tyzhu commited on
Commit
5fdd675
·
verified ·
1 Parent(s): 227bb0d

Training in progress, epoch 6, checkpoint

Browse files
checkpoint-1312/adapter_config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "auto_mapping": null,
3
- "base_model_name_or_path": "Qwen/Qwen1.5-4B",
4
  "bias": "none",
5
  "fan_in_fan_out": false,
6
  "inference_mode": true,
 
1
  {
2
  "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
4
  "bias": "none",
5
  "fan_in_fan_out": false,
6
  "inference_mode": true,
checkpoint-1312/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26b615a5233ed0312178e1e6f0741d01d2c47aca13153aeb8298a67e71a7ff72
3
+ size 143269386
checkpoint-1312/added_tokens.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "<|endoftext|>": 151643,
3
- "<|im_end|>": 151645,
4
- "<|im_start|>": 151644
5
  }
 
1
  {
2
+ "</s>": 2,
3
+ "<s>": 1,
4
+ "<unk>": 0
5
  }
checkpoint-1312/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:15e3b3bfe2f0783bdfb66c5778a3618ce9becfa3c271646a561af93de0f9f281
3
- size 224537202
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28f3dbc16f4e003422fae6aed2205bd866e74c414ab30858594b06eba1562898
3
+ size 286585234
checkpoint-1312/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b4b729c428f407cf6b1dc5cf043de723c9d572a6becee999dbd4fc4a362b48a5
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d3961ba55f68af864a992966b3a009a1dc58a5afc3d24428b45bfe9bbdd8a35
3
  size 15024
checkpoint-1312/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0d90c023ace87481c8938946860752a4592040cf1c7a9f32a6b7d82465066115
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4779a5baea73478e3675645339eedb3306cd97d45ca58de0f51a6cad68d7f8b7
3
  size 15024
checkpoint-1312/rng_state_2.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c18a7407e0c715b62036a7eef76804a0c28ba7f55ac7d5bcfdda079753d195e6
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e7bd75fcb33d2b3302f4d7ef31b9e1fbe1d703eb9977b7e0f8f78ee79676b6b
3
  size 15024
checkpoint-1312/rng_state_3.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f0505a5e0b126605aa5a33d7af77079cd15d4d72ec3715e0e43ec6af0ed0b6c2
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca4a1254ca681a38b08c66e73439b3c9c2ce7707f99a1b49d2d85a5a27875f36
3
  size 15024
checkpoint-1312/special_tokens_map.json CHANGED
@@ -1,14 +1,6 @@
1
  {
2
- "additional_special_tokens": [
3
- "<|im_start|>",
4
- "<|im_end|>"
5
- ],
6
- "eos_token": {
7
- "content": "<|endoftext|>",
8
- "lstrip": false,
9
- "normalized": false,
10
- "rstrip": false,
11
- "single_word": false
12
- },
13
- "pad_token": "<|endoftext|>"
14
  }
 
1
  {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "</s>",
5
+ "unk_token": "<unk>"
 
 
 
 
 
 
 
 
6
  }
checkpoint-1312/tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
checkpoint-1312/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-1312/tokenizer_config.json CHANGED
@@ -1,24 +1,23 @@
1
  {
2
- "add_prefix_space": false,
3
  "added_tokens_decoder": {
4
- "151643": {
5
- "content": "<|endoftext|>",
6
  "lstrip": false,
7
  "normalized": false,
8
  "rstrip": false,
9
  "single_word": false,
10
  "special": true
11
  },
12
- "151644": {
13
- "content": "<|im_start|>",
14
  "lstrip": false,
15
  "normalized": false,
16
  "rstrip": false,
17
  "single_word": false,
18
  "special": true
19
  },
20
- "151645": {
21
- "content": "<|im_end|>",
22
  "lstrip": false,
23
  "normalized": false,
24
  "rstrip": false,
@@ -26,18 +25,16 @@
26
  "special": true
27
  }
28
  },
29
- "additional_special_tokens": [
30
- "<|im_start|>",
31
- "<|im_end|>"
32
- ],
33
- "bos_token": null,
34
- "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
35
  "clean_up_tokenization_spaces": false,
36
- "eos_token": "<|endoftext|>",
37
- "errors": "replace",
38
- "model_max_length": 32768,
39
- "pad_token": "<|endoftext|>",
40
- "split_special_tokens": false,
41
- "tokenizer_class": "Qwen2Tokenizer",
42
- "unk_token": null
 
