Hugging Face's logo

philschmid
/
llama-3-1-8b-math-orca-spectrum-10k-ep1

Text Generation
Transformers
TensorBoard
Safetensors
llama
Generated from Trainer
trl
sft
conversational
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
philschmid HF staff commited on
Commit
f965aa8
•
1 Parent(s): 61f0815

Model save

Browse files
Files changed (4) hide show
  1. all_results.json +6 -6
  2. config.json +1 -1
  3. train_results.json +6 -6
  4. trainer_state.json +155 -36
all_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 0.9882352941176471,
3
- "total_flos": 8684660632584192.0,
4
- "train_loss": 0.4845818508239019,
5
- "train_runtime": 2191.8228,
6
  "train_samples": 10000,
7
- "train_samples_per_second": 2.468,
8
- "train_steps_per_second": 0.019
9
  }
 
1
  {
2
+ "epoch": 2.9647058823529413,
3
+ "total_flos": 2.6054191277408256e+16,
4
+ "train_loss": 0.28340424479000154,
5
+ "train_runtime": 5349.4515,
6
  "train_samples": 10000,
7
+ "train_samples_per_second": 3.033,
8
+ "train_steps_per_second": 0.024
9
  }
config.json CHANGED
@@ -31,6 +31,6 @@
31
  "tie_word_embeddings": false,
32
  "torch_dtype": "bfloat16",
33
  "transformers_version": "4.46.3",
34
- "use_cache": false,
35
  "vocab_size": 128256
36
  }
 
31
  "tie_word_embeddings": false,
32
  "torch_dtype": "bfloat16",
33
  "transformers_version": "4.46.3",
34
+ "use_cache": true,
35
  "vocab_size": 128256
36
  }
train_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 0.9882352941176471,
3
- "total_flos": 8684660632584192.0,
4
- "train_loss": 0.4845818508239019,
5
- "train_runtime": 2191.8228,
6
  "train_samples": 10000,
7
- "train_samples_per_second": 2.468,
8
- "train_steps_per_second": 0.019
9
  }
 
