Hugging Face's logo

justinwangx
/
vicuna-adv-robust-u50-sft-lora

TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
justinwangx commited on
Commit
717b5b3
·
1 Parent(s): 72fbda8

Model save

Browse files
Files changed (15) hide show
  1. README.md +109 -0
  2. adapter_config.json +25 -0
  3. adapter_model.safetensors +3 -0
  4. all_results.json +13 -0
  5. eval_results.json +8 -0
  6. runs/Jan01_20-57-04_f63ae9056bcd/events.out.tfevents.1704142650.f63ae9056bcd.945948.0 +3 -0
  7. runs/Jan01_20-58-10_f63ae9056bcd/events.out.tfevents.1704142713.f63ae9056bcd.947227.0 +3 -0
  8. runs/Jan01_20-58-10_f63ae9056bcd/events.out.tfevents.1704146527.f63ae9056bcd.947227.1 +3 -0
  9. special_tokens_map.json +30 -0
  10. tokenizer.json +0 -0
  11. tokenizer.model +3 -0
  12. tokenizer_config.json +43 -0
  13. train_results.json +8 -0
  14. trainer_state.json +470 -0
  15. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,109 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - generated_from_trainer
4
+ model-index:
5
+ - name: vicuna-adv-robust-u50-sft-lora
6
+ results: []
7
+ ---
8
+
9
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
10
+ should probably proofread and complete it, then remove this comment. -->
11
+
12
+ # vicuna-adv-robust-u50-sft-lora
13
+
14
+ This model was trained from scratch on an unknown dataset.
15
+ It achieves the following results on the evaluation set:
16
+ - Loss: 1.2125
17
+
18
+ ## Model description
19
+
20
+ More information needed
21
+
22
+ ## Intended uses & limitations
23
+
24
+ More information needed
25
+
26
+ ## Training and evaluation data
27
+
28
+ More information needed
29
+
30
+ ## Training procedure
31
+
32
+ ### Training hyperparameters
33
+
34
+ The following hyperparameters were used during training:
35
+ - learning_rate: 0.0003
36
+ - train_batch_size: 8
37
+ - eval_batch_size: 8
38
+ - seed: 42
39
+ - distributed_type: multi-GPU
40
+ - num_devices: 4
41
+ - gradient_accumulation_steps: 16
42
+ - total_train_batch_size: 512
43
+ - total_eval_batch_size: 32
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: cosine
46
+ - num_epochs: 50
47
+
48
+ ### Training results
49
+
50
+ | Training Loss | Epoch | Step | Validation Loss |
51
+ |:-------------:|:-----:|:----:|:---------------:|
52
+ | No log | 0 | 0 | 2.4952 |
53
+ | 2.5615 | 1.09 | 1 | 2.5270 |
54
+ | 2.5615 | 1.09 | 1 | 2.5362 |
55
+ | 2.5615 | 3.03 | 2 | 2.5342 |
56
+ | 2.5615 | 4.12 | 3 | 2.2735 |
57
+ | 2.5615 | 4.12 | 3 | 2.3209 |
58
+ | 2.5615 | 6.06 | 4 | 2.1017 |
59
+ | 2.363 | 7.15 | 5 | 2.0121 |
60
+ | 2.363 | 7.15 | 5 | 2.0751 |
61
+ | 2.363 | 9.09 | 6 | 1.9646 |
62
+ | 2.363 | 9.09 | 6 | 1.8912 |
63
+ | 2.363 | 11.03 | 7 | 1.8100 |
64
+ | 2.363 | 12.12 | 8 | 1.8144 |
65
+ | 2.363 | 12.12 | 8 | 1.7983 |
66
+ | 2.363 | 14.06 | 9 | 1.7634 |
67
+ | 1.9009 | 15.15 | 10 | 1.7628 |
68
+ | 1.9009 | 15.15 | 10 | 1.7354 |
69
+ | 1.9009 | 17.09 | 11 | 1.7343 |
70
+ | 1.9009 | 17.09 | 11 | 1.7232 |
71
+ | 1.9009 | 19.03 | 12 | 1.6737 |
72
+ | 1.9009 | 20.12 | 13 | 1.6418 |
73
+ | 1.9009 | 20.12 | 13 | 1.6635 |