 
43
  }
 
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": "<s>",
13
  "lstrip": false,
14
  "normalized": false,
15
  "rstrip": false,
16
  "single_word": false,
17
  "special": true
18
  },
19
+ "2": {
20
+ "content": "</s>",
21
  "lstrip": false,
22
  "normalized": false,
23
  "rstrip": false,
 
25
  "special": true
26
  }
27
  },
28
+ "additional_special_tokens": [],
29
+ "bos_token": "<s>",
 
 
 
 
30
  "clean_up_tokenization_spaces": false,
31
+ "eos_token": "</s>",
32
+ "legacy": false,
33
+ "model_max_length": 1000000000000000019884624838656,
34
+ "pad_token": "</s>",
35
+ "padding_side": "left",
36
+ "sp_model_kwargs": {},
37
+ "tokenizer_class": "LlamaTokenizer",
38
+ "unk_token": "<unk>",
39
+ "use_default_system_prompt": true
40
  }
checkpoint-1312/trainer_state.json CHANGED
@@ -9,221 +9,194 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.5333333333333333,
13
- "grad_norm": 0.7454617619514465,
14
  "learning_rate": 5e-05,
15
- "loss": 1.8726,
16
  "step": 100
17
  },
18
  {
19
- "epoch": 0.9973333333333333,
20
- "eval_accuracy": 0.5713333333333334,
21
- "eval_loss": 1.6331332921981812,
22
- "eval_runtime": 5.4718,
23
- "eval_samples_per_second": 91.377,
24
- "eval_steps_per_second": 11.514,
25
  "step": 187
26
  },
27
  {
28
- "epoch": 0.9973333333333333,
29
- "eval_exact_match": 15.0,
30
- "eval_f1": 25.64857142857145,
31
  "step": 187
32
  },
33
  {
34
- "epoch": 1.0666666666666667,
35
- "grad_norm": 0.9573808908462524,
36
  "learning_rate": 5e-05,
37
- "loss": 1.66,
38
  "step": 200
39
  },
40
  {
41
  "epoch": 1.6,
42
- "grad_norm": 1.0792433023452759,
43
  "learning_rate": 5e-05,
44
- "loss": 1.5764,
45
  "step": 300
46
  },
47
  {
48
  "epoch": 2.0,
49
- "eval_accuracy": 0.5741538461538461,
50
- "eval_loss": 1.6118417978286743,
51
- "eval_runtime": 5.1541,
52
- "eval_samples_per_second": 97.01,
53
- "eval_steps_per_second": 12.223,
54
  "step": 375
55
  },
56
  {
57
  "epoch": 2.0,
58
- "eval_exact_match": 16.8,
59
- "eval_f1": 26.120952380952396,
60
  "step": 375
61
  },
62
  {
63
- "epoch": 2.1333333333333333,
64
- "grad_norm": 1.2928162813186646,
65
  "learning_rate": 5e-05,
66
- "loss": 1.5364,
67
  "step": 400
68
  },
69
  {
70
- "epoch": 2.6666666666666665,
71
- "grad_norm": 1.4883633852005005,
72
  "learning_rate": 5e-05,
73
- "loss": 1.4154,
74
  "step": 500
75
  },
76
  {
77
- "epoch": 2.997333333333333,
78
- "eval_accuracy": 0.5733846153846154,
79
- "eval_loss": 1.6421321630477905,
80
- "eval_runtime": 5.23,
81
- "eval_samples_per_second": 95.603,
82
- "eval_steps_per_second": 12.046,
83
  "step": 562
84
  },
85
  {
86
- "epoch": 2.997333333333333,
87
- "eval_exact_match": 19.4,
88
- "eval_f1": 28.4652380952381,
89
  "step": 562
90
  },
91
  {
92
  "epoch": 3.2,
93
- "grad_norm": 1.7553108930587769,
94
  "learning_rate": 5e-05,
95
- "loss": 1.3426,
96
  "step": 600
97
  },
98
  {
99
- "epoch": 3.7333333333333334,
100
- "grad_norm": 2.0343973636627197,
101
  "learning_rate": 5e-05,
102
- "loss": 1.2297,
103
  "step": 700
104
  },
105
  {
106
  "epoch": 4.0,
107
