "eval_runtime": 253.9522,
1030
+ "eval_samples_per_second": 3.938,
1031
+ "eval_steps_per_second": 0.984,
1032
+ "step": 1600
1033
+ },
1034
+ {
1035
+ "epoch": 1.71,
1036
+ "learning_rate": 0.0002,
1037
+ "loss": 0.3542,
1038
+ "step": 1610
1039
+ },
1040
+ {
1041
+ "epoch": 1.72,
1042
+ "learning_rate": 0.0002,
1043
+ "loss": 0.3683,
1044
+ "step": 1620
1045
+ },
1046
+ {
1047
+ "epoch": 1.73,
1048
+ "learning_rate": 0.0002,
1049
+ "loss": 0.3604,
1050
+ "step": 1630
1051
+ },
1052
+ {
1053
+ "epoch": 1.74,
1054
+ "learning_rate": 0.0002,
1055
+ "loss": 0.3709,
1056
+ "step": 1640
1057
+ },
1058
+ {
1059
+ "epoch": 1.75,
1060
+ "learning_rate": 0.0002,
1061
+ "loss": 0.3685,
1062
+ "step": 1650
1063
+ },
1064
+ {
1065
+ "epoch": 1.76,
1066
+ "learning_rate": 0.0002,
1067
+ "loss": 0.3584,
1068
+ "step": 1660
1069
+ },
1070
+ {
1071
+ "epoch": 1.77,
1072
+ "learning_rate": 0.0002,
1073
+ "loss": 0.3475,
1074
+ "step": 1670
1075
+ },
1076
+ {
1077
+ "epoch": 1.78,
1078
+ "learning_rate": 0.0002,
1079
+ "loss": 0.3759,
1080
+ "step": 1680
1081
+ },
1082
+ {
1083
+ "epoch": 1.79,
1084
+ "learning_rate": 0.0002,
1085
+ "loss": 0.3779,
1086
+ "step": 1690
1087
+ },
1088
+ {
1089
+ "epoch": 1.8,
1090
+ "learning_rate": 0.0002,
1091
+ "loss": 0.3607,
1092
+ "step": 1700
1093
+ },
1094
+ {
1095
+ "epoch": 1.81,
1096
+ "learning_rate": 0.0002,
1097
+ "loss": 0.3639,
1098
+ "step": 1710
1099
+ },
1100
+ {
1101
+ "epoch": 1.82,
1102
+ "learning_rate": 0.0002,
1103
+ "loss": 0.3451,
1104
+ "step": 1720
1105
+ },
1106
+ {
1107
+ "epoch": 1.83,
1108
+ "learning_rate": 0.0002,
1109
+ "loss": 0.3738,
1110
+ "step": 1730
1111
+ },
1112
+ {
1113
+ "epoch": 1.85,
1114
+ "learning_rate": 0.0002,
1115
+ "loss": 0.3384,
1116
+ "step": 1740
1117
+ },
1118
+ {
1119
+ "epoch": 1.86,
1120
+ "learning_rate": 0.0002,
1121
+ "loss": 0.3548,
1122
+ "step": 1750
1123
+ },
1124
+ {
1125
+ "epoch": 1.87,
1126
+ "learning_rate": 0.0002,
1127
+ "loss": 0.3633,
1128
+ "step": 1760
1129
+ },
1130
+ {
1131
+ "epoch": 1.88,
1132
+ "learning_rate": 0.0002,
1133
+ "loss": 0.3658,
1134
+ "step": 1770
1135
+ },
1136
+ {
1137
+ "epoch": 1.89,
1138
+ "learning_rate": 0.0002,
1139
+ "loss": 0.3422,
1140
+ "step": 1780
1141
+ },
1142
+ {
1143
+ "epoch": 1.9,
1144
+ "learning_rate": 0.0002,
1145
+ "loss": 0.3682,
1146
+ "step": 1790
1147
+ },
1148
+ {
1149
+ "epoch": 1.91,
1150
+ "learning_rate": 0.0002,
1151
+ "loss": 0.3397,
1152
+ "step": 1800
1153
+ },
1154
+ {
1155
+ "epoch": 1.91,
1156
+ "eval_loss": 0.39367392659187317,
1157
+ "eval_runtime": 254.0031,
1158
+ "eval_samples_per_second": 3.937,
1159
+ "eval_steps_per_second": 0.984,
1160
