chouss commited on
Commit
41c0d6e
·
verified ·
1 Parent(s): 9bcb4ba

Uploading folder contents

Browse files
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "model/Qwen1.5-0.5B-2048",
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 5,
8
+ "eos_token_id": 6,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 1024,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 2816,
13
+ "max_position_embeddings": 32768,
14
+ "max_window_layers": 21,
15
+ "model_type": "qwen2",
16
+ "num_attention_heads": 16,
17
+ "num_hidden_layers": 24,
18
+ "num_key_value_heads": 16,
19
+ "rms_norm_eps": 1e-06,
20
+ "rope_scaling": null,
21
+ "rope_theta": 1000000.0,
22
+ "sliding_window": null,
23
+ "tie_word_embeddings": true,
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.46.1",
26
+ "use_cache": false,
27
+ "use_sliding_window": false,
28
+ "vocab_size": 10000
29
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 5,
4
+ "eos_token_id": 6,
5
+ "transformers_version": "4.46.1"
6
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba8bd4e4c65a1ac0f047d42a53361c2c940d26430334ff04179410ca9bb2f963
3
+ size 1274613488
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b58e8d7733adcacee750549fa54ebef560144a64f56327ac873a264ef4e07887
3
+ size 2549405370
rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c49abc3bdedbec1fc8e1028ef422150f19ee7470d7b542e1ad8869fc044d2af
3
+ size 15984
rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df12ca4106ff0831785a55b5da88f6c86f6f67bd3d09b2dced4f20b539b14f72
3
+ size 15984
rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05fc0786faff729a3a1582f98b806b68d4f0b76aebb25cbad4431b73176b11c1
3
+ size 15984
rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3117e1218a2dd3f7f8c516a840af48f6b93660d852cca124269f78c21f8577c
3
+ size 15984
rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cdaa30c82476bf6a65e4eb9ca2ae7b95f1b38f41a6f5b2f1cbdda9af86a4a7a0
3
+ size 15984
rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09cdde6931807139efa184e8a98108b74bb05730bc511336966b254b68dc93ee
3
+ size 15984
rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a2268136932c55b3857d38c7cf3fc4bd3cdad532c156b9addebc6d26374374a
3
+ size 15984
rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56cd9a502015b79e0ab94c92a04bd96c99aaf79ef8d64bf81d81eb702c10c2a8
3
+ size 15984
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ff9ec1355bd4ed50235dd5462de954d93a42f55247291c35ffdcf6a1f56bcf6
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "mask_token": {
17
+ "content": "[MASK]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "pad_token": {
24
+ "content": "[PAD]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "sep_token": {
31
+ "content": "[SEP]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "unk_token": {
38
+ "content": "[UNK]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ }
44
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[UNK]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[PAD]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "5": {
44
+ "content": "<s>",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "6": {
52
+ "content": "</s>",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "9006": {
60
+ "content": "<|endoftext|>",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ }
67
+ },
68