74
+ | 1.9009 | 22.06 | 14 | 1.6280 |
75
+ | 1.7031 | 23.15 | 15 | 1.6042 |
76
+ | 1.7031 | 23.15 | 15 | 1.6120 |
77
+ | 1.7031 | 25.09 | 16 | 1.5792 |
78
+ | 1.7031 | 25.09 | 16 | 1.6128 |
79
+ | 1.7031 | 27.03 | 17 | 1.5468 |
80
+ | 1.7031 | 28.12 | 18 | 1.5303 |
81
+ | 1.7031 | 28.12 | 18 | 1.5160 |
82
+ | 1.7031 | 30.06 | 19 | 1.5195 |
83
+ | 1.5968 | 31.15 | 20 | 1.5098 |
84
+ | 1.5968 | 31.15 | 20 | 1.4775 |
85
+ | 1.5968 | 33.09 | 21 | 1.4770 |
86
+ | 1.5968 | 33.09 | 21 | 1.4588 |
87
+ | 1.5968 | 35.03 | 22 | 1.4474 |
88
+ | 1.5968 | 36.12 | 23 | 1.4240 |
89
+ | 1.5968 | 36.12 | 23 | 1.4164 |
90
+ | 1.5968 | 38.06 | 24 | 1.4060 |
91
+ | 1.4776 | 39.15 | 25 | 1.3753 |
92
+ | 1.4776 | 39.15 | 25 | 1.3858 |
93
+ | 1.4776 | 41.09 | 26 | 1.3822 |
94
+ | 1.4776 | 41.09 | 26 | 1.3268 |
95
+ | 1.4776 | 43.03 | 27 | 1.3443 |
96
+ | 1.4776 | 44.12 | 28 | 1.3259 |
97
+ | 1.4776 | 44.12 | 28 | 1.3117 |
98
+ | 1.4776 | 46.06 | 29 | 1.3105 |
99
+ | 1.3585 | 47.15 | 30 | 1.2553 |
100
+ | 1.3585 | 47.15 | 30 | 1.2755 |
101
+ | 1.3585 | 49.09 | 31 | 1.2036 |
102
+
103
+
104
+ ### Framework versions
105
+
106
+ - Transformers 4.35.0
107
+ - Pytorch 2.1.0a0+32f93b1
108
+ - Datasets 2.14.6
109
+ - Tokenizers 0.14.1
adapter_config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/data/models/vicuna-7b-v1.3",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 16,
12
+ "lora_dropout": 0.1,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 64,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "q_proj",
20
+ "o_proj",
21
+ "v_proj",
22
+ "k_proj"
23
+ ],
24
+ "task_type": "CAUSAL_LM"
25
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf6e37dabf54b5340f25cf81b486d504f46e26793e27545eb0e5ab0fea10a8fd
3
+ size 134252592
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 49.09,
3
+ "eval_loss": 1.2125475406646729,
4
+ "eval_runtime": 2.6467,
5
+ "eval_samples": 234,
6
+ "eval_samples_per_second": 88.413,
7
+ "eval_steps_per_second": 3.023,
8
+ "train_loss": 1.7344687215743526,
9
+ "train_runtime": 3811.391,
10
+ "train_samples": 2097,
11
+ "train_samples_per_second": 27.51,
12
+ "train_steps_per_second": 0.052
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 49.09,
3
+ "eval_loss": 1.2125475406646729,
4
+ "eval_runtime": 2.6467,
5
+ "eval_samples": 234,
6
+ "eval_samples_per_second": 88.413,
7
+ "eval_steps_per_second": 3.023
8
+ }
runs/Jan01_20-57-04_f63ae9056bcd/events.out.tfevents.1704142650.f63ae9056bcd.945948.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6aae9189ac5944f1e0e1446936883e5ed04896dd68258d936796f415f6b2b559
3
+ size 4327
runs/Jan01_20-58-10_f63ae9056bcd/events.out.tfevents.1704142713.f63ae9056bcd.947227.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:671acf56c0cc21563050b3f8c02017e477041aff01f8ba28f9bab30bd87d750b
3
+ size 19043
runs/Jan01_20-58-10_f63ae9056bcd/events.out.tfevents.1704146527.f63ae9056bcd.947227.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff07bf6c4478ae0e940a31f5203a43c267b2ad3adedef0d11128fd370a19ca43
3
+ size 306