+ "step": 1800
1161
+ },
1162
+ {
1163
+ "epoch": 1.92,
1164
+ "learning_rate": 0.0002,
1165
+ "loss": 0.3424,
1166
+ "step": 1810
1167
+ },
1168
+ {
1169
+ "epoch": 1.93,
1170
+ "learning_rate": 0.0002,
1171
+ "loss": 0.3644,
1172
+ "step": 1820
1173
+ },
1174
+ {
1175
+ "epoch": 1.94,
1176
+ "learning_rate": 0.0002,
1177
+ "loss": 0.3605,
1178
+ "step": 1830
1179
+ },
1180
+ {
1181
+ "epoch": 1.95,
1182
+ "learning_rate": 0.0002,
1183
+ "loss": 0.3532,
1184
+ "step": 1840
1185
+ },
1186
+ {
1187
+ "epoch": 1.96,
1188
+ "learning_rate": 0.0002,
1189
+ "loss": 0.3598,
1190
+ "step": 1850
1191
+ },
1192
+ {
1193
+ "epoch": 1.97,
1194
+ "learning_rate": 0.0002,
1195
+ "loss": 0.3512,
1196
+ "step": 1860
1197
+ },
1198
+ {
1199
+ "epoch": 1.98,
1200
+ "learning_rate": 0.0002,
1201
+ "loss": 0.3539,
1202
+ "step": 1870
1203
+ },
1204
+ {
1205
+ "epoch": 1.99,
1206
+ "learning_rate": 0.0002,
1207
+ "loss": 0.3527,
1208
+ "step": 1880
1209
+ },
1210
+ {
1211
+ "epoch": 2.0,
1212
+ "learning_rate": 0.0002,
1213
+ "loss": 0.3396,
1214
+ "step": 1890
1215
+ },
1216
+ {
1217
+ "epoch": 2.01,
1218
+ "learning_rate": 0.0002,
1219
+ "loss": 0.2891,
1220
+ "step": 1900
1221
+ },
1222
+ {
1223
+ "epoch": 2.03,
1224
+ "learning_rate": 0.0002,
1225
+ "loss": 0.2841,
1226
+ "step": 1910
1227
+ },
1228
+ {
1229
+ "epoch": 2.04,
1230
+ "learning_rate": 0.0002,
1231
+ "loss": 0.2833,
1232
+ "step": 1920
1233
+ },
1234
+ {
1235
+ "epoch": 2.05,
1236
+ "learning_rate": 0.0002,
1237
+ "loss": 0.2818,
1238
+ "step": 1930
1239
+ },
1240
+ {
1241
+ "epoch": 2.06,
1242
+ "learning_rate": 0.0002,
1243
+ "loss": 0.3051,
1244
+ "step": 1940
1245
+ },
1246
+ {
1247
+ "epoch": 2.07,
1248
+ "learning_rate": 0.0002,
1249
+ "loss": 0.2943,
1250
+ "step": 1950
1251
+ },
1252
+ {
1253
+ "epoch": 2.08,
1254
+ "learning_rate": 0.0002,
1255
+ "loss": 0.2801,
1256
+ "step": 1960
1257
+ },
1258
+ {
1259
+ "epoch": 2.09,
1260
+ "learning_rate": 0.0002,
1261
+ "loss": 0.2891,
1262
+ "step": 1970
1263
+ },
1264
+ {
1265
+ "epoch": 2.1,
1266
+ "learning_rate": 0.0002,
1267
+ "loss": 0.2887,
1268
+ "step": 1980
1269
+ },
1270
+ {
1271
+ "epoch": 2.11,
1272
+ "learning_rate": 0.0002,
1273
+ "loss": 0.292,
1274
+ "step": 1990
1275
+ },
1276
+ {
1277
+ "epoch": 2.12,
1278
+ "learning_rate": 0.0002,
1279
+ "loss": 0.2921,
1280
+ "step": 2000
1281
+ },
1282
+ {
1283
+ "epoch": 2.12,
1284
+ "eval_loss": 0.4052470624446869,
1285
+ "eval_runtime": 253.9413,
1286
+ "eval_samples_per_second": 3.938,
1287
+ "eval_steps_per_second": 0.984,
1288
+ "step": 2000
1289
+ },
1290
+ {
1291
+ "epoch": 2.13,
1292
+ "learning_rate": 0.0002,
1293
+ "loss": 0.2793,