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message + '\n' }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\nAssistant:' }}{% elif message['role'] == 'assistant' %}{{ content + '<|endoftext|>' + '\n' }}{% endif %}{% endfor %}",
69
+ "clean_up_tokenization_spaces": false,
70
+ "cls_token": "[CLS]",
71
+ "eos_token": "<|endoftext|>",
72
+ "mask_token": "[MASK]",
73
+ "model_max_length": 1000000000000000019884624838656,
74
+ "pad_token": "[PAD]",
75
+ "padding_side": "right",
76
+ "sep_token": "[SEP]",
77
+ "split_special_tokens": false,
78
+ "tokenizer_class": "PreTrainedTokenizerFast",
79
+ "unk_token": "[UNK]"
80
+ }
trainer_state.json ADDED
@@ -0,0 +1,891 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 15.998185117967331,
5
+ "eval_steps": 100,
6
+ "global_step": 1100,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.14519056261343014,
13
+ "grad_norm": 20.98495864868164,
14
+ "learning_rate": 5.882352941176471e-07,
15
+ "loss": 9.4496,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.29038112522686027,
20
+ "grad_norm": 16.1613712310791,
21
+ "learning_rate": 1.1764705882352942e-06,
22
+ "loss": 8.8125,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.4355716878402904,
27
+ "grad_norm": 6.7023138999938965,
28
+ "learning_rate": 1.7647058823529414e-06,
29
+ "loss": 7.7501,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.5807622504537205,
34
+ "grad_norm": 4.253345489501953,
35
+ "learning_rate": 2.3529411764705885e-06,
36
+ "loss": 7.0378,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.7259528130671506,
41
+ "grad_norm": 2.4256012439727783,
42
+ "learning_rate": 2.9411764705882355e-06,
43
+ "loss": 6.6558,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.8711433756805808,
48
+ "grad_norm": 1.664031982421875,
49
+ "learning_rate": 3.529411764705883e-06,
50
+ "loss": 6.3545,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 1.0181488203266789,
55
+ "grad_norm": 1.206116795539856,
56
+ "learning_rate": 4.11764705882353e-06,
57
+ "loss": 6.2306,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 1.1633393829401089,
62
+ "grad_norm": 1.1027638912200928,
63
+ "learning_rate": 4.705882352941177e-06,
64
+ "loss": 5.8555,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 1.308529945553539,
69
+ "grad_norm": 1.2014143466949463,
70
+ "learning_rate": 5.294117647058824e-06,
71
+ "loss": 5.6277,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 1.453720508166969,
76
+ "grad_norm": 1.3407199382781982,
77
+ "learning_rate": 5.882352941176471e-06,
78
+ "loss": 5.4611,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 1.453720508166969,
83
+ "eval_loss": 5.380606174468994,
84
+ "eval_runtime": 14.5497,
85
+ "eval_samples_per_second": 195.88,
86
+ "eval_steps_per_second": 6.186,
87
+ "step": 100
88
+ },
89
+ {
90
+ "epoch": 1.5989110707803993,
91
+ "grad_norm": 1.6456304788589478,
92
+ "learning_rate": 6.470588235294119e-06,
93
+ "loss": 5.3117,
94
+ "step": 110
95
+ },
96
+ {
97
+ "epoch": 1.7441016333938295,
98
+ "grad_norm": 2.4888839721679688,
99
+ "learning_rate": 7.058823529411766e-06,
100
+ "loss": 5.1868,
101
+ "step": 120
102
+ },
103
+ {
104
+ "epoch": 1.8892921960072595,
105
+ "grad_norm": 2.137110471725464,
106
+ "learning_rate": 7.647058823529411e-06,
107
+ "loss": 5.0505,
108
+ "step": 130
109
+ },
110
+ {
111
