SystemAdmin123 commited on
Commit
e878739
·
verified ·
1 Parent(s): 19bbf3b

Training in progress, step 200, checkpoint

Browse files
last-checkpoint/config.json CHANGED
@@ -1,25 +1,26 @@
1
  {
2
- "_name_or_path": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
6
  "attention_bias": false,
7
  "attention_dropout": 0.0,
8
- "bos_token_id": 1,
9
  "eos_token_id": 2,
10
- "head_dim": 64,
11
  "hidden_act": "silu",
12
- "hidden_size": 2048,
13
  "initializer_range": 0.02,
14
- "intermediate_size": 5632,
15
  "max_position_embeddings": 2048,
16
  "mlp_bias": false,
17
  "model_type": "llama",
18
- "num_attention_heads": 32,
19
- "num_hidden_layers": 22,
20
  "num_key_value_heads": 4,
 
21
  "pretraining_tp": 1,
22
- "rms_norm_eps": 1e-05,
23
  "rope_scaling": null,
24
  "rope_theta": 10000.0,
25
  "tie_word_embeddings": false,
 
1
  {
2
+ "_name_or_path": "trl-internal-testing/tiny-random-LlamaForCausalLM",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
6
  "attention_bias": false,
7
  "attention_dropout": 0.0,
8
+ "bos_token_id": 0,
9
  "eos_token_id": 2,
10
+ "head_dim": 4,
11
  "hidden_act": "silu",
12
+ "hidden_size": 16,
13
  "initializer_range": 0.02,
14
+ "intermediate_size": 64,
15
  "max_position_embeddings": 2048,
16
  "mlp_bias": false,
17
  "model_type": "llama",
18
+ "num_attention_heads": 4,
19
+ "num_hidden_layers": 2,
20
  "num_key_value_heads": 4,
21
+ "pad_token_id": -1,
22
  "pretraining_tp": 1,
23
+ "rms_norm_eps": 1e-06,
24
  "rope_scaling": null,
25
  "rope_theta": 10000.0,
26
  "tie_word_embeddings": false,
last-checkpoint/generation_config.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "bos_token_id": 1,
 
3
  "do_sample": true,
4
- "eos_token_id": 2,
5
- "max_length": 2048,
6
- "pad_token_id": 0,
7
  "transformers_version": "4.48.1"
8
  }
 
1
  {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
  "do_sample": true,
5
+ "eos_token_id": 1,
6
+ "pad_token_id": 2,
 
7
  "transformers_version": "4.48.1"
8
  }
last-checkpoint/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9f733e9041aab06f212569d2073af6e6c0acf30fcc84bec18f30feab19bc086f
3
- size 2200119864
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23835aa3baa0dcf892abf69863f1f0c1ea20a2c3908d99240a722cf7177381d0
3
+ size 2066752
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2e7f6e0595a900a9079a5615b8fdd966aefa419bcc59d072afe009c68faa40b1
3
- size 2235234810
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5de0b4166267d002b9869980fe6745c1cfa3e9331d120a3e69c74cbfb28d2bac
3
+ size 2162798
last-checkpoint/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5e139d403fa01e97c7f0b1fb6f366f88e99b3ef2d1090658accfac79d4e4bf7a
3
  size 15984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9927ec66b801c0f4855203d926f0323830f03e7abef37344059118107768879f
3
  size 15984
last-checkpoint/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:163db0385f3f2c6cc1cdde3fd1262ef8c70bc95cb0742071833b93242de76dd2
3
  size 15984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:319005a7acd12212946f28795d824f4aa3bb9ae8b8a8fa8426acf8d1683c7711
3
  size 15984
last-checkpoint/rng_state_2.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a3f2223c650d0a4e2baa05aba88577beb8cf1341342d6523706f17972a2f481c
3
  size 15984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6e15e0019059647c2b2120160188600af8acd95ba49388e85735382c6fec110
3
  size 15984
last-checkpoint/rng_state_3.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f06016ff924c2dbb5f6bd90ff3c51224f11735ecac5d42e1b0d6eb800acc7b67
3
  size 15984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee6bd9483dc526698ba668e9cd1c38d218e9960bb3776fb46ce2d46403967e58
3
  size 15984
last-checkpoint/rng_state_4.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7099cb24dc8750eaaa6a5df6cead1dfa156e46f7a3a478fb7ea1f45a9599d5c1
3
  size 15984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e3c33d4b9177e702fbc3b7fe348e5bc53c0acdd468c8fa1d8d47c41439445bb
3
  size 15984
last-checkpoint/rng_state_5.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:19562befb91856d315c73b5e1fe6c1554bd4ffd5d803c7f9588b2b2a18b10eb7
3
  size 15984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:befcb54359f53141f634f3331a650f261e163fb42877249c8517ea83d36a7124
3
  size 15984
last-checkpoint/rng_state_6.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b055646d6b89d5f656bbcc0f7774bff592c8649c36beead05e7d98a901842815
3
  size 15984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d14477440f631aa977c46e3038b7d41c1e28d50b31065840e8a8470d3bdfe80
3
  size 15984
last-checkpoint/rng_state_7.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6d80f9631b5929172f6acf476e00b42058e6cff03360fb2e0c295ebe290c9cd6
3
  size 15984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a246be5cf8c2f6460956f9737bacb1b56ffc6195bb80d3c3d20f45b6c2226d5
3
  size 15984
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3a60c7d771c1fd156acee762fba03c724cb41829a3f71df370ecd1d20b134982
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5197657cfe0474e002b5e477499687f2485548beb0dd654c8444eec350c0acb8
3
  size 1064
last-checkpoint/special_tokens_map.json CHANGED
@@ -14,7 +14,7 @@
14
  "single_word": false
15
  },
16
  "pad_token": {
17
- "content": "</s>",
18
  "lstrip": false,
19
  "normalized": false,
20
  "rstrip": false,
 