1294
+ "step": 2010
1295
+ },
1296
+ {
1297
+ "epoch": 2.14,
1298
+ "learning_rate": 0.0002,
1299
+ "loss": 0.297,
1300
+ "step": 2020
1301
+ },
1302
+ {
1303
+ "epoch": 2.15,
1304
+ "learning_rate": 0.0002,
1305
+ "loss": 0.2978,
1306
+ "step": 2030
1307
+ },
1308
+ {
1309
+ "epoch": 2.16,
1310
+ "learning_rate": 0.0002,
1311
+ "loss": 0.2838,
1312
+ "step": 2040
1313
+ },
1314
+ {
1315
+ "epoch": 2.17,
1316
+ "learning_rate": 0.0002,
1317
+ "loss": 0.2981,
1318
+ "step": 2050
1319
+ },
1320
+ {
1321
+ "epoch": 2.18,
1322
+ "learning_rate": 0.0002,
1323
+ "loss": 0.2959,
1324
+ "step": 2060
1325
+ },
1326
+ {
1327
+ "epoch": 2.2,
1328
+ "learning_rate": 0.0002,
1329
+ "loss": 0.3081,
1330
+ "step": 2070
1331
+ },
1332
+ {
1333
+ "epoch": 2.21,
1334
+ "learning_rate": 0.0002,
1335
+ "loss": 0.3071,
1336
+ "step": 2080
1337
+ },
1338
+ {
1339
+ "epoch": 2.22,
1340
+ "learning_rate": 0.0002,
1341
+ "loss": 0.2848,
1342
+ "step": 2090
1343
+ },
1344
+ {
1345
+ "epoch": 2.23,
1346
+ "learning_rate": 0.0002,
1347
+ "loss": 0.2904,
1348
+ "step": 2100
1349
+ },
1350
+ {
1351
+ "epoch": 2.24,
1352
+ "learning_rate": 0.0002,
1353
+ "loss": 0.3007,
1354
+ "step": 2110
1355
+ },
1356
+ {
1357
+ "epoch": 2.25,
1358
+ "learning_rate": 0.0002,
1359
+ "loss": 0.2922,
1360
+ "step": 2120
1361
+ },
1362
+ {
1363
+ "epoch": 2.26,
1364
+ "learning_rate": 0.0002,
1365
+ "loss": 0.3092,
1366
+ "step": 2130
1367
+ },
1368
+ {
1369
+ "epoch": 2.27,
1370
+ "learning_rate": 0.0002,
1371
+ "loss": 0.2847,
1372
+ "step": 2140
1373
+ },
1374
+ {
1375
+ "epoch": 2.28,
1376
+ "learning_rate": 0.0002,
1377
+ "loss": 0.3031,
1378
+ "step": 2150
1379
+ },
1380
+ {
1381
+ "epoch": 2.29,
1382
+ "learning_rate": 0.0002,
1383
+ "loss": 0.2986,
1384
+ "step": 2160
1385
+ },
1386
+ {
1387
+ "epoch": 2.3,
1388
+ "learning_rate": 0.0002,
1389
+ "loss": 0.2858,
1390
+ "step": 2170
1391
+ },
1392
+ {
1393
+ "epoch": 2.31,
1394
+ "learning_rate": 0.0002,
1395
+ "loss": 0.3023,
1396
+ "step": 2180
1397
+ },
1398
+ {
1399
+ "epoch": 2.32,
1400
+ "learning_rate": 0.0002,
1401
+ "loss": 0.3097,
1402
+ "step": 2190
1403
+ },
1404
+ {
1405
+ "epoch": 2.33,
1406
+ "learning_rate": 0.0002,
1407
+ "loss": 0.2884,
1408
+ "step": 2200
1409
+ },
1410
+ {
1411
+ "epoch": 2.33,
1412
+ "eval_loss": 0.40587317943573,
1413
+ "eval_runtime": 254.016,
1414
+ "eval_samples_per_second": 3.937,
1415
+ "eval_steps_per_second": 0.984,
1416
+ "step": 2200
1417
+ },
1418
+ {
1419
+ "epoch": 2.34,
1420
+ "learning_rate": 0.0002,
1421
+ "loss": 0.286,
1422
+ "step": 2210
1423
+ },
1424
+ {
1425
+ "epoch": 2.35,
1426
+ "learning_rate": 0.0002,
1427
+ "loss": 0.2942,
1428