+ "epoch": 2.0362976406533577,
112
+ "grad_norm": 2.2031850814819336,
113
+ "learning_rate": 8.23529411764706e-06,
114
+ "loss": 5.0571,
115
+ "step": 140
116
+ },
117
+ {
118
+ "epoch": 2.1814882032667877,
119
+ "grad_norm": 2.2017667293548584,
120
+ "learning_rate": 8.823529411764707e-06,
121
+ "loss": 4.8034,
122
+ "step": 150
123
+ },
124
+ {
125
+ "epoch": 2.3266787658802177,
126
+ "grad_norm": 2.2998626232147217,
127
+ "learning_rate": 9.411764705882354e-06,
128
+ "loss": 4.6938,
129
+ "step": 160
130
+ },
131
+ {
132
+ "epoch": 2.471869328493648,
133
+ "grad_norm": 2.088256359100342,
134
+ "learning_rate": 1e-05,
135
+ "loss": 4.5817,
136
+ "step": 170
137
+ },
138
+ {
139
+ "epoch": 2.617059891107078,
140
+ "grad_norm": 2.1210474967956543,
141
+ "learning_rate": 1.0588235294117648e-05,
142
+ "loss": 4.4561,
143
+ "step": 180
144
+ },
145
+ {
146
+ "epoch": 2.762250453720508,
147
+ "grad_norm": 2.330993175506592,
148
+ "learning_rate": 1.1176470588235295e-05,
149
+ "loss": 4.368,
150
+ "step": 190
151
+ },
152
+ {
153
+ "epoch": 2.907441016333938,
154
+ "grad_norm": 2.2989509105682373,
155
+ "learning_rate": 1.1764705882352942e-05,
156
+ "loss": 4.268,
157
+ "step": 200
158
+ },
159
+ {
160
+ "epoch": 2.907441016333938,
161
+ "eval_loss": 4.2127814292907715,
162
+ "eval_runtime": 14.464,
163
+ "eval_samples_per_second": 197.041,
164
+ "eval_steps_per_second": 6.222,
165
+ "step": 200
166
+ },
167
+ {
168
+ "epoch": 3.0544464609800364,
169
+ "grad_norm": 1.6547863483428955,
170
+ "learning_rate": 1.235294117647059e-05,
171
+ "loss": 4.2584,
172
+ "step": 210
173
+ },
174
+ {
175
+ "epoch": 3.1996370235934664,
176
+ "grad_norm": 2.2462234497070312,
177
+ "learning_rate": 1.2941176470588238e-05,
178
+ "loss": 4.0844,
179
+ "step": 220
180
+ },
181
+ {
182
+ "epoch": 3.344827586206897,
183
+ "grad_norm": 2.176753044128418,
184
+ "learning_rate": 1.3529411764705885e-05,
185
+ "loss": 3.9943,
186
+ "step": 230
187
+ },
188
+ {
189
+ "epoch": 3.490018148820327,
190
+ "grad_norm": 2.2812905311584473,
191
+ "learning_rate": 1.4117647058823532e-05,
192
+ "loss": 3.9237,
193
+ "step": 240
194
+ },
195
+ {
196
+ "epoch": 3.635208711433757,
197
+ "grad_norm": 2.2799627780914307,
198
+ "learning_rate": 1.4705882352941179e-05,
199
+ "loss": 3.849,
200
+ "step": 250
201
+ },
202
+ {
203
+ "epoch": 3.780399274047187,
204
+ "grad_norm": 2.3910913467407227,
205
+ "learning_rate": 1.5294117647058822e-05,
206
+ "loss": 3.7829,
207
+ "step": 260
208
+ },
209
+ {
210
+ "epoch": 3.925589836660617,
211
+ "grad_norm": 2.793339490890503,
212
+ "learning_rate": 1.5882352941176473e-05,
213
+ "loss": 3.7159,
214
+ "step": 270
215
+ },
216
+ {
217
+ "epoch": 4.072595281306715,
218
+ "grad_norm": 4.1607465744018555,
219
+ "learning_rate": 1.647058823529412e-05,
220
+ "loss": 3.7605,
221
+ "step": 280
222
+ },
223
+ {
224
+ "epoch": 4.217785843920145,
225
+ "grad_norm": 2.6675713062286377,
226
+ "learning_rate": 1.7058823529411767e-05,
227
+ "loss": 3.6045,
228
+ "step": 290
229
+ },
230
+ {
231
+ "epoch": 4.362976406533575,
232
+ "grad_norm": 3.1864140033721924,
233
+ "learning_rate": 1.7647058823529414e-05,
234
+ "loss": 3.5499,
235
+ "step": 300
236
+ },
237
+ {
238
+ "epoch": 4.362976406533575,
239
+ "eval_loss": 3.5333378314971924,
240
+ "eval_runtime": 14.5283,
241
+ "eval_samples_per_second": 196.168,