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<unk>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": true,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "bos_token": "<s>",
31
+ "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": true,
35
+ "model_max_length": 2048,
36
+ "pad_token": "<unk>",
37
+ "padding_side": "right",
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 49.09,
3
+ "train_loss": 1.7344687215743526,
4
+ "train_runtime": 3811.391,
5
+ "train_samples": 2097,
6
+ "train_samples_per_second": 27.51,
7
+ "train_steps_per_second": 0.052
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,470 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 49.09090909090909,
5
+ "eval_steps": 500,
6
+ "global_step": 31,
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,
13
+ "eval_loss": 2.495208501815796,
14
+ "eval_runtime": 2.7046,
15
+ "eval_samples_per_second": 86.518,
16
+ "eval_steps_per_second": 2.958,
17
+ "step": 0
18
+ },
19
+ {
20
+ "epoch": 1.09,
21
+ "learning_rate": 0.0002999814948722491,
22
+ "loss": 2.5615,
23
+ "step": 1
24
+ },
25
+ {
26
+ "epoch": 1.09,
27
+ "eval_loss": 2.527021646499634,
28
+ "eval_runtime": 2.6357,
29
+ "eval_samples_per_second": 88.78,
30
+ "eval_steps_per_second": 3.035,
31
+ "step": 1
32
+ },
33
+ {
34
+ "epoch": 1.09,
35
+ "eval_loss": 2.5362284183502197,
36
+ "eval_runtime": 2.6589,
37
+ "eval_samples_per_second": 88.005,
38
+ "eval_steps_per_second": 3.009,
39
+ "step": 1
40
+ },
41
+ {
42
+ "epoch": 3.03,
43
+ "eval_loss": 2.5341787338256836,
44
+ "eval_runtime": 2.6589,
45
+ "eval_samples_per_second": 88.007,
46
+ "eval_steps_per_second": 3.009,
47
+ "step": 2
48
+ },
49
+ {
50
+ "epoch": 4.12,
51
+ "eval_loss": 2.2734625339508057,
52
+ "eval_runtime": 2.6648,
53
+ "eval_samples_per_second": 87.812,
54
+ "eval_steps_per_second": 3.002,
55
+ "step": 3
56
+ },
57
+ {
58
+ "epoch": 4.12,
59
+ "eval_loss": 2.3209266662597656,
60
+ "eval_runtime": 2.6531,
61
+ "eval_samples_per_second": 88.198,
62
+ "eval_steps_per_second": 3.015,
63
+ "step": 3
64
+ },
65
+ {
66
+ "epoch": 6.06,
67
+ "eval_loss": 2.1017019748687744,
68
+ "eval_runtime": 2.6605,
69
+ "eval_samples_per_second": 87.954,
70
+ "eval_steps_per_second": 3.007,
71
+ "step": 4
72
+ },
73
+ {
74
+ "epoch": 7.15,
75
+ "learning_rate": 0.00029953760005996916,
76
+ "loss": 2.363,
77
+ "step": 5
78
+ },
79
+ {
80
+ "epoch": 7.15,
81
+ "eval_loss": 2.012136697769165,
82
+ "eval_runtime": 2.6573,
83
+ "eval_samples_per_second": 88.061,
84
+ "eval_steps_per_second": 3.011,
85
+ "step": 5
86
+ },
87
+ {
88
+ "epoch": 7.15,
89
+ "eval_loss": 2.0751442909240723,
90
+ "eval_runtime": 2.6638,
91
+ "eval_samples_per_second": 87.844,
92
+ "eval_steps_per_second": 3.003,
93
+ "step": 5
94
+ },
95
+ {
96
+ "epoch": 9.09,
97
+ "eval_loss": 1.964595079421997,
98
+ "eval_runtime": 2.7098,
99
+ "eval_samples_per_second": 86.353,
100
+ "eval_steps_per_second": 2.952,
101
+ "step": 6
102
+ },
103
+ {
104
+ "epoch": 9.09,
105
+ "eval_loss": 1.8911688327789307,
106
+ "eval_runtime": 2.6643,
107
+ "eval_samples_per_second": 87.829,
108