14
  "single_word": false
15
  },
16
  "pad_token": {
17
+ "content": "<unk>",
18
  "lstrip": false,
19
  "normalized": false,
20
  "rstrip": false,
last-checkpoint/tokenizer.json CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c88bda6bdd84543eadebdf4bd2ec325ae43b71a3ff1fead6a766562c00b29bd8
3
- size 3619016
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d606f46a8aa6f29f0a0abdec7c3ffddefc9f9bfe26919532d209a0a850e25029
3
+ size 3619013
last-checkpoint/tokenizer_config.json CHANGED
@@ -29,17 +29,15 @@
29
  }
30
  },
31
  "bos_token": "<s>",
32
- "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 %}",
33
  "clean_up_tokenization_spaces": false,
34
  "eos_token": "</s>",
35
  "extra_special_tokens": {},
36
- "legacy": false,
37
  "model_max_length": 2048,
38
- "pad_token": "</s>",
39
- "padding_side": "right",
40
- "sp_model_kwargs": {},
41
  "tokenizer_class": "LlamaTokenizer",
42
  "unk_token": "<unk>",
43
- "use_default_system_prompt": false,
44
  "use_fast": true
45
  }
 
29
  }
30
  },
31
  "bos_token": "<s>",
32
+ "chat_template": "{% for message in messages %}{% if message['role'] == 'user' %}{{ '### Instruction: ' + message['content'] + '\n\n' }}{% elif message['role'] == 'assistant' %}{{ '### Response: ' + message['content'] + eos_token}}{% endif %}{% endfor %}",
33
  "clean_up_tokenization_spaces": false,
34
  "eos_token": "</s>",
35
  "extra_special_tokens": {},
36
+ "legacy": true,
37
  "model_max_length": 2048,
38
+ "pad_token": "<unk>",
 