242
+ "eval_steps_per_second": 6.195,
243
+ "step": 300
244
+ },
245
+ {
246
+ "epoch": 4.508166969147005,
247
+ "grad_norm": 3.361107110977173,
248
+ "learning_rate": 1.823529411764706e-05,
249
+ "loss": 3.4898,
250
+ "step": 310
251
+ },
252
+ {
253
+ "epoch": 4.653357531760435,
254
+ "grad_norm": 3.826258897781372,
255
+ "learning_rate": 1.8823529411764708e-05,
256
+ "loss": 3.4595,
257
+ "step": 320
258
+ },
259
+ {
260
+ "epoch": 4.798548094373865,
261
+ "grad_norm": 3.7704880237579346,
262
+ "learning_rate": 1.9411764705882355e-05,
263
+ "loss": 3.4113,
264
+ "step": 330
265
+ },
266
+ {
267
+ "epoch": 4.943738656987296,
268
+ "grad_norm": 3.5223851203918457,
269
+ "learning_rate": 2e-05,
270
+ "loss": 3.3734,
271
+ "step": 340
272
+ },
273
+ {
274
+ "epoch": 5.090744101633394,
275
+ "grad_norm": 3.332577705383301,
276
+ "learning_rate": 1.9999472984871734e-05,
277
+ "loss": 3.4133,
278
+ "step": 350
279
+ },
280
+ {
281
+ "epoch": 5.235934664246824,
282
+ "grad_norm": 4.306556701660156,
283
+ "learning_rate": 1.9997891995035914e-05,
284
+ "loss": 3.2955,
285
+ "step": 360
286
+ },
287
+ {
288
+ "epoch": 5.381125226860254,
289
+ "grad_norm": 4.148169040679932,
290
+ "learning_rate": 1.999525719713366e-05,
291
+ "loss": 3.2456,
292
+ "step": 370
293
+ },
294
+ {
295
+ "epoch": 5.526315789473684,
296
+ "grad_norm": 4.137167930603027,
297
+ "learning_rate": 1.999156886888064e-05,
298
+ "loss": 3.2064,
299
+ "step": 380
300
+ },
301
+ {
302
+ "epoch": 5.671506352087114,
303
+ "grad_norm": 3.123608350753784,
304
+ "learning_rate": 1.998682739903781e-05,
305
+ "loss": 3.1841,
306
+ "step": 390
307
+ },
308
+ {
309
+ "epoch": 5.816696914700545,
310
+ "grad_norm": 4.209367752075195,
311
+ "learning_rate": 1.9981033287370443e-05,
312
+ "loss": 3.1453,
313
+ "step": 400
314
+ },
315
+ {
316
+ "epoch": 5.816696914700545,
317
+ "eval_loss": 3.1412713527679443,
318
+ "eval_runtime": 14.4671,
319
+ "eval_samples_per_second": 196.998,
320
+ "eval_steps_per_second": 6.221,
321
+ "step": 400
322
+ },
323
+ {
324
+ "epoch": 5.961887477313975,
325
+ "grad_norm": 3.708157539367676,
326
+ "learning_rate": 1.9974187144595433e-05,
327
+ "loss": 3.1183,
328
+ "step": 410
329
+ },
330
+ {
331
+ "epoch": 6.108892921960073,
332
+ "grad_norm": 6.567568778991699,
333
+ "learning_rate": 1.9966289692316944e-05,
334
+ "loss": 3.166,
335
+ "step": 420
336
+ },
337
+ {
338
+ "epoch": 6.254083484573503,
339
+ "grad_norm": 4.073953151702881,
340
+ "learning_rate": 1.9957341762950346e-05,
341
+ "loss": 3.0523,
342
+ "step": 430
343
+ },
344
+ {
345
+ "epoch": 6.399274047186933,
346
+ "grad_norm": 4.078774452209473,
347
+ "learning_rate": 1.9947344299634464e-05,
348
+ "loss": 3.018,
349
+ "step": 440
350
+ },
351
+ {
352
+ "epoch": 6.544464609800363,
353
+ "grad_norm": 2.7505741119384766,
354
+ "learning_rate": 1.993629835613218e-05,
355
+ "loss": 2.9874,
356
+ "step": 450
357
+ },
358
+ {
359
+ "epoch": 6.689655172413794,
360
+ "grad_norm": 4.441661834716797,
361
+ "learning_rate": 1.992420509671936e-05,
362
+ "loss": 2.9679,
363
+ "step": 460
364
+ },
365
+ {
366
+ "epoch": 6.834845735027224,
367
+ "grad_norm": 3.656827211380005,
368
+ "learning_rate": 1.9911065796062137e-05,
369
+ "loss": 2.9333,
370
+ "step": 470
371
+ },
372
+ {
373
+ "epoch": 6.980036297640654,
374