+ "eval_steps_per_second": 3.003,
109
+ "step": 6
110
+ },
111
+ {
112
+ "epoch": 11.03,
113
+ "eval_loss": 1.809972882270813,
114
+ "eval_runtime": 2.6547,
115
+ "eval_samples_per_second": 88.146,
116
+ "eval_steps_per_second": 3.014,
117
+ "step": 7
118
+ },
119
+ {
120
+ "epoch": 12.12,
121
+ "eval_loss": 1.8143646717071533,
122
+ "eval_runtime": 2.6683,
123
+ "eval_samples_per_second": 87.697,
124
+ "eval_steps_per_second": 2.998,
125
+ "step": 8
126
+ },
127
+ {
128
+ "epoch": 12.12,
129
+ "eval_loss": 1.7983335256576538,
130
+ "eval_runtime": 2.6503,
131
+ "eval_samples_per_second": 88.291,
132
+ "eval_steps_per_second": 3.018,
133
+ "step": 8
134
+ },
135
+ {
136
+ "epoch": 14.06,
137
+ "eval_loss": 1.7633870840072632,
138
+ "eval_runtime": 2.6612,
139
+ "eval_samples_per_second": 87.931,
140
+ "eval_steps_per_second": 3.006,
141
+ "step": 9
142
+ },
143
+ {
144
+ "epoch": 15.15,
145
+ "learning_rate": 0.00029815325108927063,
146
+ "loss": 1.9009,
147
+ "step": 10
148
+ },
149
+ {
150
+ "epoch": 15.15,
151
+ "eval_loss": 1.762792706489563,
152
+ "eval_runtime": 2.6498,
153
+ "eval_samples_per_second": 88.31,
154
+ "eval_steps_per_second": 3.019,
155
+ "step": 10
156
+ },
157
+ {
158
+ "epoch": 15.15,
159
+ "eval_loss": 1.7354298830032349,
160
+ "eval_runtime": 2.6595,
161
+ "eval_samples_per_second": 87.986,
162
+ "eval_steps_per_second": 3.008,
163
+ "step": 10
164
+ },
165
+ {
166
+ "epoch": 17.09,
167
+ "eval_loss": 1.7343316078186035,
168
+ "eval_runtime": 2.6543,
169
+ "eval_samples_per_second": 88.159,
170
+ "eval_steps_per_second": 3.014,
171
+ "step": 11
172
+ },
173
+ {
174
+ "epoch": 17.09,
175
+ "eval_loss": 1.7231522798538208,
176
+ "eval_runtime": 2.6679,
177
+ "eval_samples_per_second": 87.709,
178
+ "eval_steps_per_second": 2.999,
179
+ "step": 11
180
+ },
181
+ {
182
+ "epoch": 19.03,
183
+ "eval_loss": 1.6737045049667358,
184
+ "eval_runtime": 2.6731,
185
+ "eval_samples_per_second": 87.538,
186
+ "eval_steps_per_second": 2.993,
187
+ "step": 12
188
+ },
189
+ {
190
+ "epoch": 20.12,
191
+ "eval_loss": 1.6417571306228638,
192
+ "eval_runtime": 2.6611,
193
+ "eval_samples_per_second": 87.935,
194
+ "eval_steps_per_second": 3.006,
195
+ "step": 13
196
+ },
197
+ {
198
+ "epoch": 20.12,
199
+ "eval_loss": 1.663546085357666,
200
+ "eval_runtime": 2.7084,
201
+ "eval_samples_per_second": 86.399,
202
+ "eval_steps_per_second": 2.954,
203
+ "step": 13
204
+ },
205
+ {
206
+ "epoch": 22.06,
207
+ "eval_loss": 1.6280120611190796,
208
+ "eval_runtime": 2.6541,
209
+ "eval_samples_per_second": 88.166,
210
+ "eval_steps_per_second": 3.014,
211
+ "step": 14
212
+ },
213
+ {
214
+ "epoch": 23.15,
215
+ "learning_rate": 0.0002958554880596515,
216
+ "loss": 1.7031,
217
+ "step": 15
218
+ },
219
+ {
220
+ "epoch": 23.15,
221
+ "eval_loss": 1.6042001247406006,
222
+ "eval_runtime": 2.6431,
223
+ "eval_samples_per_second": 88.533,
224
+ "eval_steps_per_second": 3.027,
225
+ "step": 15
226
+ },
227
+ {
228
+ "epoch": 23.15,
229
+ "eval_loss": 1.6120343208312988,
230
+ "eval_runtime": 2.6568,
231