 
39
  "tokenizer_class": "LlamaTokenizer",
40
  "unk_token": "<unk>",
41
+ "use_default_system_prompt": true,
42
  "use_fast": true
43
  }
last-checkpoint/trainer_state.json CHANGED
@@ -1,96 +1,174 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 12.5,
5
  "eval_steps": 200,
6
- "global_step": 100,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.125,
13
- "eval_loss": 2.0389645099639893,
14
- "eval_runtime": 3.3031,
15
- "eval_samples_per_second": 454.427,
16
- "eval_steps_per_second": 4.844,
17
  "step": 1
18
  },
19
  {
20
- "epoch": 1.25,
21
- "grad_norm": 2.234375,
22
- "learning_rate": 0.00019863613034027224,
23
- "loss": 1.8385,
24
  "step": 10
25
  },
26
  {
27
- "epoch": 2.5,
28
- "grad_norm": 1.90625,
29
- "learning_rate": 0.0001879473751206489,
30
- "loss": 1.1862,
31
  "step": 20
32
  },
33
  {
34
- "epoch": 3.75,
35
- "grad_norm": 1.375,
36
- "learning_rate": 0.00016772815716257412,
37
- "loss": 0.615,
38
  "step": 30
39
  },
40
  {
41
- "epoch": 5.0,
42
- "grad_norm": 0.82421875,
43
- "learning_rate": 0.00014016954246529696,
44
- "loss": 0.2473,
45
  "step": 40
46
  },
47
  {
48
- "epoch": 6.25,
49
- "grad_norm": 0.498046875,
50
- "learning_rate": 0.00010825793454723325,
51
- "loss": 0.0731,
52
  "step": 50
53
  },
54
  {
55
- "epoch": 7.5,
56
- "grad_norm": 0.29296875,
57
- "learning_rate": 7.54514512859201e-05,
58
- "loss": 0.0275,
59
  "step": 60
60
  },
61
  {
62
- "epoch": 8.75,
63
- "grad_norm": 0.11962890625,
64
- "learning_rate": 4.530518418775733e-05,
65
- "loss": 0.0112,
66
  "step": 70
67
  },
68
  {
69
- "epoch": 10.0,
70
- "grad_norm": 0.06787109375,
71
- "learning_rate": 2.1085949060360654e-05,
72
- "loss": 0.0078,
73
  "step": 80
74
  },
75
  {
76
- "epoch": 11.25,
77
- "grad_norm": 0.061767578125,
78
- "learning_rate": 5.418275829936537e-06,
79
- "loss": 0.0072,
80
  "step": 90
81
  },
82
  {
83
- "epoch": 12.5,
84
- "grad_norm": 0.0615234375,
85
- "learning_rate": 0.0,
86
- "loss": 0.007,
87
  "step": 100
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
88
  }
89
  ],
90
  "logging_steps": 10,
91
- "max_steps": 100,
92
  "num_input_tokens_seen": 0,
93
- "num_train_epochs": 13,
94
  "save_steps": 200,
95
  "stateful_callbacks": {
96
  "TrainerControl": {
@@ -99,13 +177,13 @@
99
  "should_evaluate": false,
100
  "should_log": false,
101
  "should_save": true,
102
- "should_training_stop": true
103
  },
104
  "attributes": {}
105
  }
106
  },
107
- "total_flos": 1.2203604110737408e+17,
108
- "train_batch_size": 12,
109
  "trial_name": null,
110
  "trial_params": null
111
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 66.66666666666667,
5
  "eval_steps": 200,
6
+ "global_step": 200,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.3333333333333333,
13
+ "eval_loss": 10.375551223754883,
14
+ "eval_runtime": 1.0096,
15
+ "eval_samples_per_second": 1486.793,
16
+ "eval_steps_per_second": 5.943,
17
  "step": 1
18
  },
19
  {
20
+ "epoch": 3.3333333333333335,
21
+ "grad_norm": 0.1015625,
22
+ "learning_rate": 0.00019945218953682734,
23
+ "loss": 10.3718,
24
  "step": 10
25
  },
26
  {
27
+ "epoch": 6.666666666666667,
28
+ "grad_norm": 0.162109375,
29
+ "learning_rate": 0.00019781476007338058,
30
+ "loss": 10.3553,
31
  "step": 20
32
  },
33
  {
34
+ "epoch": 10.0,
35
+ "grad_norm": 0.291015625,
36
+ "learning_rate": 0.00019510565162951537,
37
+ "loss": 10.3251,
38
  "step": 30
39
  },
40
  {
41
+ "epoch": 13.333333333333334,
42
+ "grad_norm": 0.330078125,
43
+ "learning_rate": 0.0001913545457642601,
44
+ "loss": 10.2723,
45
  "step": 40
46
  },
47
  {