+ "step": 2220
1429
+ },
1430
+ {
1431
+ "epoch": 2.36,
1432
+ "learning_rate": 0.0002,
1433
+ "loss": 0.2947,
1434
+ "step": 2230
1435
+ },
1436
+ {
1437
+ "epoch": 2.38,
1438
+ "learning_rate": 0.0002,
1439
+ "loss": 0.3024,
1440
+ "step": 2240
1441
+ },
1442
+ {
1443
+ "epoch": 2.39,
1444
+ "learning_rate": 0.0002,
1445
+ "loss": 0.2958,
1446
+ "step": 2250
1447
+ },
1448
+ {
1449
+ "epoch": 2.4,
1450
+ "learning_rate": 0.0002,
1451
+ "loss": 0.3074,
1452
+ "step": 2260
1453
+ },
1454
+ {
1455
+ "epoch": 2.41,
1456
+ "learning_rate": 0.0002,
1457
+ "loss": 0.2943,
1458
+ "step": 2270
1459
+ },
1460
+ {
1461
+ "epoch": 2.42,
1462
+ "learning_rate": 0.0002,
1463
+ "loss": 0.2938,
1464
+ "step": 2280
1465
+ },
1466
+ {
1467
+ "epoch": 2.43,
1468
+ "learning_rate": 0.0002,
1469
+ "loss": 0.2842,
1470
+ "step": 2290
1471
+ },
1472
+ {
1473
+ "epoch": 2.44,
1474
+ "learning_rate": 0.0002,
1475
+ "loss": 0.3056,
1476
+ "step": 2300
1477
+ },
1478
+ {
1479
+ "epoch": 2.45,
1480
+ "learning_rate": 0.0002,
1481
+ "loss": 0.3002,
1482
+ "step": 2310
1483
+ },
1484
+ {
1485
+ "epoch": 2.46,
1486
+ "learning_rate": 0.0002,
1487
+ "loss": 0.3111,
1488
+ "step": 2320
1489
+ },
1490
+ {
1491
+ "epoch": 2.47,
1492
+ "learning_rate": 0.0002,
1493
+ "loss": 0.2953,
1494
+ "step": 2330
1495
+ },
1496
+ {
1497
+ "epoch": 2.48,
1498
+ "learning_rate": 0.0002,
1499
+ "loss": 0.3031,
1500
+ "step": 2340
1501
+ },
1502
+ {
1503
+ "epoch": 2.49,
1504
+ "learning_rate": 0.0002,
1505
+ "loss": 0.2929,
1506
+ "step": 2350
1507
+ },
1508
+ {
1509
+ "epoch": 2.5,
1510
+ "learning_rate": 0.0002,
1511
+ "loss": 0.2933,
1512
+ "step": 2360
1513
+ },
1514
+ {
1515
+ "epoch": 2.51,
1516
+ "learning_rate": 0.0002,
1517
+ "loss": 0.3226,
1518
+ "step": 2370
1519
+ },
1520
+ {
1521
+ "epoch": 2.52,
1522
+ "learning_rate": 0.0002,
1523
+ "loss": 0.2806,
1524
+ "step": 2380
1525
+ },
1526
+ {
1527
+ "epoch": 2.53,
1528
+ "learning_rate": 0.0002,
1529
+ "loss": 0.3094,
1530
+ "step": 2390
1531
+ },
1532
+ {
1533
+ "epoch": 2.55,
1534
+ "learning_rate": 0.0002,
1535
+ "loss": 0.2736,
1536
+ "step": 2400
1537
+ },
1538
+ {
1539
+ "epoch": 2.55,
1540
+ "eval_loss": 0.4049573540687561,
1541
+ "eval_runtime": 254.127,
1542
+ "eval_samples_per_second": 3.935,
1543
+ "eval_steps_per_second": 0.984,
1544
+ "step": 2400
1545
+ }
1546
+ ],
1547
+ "max_steps": 5000,
1548
+ "num_train_epochs": 6,
1549
+ "total_flos": 6.581138123797955e+17,
1550
+ "trial_name": null,
1551
+ "trial_params": null
1552
+ }
checkpoint-2400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d43bb46b7e18307cd325d889344b3c5a6b0f12a2594434f5acb40117cbab761
3
+ size 6011