+ "grad_norm": 3.519759178161621,
375
+ "learning_rate": 1.9896881839082554e-05,
376
+ "loss": 2.9003,
377
+ "step": 480
378
+ },
379
+ {
380
+ "epoch": 7.127041742286751,
381
+ "grad_norm": 2.58138108253479,
382
+ "learning_rate": 1.9881654720812594e-05,
383
+ "loss": 2.9466,
384
+ "step": 490
385
+ },
386
+ {
387
+ "epoch": 7.272232304900181,
388
+ "grad_norm": 3.1261541843414307,
389
+ "learning_rate": 1.9865386046236597e-05,
390
+ "loss": 2.865,
391
+ "step": 500
392
+ },
393
+ {
394
+ "epoch": 7.272232304900181,
395
+ "eval_loss": 2.8578507900238037,
396
+ "eval_runtime": 14.4738,
397
+ "eval_samples_per_second": 196.908,
398
+ "eval_steps_per_second": 6.218,
399
+ "step": 500
400
+ },
401
+ {
402
+ "epoch": 7.417422867513611,
403
+ "grad_norm": 4.163350582122803,
404
+ "learning_rate": 1.9848077530122083e-05,
405
+ "loss": 2.8359,
406
+ "step": 510
407
+ },
408
+ {
409
+ "epoch": 7.562613430127042,
410
+ "grad_norm": 2.9917635917663574,
411
+ "learning_rate": 1.982973099683902e-05,
412
+ "loss": 2.7953,
413
+ "step": 520
414
+ },
415
+ {
416
+ "epoch": 7.707803992740472,
417
+ "grad_norm": 3.293595790863037,
418
+ "learning_rate": 1.9810348380167527e-05,
419
+ "loss": 2.7838,
420
+ "step": 530
421
+ },
422
+ {
423
+ "epoch": 7.852994555353902,
424
+ "grad_norm": 3.7709453105926514,
425
+ "learning_rate": 1.9789931723094046e-05,
426
+ "loss": 2.7497,
427
+ "step": 540
428
+ },
429
+ {
430
+ "epoch": 7.998185117967332,
431
+ "grad_norm": 3.3971333503723145,
432
+ "learning_rate": 1.9768483177596008e-05,
433
+ "loss": 2.8238,
434
+ "step": 550
435
+ },
436
+ {
437
+ "epoch": 8.14519056261343,
438
+ "grad_norm": 4.206657886505127,
439
+ "learning_rate": 1.9746005004415004e-05,
440
+ "loss": 2.7141,
441
+ "step": 560
442
+ },
443
+ {
444
+ "epoch": 8.29038112522686,
445
+ "grad_norm": 3.42154598236084,
446
+ "learning_rate": 1.9722499572818496e-05,
447
+ "loss": 2.7061,
448
+ "step": 570
449
+ },
450
+ {
451
+ "epoch": 8.43557168784029,
452
+ "grad_norm": 2.6466500759124756,
453
+ "learning_rate": 1.9697969360350098e-05,
454
+ "loss": 2.6849,
455
+ "step": 580
456
+ },
457
+ {
458
+ "epoch": 8.58076225045372,
459
+ "grad_norm": 3.4602091312408447,
460
+ "learning_rate": 1.9672416952568416e-05,
461
+ "loss": 2.6546,
462
+ "step": 590
463
+ },
464
+ {
465
+ "epoch": 8.72595281306715,
466
+ "grad_norm": 2.882288694381714,
467
+ "learning_rate": 1.9645845042774555e-05,
468
+ "loss": 2.6592,
469
+ "step": 600
470
+ },
471
+ {
472
+ "epoch": 8.72595281306715,
473
+ "eval_loss": 2.655930280685425,
474
+ "eval_runtime": 14.4429,
475
+ "eval_samples_per_second": 197.329,
476
+ "eval_steps_per_second": 6.231,
477
+ "step": 600
478
+ },
479
+ {
480
+ "epoch": 8.87114337568058,
481
+ "grad_norm": 2.86531662940979,
482
+ "learning_rate": 1.961825643172819e-05,
483
+ "loss": 2.6246,
484
+ "step": 610
485
+ },
486
+ {
487
+ "epoch": 9.01814882032668,
488
+ "grad_norm": 2.0800743103027344,
489
+ "learning_rate": 1.9589654027352412e-05,
490
+ "loss": 2.6887,
491
+ "step": 620
492
+ },
493
+ {
494
+ "epoch": 9.163339382940109,
495
+ "grad_norm": 2.1428987979888916,
496
+ "learning_rate": 1.956004084442718e-05,
497
+ "loss": 2.6034,
498
+ "step": 630
499
+ },
500
+ {
501
+ "epoch": 9.30852994555354,
502
+ "grad_norm": 2.0337836742401123,
503
+ "learning_rate": 1.9529420004271568e-05,
504