- "eval_accuracy": 0.5687692307692308,
108
- "eval_loss": 1.7009365558624268,
109
- "eval_runtime": 5.4367,
110
- "eval_samples_per_second": 91.968,
111
- "eval_steps_per_second": 11.588,
112
  "step": 750
113
  },
114
  {
115
  "epoch": 4.0,
116
- "eval_exact_match": 17.6,
117
- "eval_f1": 26.490952380952383,
118
  "step": 750
119
  },
120
  {
121
- "epoch": 4.266666666666667,
122
- "grad_norm": 2.3617172241210938,
123
  "learning_rate": 5e-05,
124
- "loss": 1.1613,
125
  "step": 800
126
  },
127
  {
128
  "epoch": 4.8,
129
- "grad_norm": 2.357593297958374,
130
  "learning_rate": 5e-05,
131
- "loss": 1.0779,
132
  "step": 900
133
  },
134
  {
135
- "epoch": 4.997333333333334,
136
- "eval_accuracy": 0.567025641025641,
137
- "eval_loss": 1.791063666343689,
138
- "eval_runtime": 6.0902,
139
- "eval_samples_per_second": 82.099,
140
- "eval_steps_per_second": 10.345,
141
  "step": 937
142
  },
143
  {
144
- "epoch": 4.997333333333334,
145
- "eval_exact_match": 16.0,
146
- "eval_f1": 23.740000000000013,
147
  "step": 937
148
  },
149
  {
150
- "epoch": 5.333333333333333,
151
- "grad_norm": 2.7846109867095947,
152
  "learning_rate": 5e-05,
153
- "loss": 0.9807,
154
  "step": 1000
155
  },
156
  {
157
- "epoch": 5.866666666666667,
158
- "grad_norm": 2.713942050933838,
159
  "learning_rate": 5e-05,
160
- "loss": 0.9547,
161
  "step": 1100
162
  },
163
  {
164
  "epoch": 6.0,
165
- "eval_accuracy": 0.563076923076923,
166
- "eval_loss": 1.8913958072662354,
167
- "eval_runtime": 5.2006,
168
- "eval_samples_per_second": 96.142,
169
- "eval_steps_per_second": 12.114,
170
  "step": 1125
171
  },
172
  {
173
  "epoch": 6.0,
174
- "eval_exact_match": 15.6,
175
- "eval_f1": 24.21666666666668,
176
  "step": 1125
177
  },
178
  {
179
  "epoch": 6.4,
180
- "grad_norm": 2.6644272804260254,
181
  "learning_rate": 5e-05,
182
- "loss": 0.8401,
183
  "step": 1200
184
  },
185
  {
186
- "epoch": 6.933333333333334,
187
- "grad_norm": 3.0467491149902344,
188
  "learning_rate": 5e-05,
189
- "loss": 0.8414,
190
  "step": 1300
191
  },
192
  {
193
- "epoch": 6.997333333333334,
194
- "eval_accuracy": 0.5606666666666666,
195
- "eval_loss": 1.987471342086792,
196
- "eval_runtime": 5.1815,
197
- "eval_samples_per_second": 96.497,
198
- "eval_steps_per_second": 12.159,
199
  "step": 1312
200
  },
201
  {
202
- "epoch": 6.997333333333334,
203
- "eval_exact_match": 15.4,
204
- "eval_f1": 24.453174603174613,
205
  "step": 1312
206
  }
207
  ],
208
  "logging_steps": 100,
209
  "max_steps": 9350,
210
- "num_input_tokens_seen": 0,
211
  "num_train_epochs": 50,
212
  "save_steps": 500,
213
- "stateful_callbacks": {
214
- "TrainerControl": {
215
- "args": {
216
- "should_epoch_stop": false,
217
- "should_evaluate": false,
218
- "should_log": false,
219
- "should_save": true,
220
- "should_training_stop": false
221
- },
222
- "attributes": {}
223
- }
224
- },
225
- "total_flos": 2.285823521193984e+16,
226
- "train_batch_size": 1,
227
  "trial_name": null,
228
  "trial_params": null
229
  }
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.53,
 