+ "eval_samples_per_second": 88.076,
232
+ "eval_steps_per_second": 3.011,
233
+ "step": 15
234
+ },
235
+ {
236
+ "epoch": 25.09,
237
+ "eval_loss": 1.579213261604309,
238
+ "eval_runtime": 2.6609,
239
+ "eval_samples_per_second": 87.94,
240
+ "eval_steps_per_second": 3.007,
241
+ "step": 16
242
+ },
243
+ {
244
+ "epoch": 25.09,
245
+ "eval_loss": 1.6127510070800781,
246
+ "eval_runtime": 2.6566,
247
+ "eval_samples_per_second": 88.082,
248
+ "eval_steps_per_second": 3.011,
249
+ "step": 16
250
+ },
251
+ {
252
+ "epoch": 27.03,
253
+ "eval_loss": 1.5467751026153564,
254
+ "eval_runtime": 2.655,
255
+ "eval_samples_per_second": 88.136,
256
+ "eval_steps_per_second": 3.013,
257
+ "step": 17
258
+ },
259
+ {
260
+ "epoch": 28.12,
261
+ "eval_loss": 1.530348539352417,
262
+ "eval_runtime": 2.6531,
263
+ "eval_samples_per_second": 88.197,
264
+ "eval_steps_per_second": 3.015,
265
+ "step": 18
266
+ },
267
+ {
268
+ "epoch": 28.12,
269
+ "eval_loss": 1.5159918069839478,
270
+ "eval_runtime": 2.6518,
271
+ "eval_samples_per_second": 88.241,
272
+ "eval_steps_per_second": 3.017,
273
+ "step": 18
274
+ },
275
+ {
276
+ "epoch": 30.06,
277
+ "eval_loss": 1.5194865465164185,
278
+ "eval_runtime": 2.6595,
279
+ "eval_samples_per_second": 87.987,
280
+ "eval_steps_per_second": 3.008,
281
+ "step": 19
282
+ },
283
+ {
284
+ "epoch": 31.15,
285
+ "learning_rate": 0.00029265847744427303,
286
+ "loss": 1.5968,
287
+ "step": 20
288
+ },
289
+ {
290
+ "epoch": 31.15,
291
+ "eval_loss": 1.5098381042480469,
292
+ "eval_runtime": 2.6396,
293
+ "eval_samples_per_second": 88.651,
294
+ "eval_steps_per_second": 3.031,
295
+ "step": 20
296
+ },
297
+ {
298
+ "epoch": 31.15,
299
+ "eval_loss": 1.4774686098098755,
300
+ "eval_runtime": 2.6606,
301
+ "eval_samples_per_second": 87.951,
302
+ "eval_steps_per_second": 3.007,
303
+ "step": 20
304
+ },
305
+ {
306
+ "epoch": 33.09,
307
+ "eval_loss": 1.4770317077636719,
308
+ "eval_runtime": 2.6523,
309
+ "eval_samples_per_second": 88.225,
310
+ "eval_steps_per_second": 3.016,
311
+ "step": 21
312
+ },
313
+ {
314
+ "epoch": 33.09,
315
+ "eval_loss": 1.4588351249694824,
316
+ "eval_runtime": 2.6529,
317
+ "eval_samples_per_second": 88.205,
318
+ "eval_steps_per_second": 3.016,
319
+ "step": 21
320
+ },
321
+ {
322
+ "epoch": 35.03,
323
+ "eval_loss": 1.4474384784698486,
324
+ "eval_runtime": 2.6678,
325
+ "eval_samples_per_second": 87.711,
326
+ "eval_steps_per_second": 2.999,
327
+ "step": 22
328
+ },
329
+ {
330
+ "epoch": 36.12,
331
+ "eval_loss": 1.424033761024475,
332
+ "eval_runtime": 2.6514,
333
+ "eval_samples_per_second": 88.254,
334
+ "eval_steps_per_second": 3.017,
335
+ "step": 23
336
+ },
337
+ {
338
+ "epoch": 36.12,
339
+ "eval_loss": 1.4164339303970337,
340
+ "eval_runtime": 2.6554,
341
+ "eval_samples_per_second": 88.121,
342
+ "eval_steps_per_second": 3.013,
343
+ "step": 23
344
+ },
345
+ {
346
+ "epoch": 38.06,
347
+ "eval_loss": 1.4059854745864868,
348
+ "eval_runtime": 2.6536,
349
+ "eval_samples_per_second": 88.181,
350
+ "eval_steps_per_second": 3.015,
351
+ "step": 24
352
+ },
353
+ {