+ "loss": 2.6018,
505
+ "step": 640
506
+ },
507
+ {
508
+ "epoch": 9.453720508166969,
509
+ "grad_norm": 2.25555419921875,
510
+ "learning_rate": 1.9497794734414782e-05,
511
+ "loss": 2.5723,
512
+ "step": 650
513
+ },
514
+ {
515
+ "epoch": 9.5989110707804,
516
+ "grad_norm": 2.281365156173706,
517
+ "learning_rate": 1.9465168368255946e-05,
518
+ "loss": 2.5639,
519
+ "step": 660
520
+ },
521
+ {
522
+ "epoch": 9.744101633393829,
523
+ "grad_norm": 1.9673478603363037,
524
+ "learning_rate": 1.9431544344712776e-05,
525
+ "loss": 2.5486,
526
+ "step": 670
527
+ },
528
+ {
529
+ "epoch": 9.88929219600726,
530
+ "grad_norm": 2.3862695693969727,
531
+ "learning_rate": 1.9396926207859085e-05,
532
+ "loss": 2.5319,
533
+ "step": 680
534
+ },
535
+ {
536
+ "epoch": 10.036297640653357,
537
+ "grad_norm": 2.3560924530029297,
538
+ "learning_rate": 1.936131760655124e-05,
539
+ "loss": 2.5827,
540
+ "step": 690
541
+ },
542
+ {
543
+ "epoch": 10.181488203266788,
544
+ "grad_norm": 3.1034605503082275,
545
+ "learning_rate": 1.932472229404356e-05,
546
+ "loss": 2.514,
547
+ "step": 700
548
+ },
549
+ {
550
+ "epoch": 10.181488203266788,
551
+ "eval_loss": 2.5237531661987305,
552
+ "eval_runtime": 14.6638,
553
+ "eval_samples_per_second": 194.356,
554
+ "eval_steps_per_second": 6.138,
555
+ "step": 700
556
+ },
557
+ {
558
+ "epoch": 10.326678765880217,
559
+ "grad_norm": 2.507720470428467,
560
+ "learning_rate": 1.9287144127592704e-05,
561
+ "loss": 2.5069,
562
+ "step": 710
563
+ },
564
+ {
565
+ "epoch": 10.471869328493648,
566
+ "grad_norm": 2.375530481338501,
567
+ "learning_rate": 1.924858706805112e-05,
568
+ "loss": 2.4914,
569
+ "step": 720
570
+ },
571
+ {
572
+ "epoch": 10.617059891107077,
573
+ "grad_norm": 2.781869649887085,
574
+ "learning_rate": 1.920905517944954e-05,
575
+ "loss": 2.4731,
576
+ "step": 730
577
+ },
578
+ {
579
+ "epoch": 10.762250453720508,
580
+ "grad_norm": 2.3014352321624756,
581
+ "learning_rate": 1.9168552628568632e-05,
582
+ "loss": 2.4679,
583
+ "step": 740
584
+ },
585
+ {
586
+ "epoch": 10.907441016333939,
587
+ "grad_norm": 2.277211904525757,
588
+ "learning_rate": 1.9127083684499805e-05,
589
+ "loss": 2.4708,
590
+ "step": 750
591
+ },
592
+ {
593
+ "epoch": 11.054446460980037,
594
+ "grad_norm": 2.731947660446167,
595
+ "learning_rate": 1.9084652718195237e-05,
596
+ "loss": 2.5219,
597
+ "step": 760
598
+ },
599
+ {
600
+ "epoch": 11.199637023593466,
601
+ "grad_norm": 2.070516347885132,
602
+ "learning_rate": 1.9041264202007158e-05,
603
+ "loss": 2.4407,
604
+ "step": 770
605
+ },
606
+ {
607
+ "epoch": 11.344827586206897,
608
+ "grad_norm": 2.526477098464966,
609
+ "learning_rate": 1.8996922709216456e-05,
610
+ "loss": 2.4408,
611
+ "step": 780
612
+ },
613
+ {
614
+ "epoch": 11.490018148820326,
615
+ "grad_norm": 2.280230760574341,
616
+ "learning_rate": 1.8951632913550625e-05,
617
+ "loss": 2.4217,
618
+ "step": 790
619
+ },
620
+ {
621
+ "epoch": 11.635208711433757,
622
+ "grad_norm": 2.5102462768554688,
623
+ "learning_rate": 1.8905399588691165e-05,
624
+ "loss": 2.4129,
625
+ "step": 800
626
+ },
627
+ {
628
+ "epoch": 11.635208711433757,
629
+ "eval_loss": 2.4271934032440186,
630
+ "eval_runtime": 14.4721,
631
+ "eval_samples_per_second": 196.93,
632
+ "eval_steps_per_second": 6.219,
633
+ "step": 800
634
+ },
635
+ {
636