13
  "learning_rate": 5e-05,
14
+ "loss": 1.7923,
15
  "step": 100
16
  },
17
  {
18
+ "epoch": 1.0,
19
+ "eval_accuracy": 0.6128205128205129,
20
+ "eval_loss": 1.2805120944976807,
21
+ "eval_runtime": 4.4685,
22
+ "eval_samples_per_second": 111.895,
23
+ "eval_steps_per_second": 14.099,
24
  "step": 187
25
  },
26
  {
27
+ "epoch": 1.0,
28
+ "eval_exact_match": 26.2,
29
+ "eval_f1": 37.42854700854702,
30
  "step": 187
31
  },
32
  {
33
+ "epoch": 1.07,
 
34
  "learning_rate": 5e-05,
35
+ "loss": 1.306,
36
  "step": 200
37
  },
38
  {
39
  "epoch": 1.6,
 
40
  "learning_rate": 5e-05,
41
+ "loss": 1.2488,
42
  "step": 300
43
  },
44
  {
45
  "epoch": 2.0,
46
+ "eval_accuracy": 0.6167692307692307,
47
+ "eval_loss": 1.2676523923873901,
48
+ "eval_runtime": 4.1151,
49
+ "eval_samples_per_second": 121.504,
50
+ "eval_steps_per_second": 15.309,
51
  "step": 375
52
  },
53
  {
54
  "epoch": 2.0,
55
+ "eval_exact_match": 26.0,
56
+ "eval_f1": 39.16730158730158,
57
  "step": 375
58
  },
59
  {
60
+ "epoch": 2.13,
 
61
  "learning_rate": 5e-05,
62
+ "loss": 1.1905,
63
  "step": 400
64
  },
65
  {
66
+ "epoch": 2.67,
 
67
  "learning_rate": 5e-05,
68
+ "loss": 1.1097,
69
  "step": 500
70
  },
71
  {
72
+ "epoch": 3.0,
73
+ "eval_accuracy": 0.6161538461538462,
74
+ "eval_loss": 1.2942619323730469,
75
+ "eval_runtime": 4.2002,
76
+ "eval_samples_per_second": 119.043,
77
+ "eval_steps_per_second": 14.999,
78
  "step": 562
79
  },
80
  {
81
+ "epoch": 3.0,
82
+ "eval_exact_match": 25.8,
83
+ "eval_f1": 38.45095238095238,
84
  "step": 562
85
  },
86
  {
87
  "epoch": 3.2,
 
88
  "learning_rate": 5e-05,
89
+ "loss": 1.0433,
90
  "step": 600
91
  },
92
  {
93
+ "epoch": 3.73,
 