354
+ "epoch": 39.15,
355
+ "learning_rate": 0.000288581929876693,
356
+ "loss": 1.4776,
357
+ "step": 25
358
+ },
359
+ {
360
+ "epoch": 39.15,
361
+ "eval_loss": 1.3752561807632446,
362
+ "eval_runtime": 2.6459,
363
+ "eval_samples_per_second": 88.439,
364
+ "eval_steps_per_second": 3.024,
365
+ "step": 25
366
+ },
367
+ {
368
+ "epoch": 39.15,
369
+ "eval_loss": 1.385780930519104,
370
+ "eval_runtime": 2.667,
371
+ "eval_samples_per_second": 87.738,
372
+ "eval_steps_per_second": 3.0,
373
+ "step": 25
374
+ },
375
+ {
376
+ "epoch": 41.09,
377
+ "eval_loss": 1.3821604251861572,
378
+ "eval_runtime": 2.6548,
379
+ "eval_samples_per_second": 88.141,
380
+ "eval_steps_per_second": 3.013,
381
+ "step": 26
382
+ },
383
+ {
384
+ "epoch": 41.09,
385
+ "eval_loss": 1.3268494606018066,
386
+ "eval_runtime": 2.6901,
387
+ "eval_samples_per_second": 86.986,
388
+ "eval_steps_per_second": 2.974,
389
+ "step": 26
390
+ },
391
+ {
392
+ "epoch": 43.03,
393
+ "eval_loss": 1.3443068265914917,
394
+ "eval_runtime": 2.6512,
395
+ "eval_samples_per_second": 88.263,
396
+ "eval_steps_per_second": 3.018,
397
+ "step": 27
398
+ },
399
+ {
400
+ "epoch": 44.12,
401
+ "eval_loss": 1.3258930444717407,
402
+ "eval_runtime": 2.6544,
403
+ "eval_samples_per_second": 88.156,
404
+ "eval_steps_per_second": 3.014,
405
+ "step": 28
406
+ },
407
+ {
408
+ "epoch": 44.12,
409
+ "eval_loss": 1.311697006225586,
410
+ "eval_runtime": 2.6815,
411
+ "eval_samples_per_second": 87.264,
412
+ "eval_steps_per_second": 2.983,
413
+ "step": 28
414
+ },
415
+ {
416
+ "epoch": 46.06,
417
+ "eval_loss": 1.3104833364486694,
418
+ "eval_runtime": 2.6829,
419
+ "eval_samples_per_second": 87.218,
420
+ "eval_steps_per_second": 2.982,
421
+ "step": 29
422
+ },
423
+ {
424
+ "epoch": 47.15,
425
+ "learning_rate": 0.00028365097862825513,
426
+ "loss": 1.3585,
427
+ "step": 30
428
+ },
429
+ {
430
+ "epoch": 47.15,
431
+ "eval_loss": 1.2553305625915527,
432
+ "eval_runtime": 2.8251,
433
+ "eval_samples_per_second": 82.83,
434
+ "eval_steps_per_second": 2.832,
435
+ "step": 30
436
+ },
437
+ {
438
+ "epoch": 47.15,
439
+ "eval_loss": 1.275472640991211,
440
+ "eval_runtime": 2.6596,
441
+ "eval_samples_per_second": 87.983,
442
+ "eval_steps_per_second": 3.008,
443
+ "step": 30
444
+ },
445
+ {
446
+ "epoch": 49.09,
447
+ "eval_loss": 1.2036432027816772,
448
+ "eval_runtime": 2.6726,
449
+ "eval_samples_per_second": 87.554,
450
+ "eval_steps_per_second": 2.993,
451
+ "step": 31
452
+ },
453
+ {
454
+ "epoch": 49.09,
455
+ "step": 31,
456
+ "total_flos": 8700902454067200.0,
457
+ "train_loss": 1.7344687215743526,
458
+ "train_runtime": 3811.391,
459
+ "train_samples_per_second": 27.51,
460
+ "train_steps_per_second": 0.052
461
+ }
462
+ ],
463
+ "logging_steps": 5,
464
+ "max_steps": 200,
465
+ "num_train_epochs": 50,
466
+ "save_steps": 500,
467
+ "total_flos": 8700902454067200.0,
468
+ "trial_name": null,
469
+ "trial_params": null
470
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fa5152ae64c2764ded13bdb72625ad178fb28b8ba6f06f38d58ed5a17690ba8
3
+ size 5688