+ "epoch": 11.780399274047188,
637
+ "grad_norm": 2.3329319953918457,
638
+ "learning_rate": 1.8858227607770398e-05,
639
+ "loss": 2.4068,
640
+ "step": 810
641
+ },
642
+ {
643
+ "epoch": 11.925589836660617,
644
+ "grad_norm": 2.9200258255004883,
645
+ "learning_rate": 1.8810121942857848e-05,
646
+ "loss": 2.393,
647
+ "step": 820
648
+ },
649
+ {
650
+ "epoch": 12.072595281306715,
651
+ "grad_norm": 2.835029363632202,
652
+ "learning_rate": 1.8761087664436137e-05,
653
+ "loss": 2.4508,
654
+ "step": 830
655
+ },
656
+ {
657
+ "epoch": 12.217785843920145,
658
+ "grad_norm": 2.7595760822296143,
659
+ "learning_rate": 1.8711129940866577e-05,
660
+ "loss": 2.3873,
661
+ "step": 840
662
+ },
663
+ {
664
+ "epoch": 12.362976406533575,
665
+ "grad_norm": 2.940290689468384,
666
+ "learning_rate": 1.866025403784439e-05,
667
+ "loss": 2.3817,
668
+ "step": 850
669
+ },
670
+ {
671
+ "epoch": 12.508166969147005,
672
+ "grad_norm": 2.936760902404785,
673
+ "learning_rate": 1.860846531784368e-05,
674
+ "loss": 2.3642,
675
+ "step": 860
676
+ },
677
+ {
678
+ "epoch": 12.653357531760436,
679
+ "grad_norm": 3.160423994064331,
680
+ "learning_rate": 1.8555769239552232e-05,
681
+ "loss": 2.3586,
682
+ "step": 870
683
+ },
684
+ {
685
+ "epoch": 12.798548094373865,
686
+ "grad_norm": 2.8737099170684814,
687
+ "learning_rate": 1.8502171357296144e-05,
688
+ "loss": 2.3481,
689
+ "step": 880
690
+ },
691
+ {
692
+ "epoch": 12.943738656987296,
693
+ "grad_norm": 2.706122398376465,
694
+ "learning_rate": 1.8447677320454367e-05,
695
+ "loss": 2.3496,
696
+ "step": 890
697
+ },
698
+ {
699
+ "epoch": 13.090744101633394,
700
+ "grad_norm": 2.1658377647399902,
701
+ "learning_rate": 1.839229287286327e-05,
702
+ "loss": 2.3982,
703
+ "step": 900
704
+ },
705
+ {
706
+ "epoch": 13.090744101633394,
707
+ "eval_loss": 2.3612313270568848,
708
+ "eval_runtime": 14.4663,
709
+ "eval_samples_per_second": 197.009,
710
+ "eval_steps_per_second": 6.221,
711
+ "step": 900
712
+ },
713
+ {
714
+ "epoch": 13.235934664246823,
715
+ "grad_norm": 2.8399763107299805,
716
+ "learning_rate": 1.8336023852211197e-05,
717
+ "loss": 2.3385,
718
+ "step": 910
719
+ },
720
+ {
721
+ "epoch": 13.381125226860254,
722
+ "grad_norm": 2.8449740409851074,
723
+ "learning_rate": 1.827887618942318e-05,
724
+ "loss": 2.329,
725
+ "step": 920
726
+ },
727
+ {
728
+ "epoch": 13.526315789473685,
729
+ "grad_norm": 3.4475836753845215,
730
+ "learning_rate": 1.8220855908035783e-05,
731
+ "loss": 2.3102,
732
+ "step": 930
733
+ },
734
+ {
735
+ "epoch": 13.671506352087114,
736
+ "grad_norm": 2.820624589920044,
737
+ "learning_rate": 1.816196912356222e-05,
738
+ "loss": 2.3118,
739
+ "step": 940
740
+ },
741
+ {
742
+ "epoch": 13.816696914700545,
743
+ "grad_norm": 2.9867615699768066,
744
+ "learning_rate": 1.8102222042847735e-05,
745
+ "loss": 2.3077,
746
+ "step": 950
747
+ },
748
+ {
749
+ "epoch": 13.961887477313974,
750
+ "grad_norm": 4.323665142059326,
751
+ "learning_rate": 1.8041620963415418e-05,
752
+ "loss": 2.3013,
753
+ "step": 960
754
+ },
755
+ {
756
+ "epoch": 14.108892921960072,
757
+ "grad_norm": 4.4349236488342285,
758
+ "learning_rate": 1.7980172272802398e-05,
759
+ "loss": 2.3514,
760
+ "step": 970
761
+ },
762
+ {
763
+ "epoch": 14.254083484573503,
764
+ "grad_norm": 4.404689311981201,
765
+ "learning_rate": 1.7917882447886585e-05,