94
  "learning_rate": 5e-05,
95
+ "loss": 0.9244,
96
  "step": 700
97
  },
98
  {
99
  "epoch": 4.0,
100
+ "eval_accuracy": 0.6126153846153846,
101
+ "eval_loss": 1.359810709953308,
102
+ "eval_runtime": 4.0966,
103
+ "eval_samples_per_second": 122.052,
104
+ "eval_steps_per_second": 15.379,
105
  "step": 750
106
  },
107
  {
108
  "epoch": 4.0,
109
+ "eval_exact_match": 24.2,
110
+ "eval_f1": 37.012380952380965,
111
  "step": 750
112
  },
113
  {
114
+ "epoch": 4.27,
 
115
  "learning_rate": 5e-05,
116
+ "loss": 0.8571,
117
  "step": 800
118
  },
119
  {
120
  "epoch": 4.8,
 
121
  "learning_rate": 5e-05,
122
+ "loss": 0.7924,
123
  "step": 900
124
  },
125
  {
126
+ "epoch": 5.0,
127
+ "eval_accuracy": 0.6088717948717949,
128
+ "eval_loss": 1.4713767766952515,
129
+ "eval_runtime": 4.1965,
130
+ "eval_samples_per_second": 119.148,
131
+ "eval_steps_per_second": 15.013,
132
  "step": 937
133
  },
134
  {
135
+ "epoch": 5.0,
136
+ "eval_exact_match": 22.4,
137
+ "eval_f1": 34.34000000000002,
138
  "step": 937
139
  },
140
  {
141
+ "epoch": 5.33,
 
142
  "learning_rate": 5e-05,
143
+ "loss": 0.721,
144
  "step": 1000
145
  },
146
  {
147
+ "epoch": 5.87,
 
148
  "learning_rate": 5e-05,
149
+ "loss": 0.6864,
150
  "step": 1100
151
  },
152
  {
153
  "epoch": 6.0,
154
+ "eval_accuracy": 0.6044615384615385,
155
+ "eval_loss": 1.5761321783065796,
156
+ "eval_runtime": 4.4048,
157
+ "eval_samples_per_second": 113.513,
158
+ "eval_steps_per_second": 14.303,
159
  "step": 1125
160
  },
161
  {
162
  "epoch": 6.0,
163
+ "eval_exact_match": 22.8,
164
+ "eval_f1": 34.564761904761916,
165
  "step": 1125
166
  },
167
  {
168
  "epoch": 6.4,
 
169
  "learning_rate": 5e-05,
170
+ "loss": 0.6126,
171
  "step": 1200
172
  },
173
  {
174
+ "epoch": 6.93,
 
175
  "learning_rate": 5e-05,
176
+ "loss": 0.6101,
177
  "step": 1300
178
  },
179
  {
180
+ "epoch": 7.0,
181
+ "eval_accuracy": 0.6028717948717949,
182
+ "eval_loss": 1.6553617715835571,
183
+ "eval_runtime": 4.3743,
184
+ "eval_samples_per_second": 114.305,
185
+ "eval_steps_per_second": 14.402,
186
  "step": 1312
187
  },
188
  {
189
+ "epoch": 7.0,
190
+ "eval_exact_match": 20.4,
191
+ "eval_f1": 31.70952380952382,
192
  "step": 1312
193
  }
194
  ],
195
  "logging_steps": 100,
196
  "max_steps": 9350,
 
197
  "num_train_epochs": 50,
198
  "save_steps": 500,
199
+ "total_flos": 4.899502260276429e+16,
 
 
 
 
 
 
 
 
 
 
 
 
 
200
  "trial_name": null,
201
  "trial_params": null
202
  }
checkpoint-1312/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dd2191b7045c2a2985850d3b9f1472eaf3359ca38b9bc68efae9970e68b817ab
3
- size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:742db4eb1b1050153055c5b82de0ef39604f0c615b7ee389916bfc19979a9577
3
+ size 4728