48
+ "epoch": 16.666666666666668,
49
+ "grad_norm": 0.328125,
50
+ "learning_rate": 0.00018660254037844388,
51
+ "loss": 10.2096,
52
  "step": 50
53
  },
54
  {
55
+ "epoch": 20.0,
56
+ "grad_norm": 0.328125,
57
+ "learning_rate": 0.00018090169943749476,
58
+ "loss": 10.1499,
59
  "step": 60
60
  },
61
  {
62
+ "epoch": 23.333333333333332,
63
+ "grad_norm": 0.326171875,
64
+ "learning_rate": 0.00017431448254773944,
65
+ "loss": 10.0935,
66
  "step": 70
67
  },
68
  {
69
+ "epoch": 26.666666666666668,
70
+ "grad_norm": 0.333984375,
71
+ "learning_rate": 0.00016691306063588583,
72
+ "loss": 10.0398,
73
  "step": 80
74
  },
75
  {
76
+ "epoch": 30.0,
77
+ "grad_norm": 0.337890625,
78
+ "learning_rate": 0.00015877852522924732,
79
+ "loss": 9.9895,
80
  "step": 90
81
  },
82
  {
83
+ "epoch": 33.333333333333336,
84
+ "grad_norm": 0.341796875,
85
+ "learning_rate": 0.00015000000000000001,
86
+ "loss": 9.9424,
87
  "step": 100
88
+ },
89
+ {
90
+ "epoch": 36.666666666666664,
91
+ "grad_norm": 0.34375,
92
+ "learning_rate": 0.00014067366430758004,
93
+ "loss": 9.8995,
94
+ "step": 110
95
+ },
96
+ {
97
+ "epoch": 40.0,
98
+ "grad_norm": 0.345703125,
99
+ "learning_rate": 0.00013090169943749476,
100
+ "loss": 9.859,
101
+ "step": 120
102
+ },
103
+ {
104
+ "epoch": 43.333333333333336,
105
+ "grad_norm": 0.34765625,
106
+ "learning_rate": 0.00012079116908177593,
107
+ "loss": 9.8216,
108
+ "step": 130
109
+ },
110
+ {
111
+ "epoch": 46.666666666666664,
112
+ "grad_norm": 0.3515625,
113
+ "learning_rate": 0.00011045284632676536,
114
+ "loss": 9.7872,
115
+ "step": 140
116
+ },
117
+ {
118
+ "epoch": 50.0,
119
+ "grad_norm": 0.357421875,
120
+ "learning_rate": 0.0001,
121
+ "loss": 9.7569,
122
+ "step": 150
123
+ },
124
+ {
125
+ "epoch": 53.333333333333336,
126
+ "grad_norm": 0.35546875,
127
+ "learning_rate": 8.954715367323468e-05,
128
+ "loss": 9.7325,
129
+ "step": 160
130
+ },
131
+ {
132
+ "epoch": 56.666666666666664,
133
+ "grad_norm": 0.359375,
134
+ "learning_rate": 7.920883091822408e-05,
135
+ "loss": 9.712,
136
+ "step": 170
137
+ },
138
+ {
139
+ "epoch": 60.0,
140
+ "grad_norm": 0.359375,
141
+ "learning_rate": 6.909830056250527e-05,
142
+ "loss": 9.697,
143
+ "step": 180
144
+ },
145
+ {
146
+ "epoch": 63.333333333333336,
147
+ "grad_norm": 0.361328125,
148
+ "learning_rate": 5.9326335692419995e-05,
149
+ "loss": 9.6841,
150
+ "step": 190
151
+ },
152
+ {
153
+ "epoch": 66.66666666666667,
154
+ "grad_norm": 0.361328125,
155
+ "learning_rate": 5.000000000000002e-05,
156
+ "loss": 9.6746,
157
+ "step": 200
158
+ },
159
+ {
160
+ "epoch": 66.66666666666667,
161
+ "eval_loss": 9.681697845458984,
162
+ "eval_runtime": 0.9466,
163
+ "eval_samples_per_second": 1585.661,
164
+ "eval_steps_per_second": 6.338,
165
+ "step": 200
166
  }
167
  ],
168
  "logging_steps": 10,
169
+ "max_steps": 300,
170
  "num_input_tokens_seen": 0,
171
+ "num_train_epochs": 100,
172
  "save_steps": 200,
173
  "stateful_callbacks": {
174
  "TrainerControl": {
 
177
  "should_evaluate": false,
178
  "should_log": false,
179
  "should_save": true,
180
+ "should_training_stop": false
181
  },
182
  "attributes": {}
183
  }
184
  },
185
+ "total_flos": 327326839603200.0,
186
+ "train_batch_size": 32,
187
  "trial_name": null,
188
  "trial_params": null
189
  }
last-checkpoint/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:917005e54e6a74be4d5bc3c0309ff36483f7719e494b126c6cee70ee375d0421
3
- size 7032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9e9fca0c23d9cd44121007d8469411a44af500190e6fe6533d8ba7e8ebaeafa
3
+ size 6840