766
+ "loss": 2.2881,
767
+ "step": 980
768
+ },
769
+ {
770
+ "epoch": 14.399274047186934,
771
+ "grad_norm": 4.489727020263672,
772
+ "learning_rate": 1.785475805420399e-05,
773
+ "loss": 2.2839,
774
+ "step": 990
775
+ },
776
+ {
777
+ "epoch": 14.544464609800363,
778
+ "grad_norm": 3.8734374046325684,
779
+ "learning_rate": 1.7790805745256703e-05,
780
+ "loss": 2.2785,
781
+ "step": 1000
782
+ },
783
+ {
784
+ "epoch": 14.544464609800363,
785
+ "eval_loss": 2.2907564640045166,
786
+ "eval_runtime": 14.4981,
787
+ "eval_samples_per_second": 196.577,
788
+ "eval_steps_per_second": 6.208,
789
+ "step": 1000
790
+ },
791
+ {
792
+ "epoch": 14.689655172413794,
793
+ "grad_norm": 4.450066089630127,
794
+ "learning_rate": 1.772603226181159e-05,
795
+ "loss": 2.2566,
796
+ "step": 1010
797
+ },
798
+ {
799
+ "epoch": 14.834845735027223,
800
+ "grad_norm": 4.106595993041992,
801
+ "learning_rate": 1.766044443118978e-05,
802
+ "loss": 2.261,
803
+ "step": 1020
804
+ },
805
+ {
806
+ "epoch": 14.980036297640654,
807
+ "grad_norm": 3.8136234283447266,
808
+ "learning_rate": 1.7594049166547073e-05,
809
+ "loss": 2.2535,
810
+ "step": 1030
811
+ },
812
+ {
813
+ "epoch": 15.127041742286751,
814
+ "grad_norm": 5.098197937011719,
815
+ "learning_rate": 1.7526853466145248e-05,
816
+ "loss": 2.3093,
817
+ "step": 1040
818
+ },
819
+ {
820
+ "epoch": 15.272232304900182,
821
+ "grad_norm": 3.752929210662842,
822
+ "learning_rate": 1.7458864412614436e-05,
823
+ "loss": 2.2377,
824
+ "step": 1050
825
+ },
826
+ {
827
+ "epoch": 15.417422867513611,
828
+ "grad_norm": 3.992673397064209,
829
+ "learning_rate": 1.7390089172206594e-05,
830
+ "loss": 2.2416,
831
+ "step": 1060
832
+ },
833
+ {
834
+ "epoch": 15.562613430127042,
835
+ "grad_norm": 3.857272148132324,
836
+ "learning_rate": 1.7320534994040148e-05,
837
+ "loss": 2.2251,
838
+ "step": 1070
839
+ },
840
+ {
841
+ "epoch": 15.707803992740471,
842
+ "grad_norm": 4.833571434020996,
843
+ "learning_rate": 1.725020920933593e-05,
844
+ "loss": 2.2262,
845
+ "step": 1080
846
+ },
847
+ {
848
+ "epoch": 15.852994555353902,
849
+ "grad_norm": 2.919546127319336,
850
+ "learning_rate": 1.717911923064442e-05,
851
+ "loss": 2.2204,
852
+ "step": 1090
853
+ },
854
+ {
855
+ "epoch": 15.998185117967331,
856
+ "grad_norm": 4.030925273895264,
857
+ "learning_rate": 1.710727255106447e-05,
858
+ "loss": 2.274,
859
+ "step": 1100
860
+ },
861
+ {
862
+ "epoch": 15.998185117967331,
863
+ "eval_loss": 2.2391433715820312,
864
+ "eval_runtime": 14.4613,
865
+ "eval_samples_per_second": 197.077,
866
+ "eval_steps_per_second": 6.223,
867
+ "step": 1100
868
+ }
869
+ ],
870
+ "logging_steps": 10,
871
+ "max_steps": 3400,
872
+ "num_input_tokens_seen": 0,
873
+ "num_train_epochs": 50,
874
+ "save_steps": 100,
875
+ "stateful_callbacks": {
876
+ "TrainerControl": {
877
+ "args": {
878
+ "should_epoch_stop": false,
879
+ "should_evaluate": false,
880
+ "should_log": false,
881
+ "should_save": true,
882
+ "should_training_stop": false
883
+ },
884
+ "attributes": {}
885
+ }
886
+ },
887
+ "total_flos": 3.414959764411515e+19,
888
+ "train_batch_size": 16,
889
+ "trial_name": null,
890
+ "trial_params": null
891
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1797df1d4cbc33dde400269e01051c7922eb6ffff263d8c275570953d2b32085
3
+ size 5368