1
  {
2
+ "epoch": 2.9647058823529413,
3
+ "total_flos": 2.6054191277408256e+16,
4
+ "train_loss": 0.28340424479000154,
5
+ "train_runtime": 5349.4515,
6
  "train_samples": 10000,
7
+ "train_samples_per_second": 3.033,
8
+ "train_steps_per_second": 0.024
9
  }
trainer_state.json CHANGED
@@ -1,83 +1,202 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 0.9882352941176471,
5
  "eval_steps": 500,
6
- "global_step": 42,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.11764705882352941,
13
- "grad_norm": 3.502065765112122,
14
- "learning_rate": 5e-05,
15
- "loss": 1.2955,
16
  "step": 5
17
  },
18
  {
19
  "epoch": 0.23529411764705882,
20
- "grad_norm": 1.0166353187103343,
21
- "learning_rate": 4.77807122597034e-05,
22
- "loss": 0.5761,
23
  "step": 10
24
  },
25
  {
26
  "epoch": 0.35294117647058826,
27
- "grad_norm": 2.0609265403495867,
28
- "learning_rate": 4.151686808475204e-05,
29
- "loss": 0.4265,
30
  "step": 15
31
  },
32
  {
33
  "epoch": 0.47058823529411764,
34
- "grad_norm": 0.42584227426186005,
35
- "learning_rate": 3.232056928191376e-05,
36
- "loss": 0.3626,
37
  "step": 20
38
  },
39
  {
40
  "epoch": 0.5882352941176471,
41
- "grad_norm": 0.31196594170810965,
42
- "learning_rate": 2.182455450632803e-05,
43
- "loss": 0.337,
44
  "step": 25
45
  },
46
  {
47
  "epoch": 0.7058823529411765,
48
- "grad_norm": 0.2951365308766864,
49
- "learning_rate": 1.1892317911069212e-05,
50
- "loss": 0.3255,
51
  "step": 30
52
  },
53
  {
54
  "epoch": 0.8235294117647058,
55
- "grad_norm": 0.2794268174086481,
56
- "learning_rate": 4.2872587689039484e-06,
57
- "loss": 0.3152,
58
  "step": 35
59
  },
60
  {
61
  "epoch": 0.9411764705882353,
62
- "grad_norm": 0.27293344082468357,
63
- "learning_rate": 3.5960224130728857e-07,
64
- "loss": 0.3135,
65
  "step": 40
66
  },
67
  {
68
- "epoch": 0.9882352941176471,
69
- "step": 42,
70
- "total_flos": 8684660632584192.0,
71
- "train_loss": 0.4845818508239019,
72
- "train_runtime": 2191.8228,
73
- "train_samples_per_second": 2.468,
74
- "train_steps_per_second": 0.019
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
75
  }
76
  ],
77
  "logging_steps": 5,
78
- "max_steps": 42,
79
  "num_input_tokens_seen": 0,
80
- "num_train_epochs": 1,
81
  "save_steps": 500,
82
  "stateful_callbacks": {
83
  "TrainerControl": {
@@ -91,7 +210,7 @@
91
  "attributes": {}
92
  }
93
  },
94
- "total_flos": 8684660632584192.0,
95
  "train_batch_size": 8,
96
  "trial_name": null,
97
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 2.9647058823529413,
5
  "eval_steps": 500,
6
+ "global_step": 126,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.11764705882352941,
13
+ "grad_norm": 7.926609856300782,
14
+ "learning_rate": 1.923076923076923e-05,
15
+ "loss": 1.3952,
16
  "step": 5
17
  },
18
  {
19
  "epoch": 0.23529411764705882,
20
+ "grad_norm": 1.8405266678960521,
21
+ "learning_rate": 3.846153846153846e-05,
22
+ "loss": 0.7195,
23
  "step": 10
24
  },
25
  {
26
  "epoch": 0.35294117647058826,
27
+ "grad_norm": 0.9213360863093222,
28
+ "learning_rate": 4.99613632163459e-05,
29
+ "loss": 0.4751,
30
  "step": 15
31
  },
32
  {
33
  "epoch": 0.47058823529411764,
34
+ "grad_norm": 0.5013556648810926,
35
+ "learning_rate": 4.952806974561518e-05,
36
+ "loss": 0.3764,
37
  "step": 20
38
  },
39
  {
40
  "epoch": 0.5882352941176471,
41
+ "grad_norm": 0.36651550460940663,
42
+ "learning_rate": 4.862157403595598e-05,
43
+ "loss": 0.3376,
44
  "step": 25
45
  },
46
  {
47
  "epoch": 0.7058823529411765,
48
+ "grad_norm": 0.2990382699652553,
49
+ "learning_rate": 4.7259364450857096e-05,
50
+ "loss": 0.319,
51
  "step": 30
52
  },
53
  {
54
  "epoch": 0.8235294117647058,
55
+ "grad_norm": 0.3237256028681643,
56
+ "learning_rate": 4.5467721110696685e-05,
57
+ "loss": 0.3004,
58
  "step": 35
59
  },
60
  {
61
  "epoch": 0.9411764705882353,
62
+ "grad_norm": 0.279108419607827,
63
+ "learning_rate": 4.3281208889462715e-05,
64
+ "loss": 0.2876,
65
  "step": 40
66
  },
67
  {
68
+ "epoch": 1.0588235294117647,
69
+ "grad_norm": 0.2913644070197406,
70
+ "learning_rate": 4.0742010579737855e-05,
71
+ "loss": 0.2781,
72
+ "step": 45
73
+ },
74
+ {
75
+ "epoch": 1.1764705882352942,
76
+ "grad_norm": 0.31192283061108456,
77
+ "learning_rate": 3.7899113090712526e-05,
78
+ "loss": 0.2299,
79
+ "step": 50
80
+ },
81
+ {
82
+ "epoch": 1.2941176470588236,
83
+ "grad_norm": 0.26219335707946045,
84
+ "learning_rate": 3.4807362379317025e-05,
85
+ "loss": 0.2162,
86
+ "step": 55
87
+ },
88
+ {
89
+ "epoch": 1.4117647058823528,
90
+ "grad_norm": 0.29615778350655847,
91
+ "learning_rate": 3.1526405346999946e-05,
92
+ "loss": 0.2084,
93
+ "step": 60
94
+ },
95
+ {
96
+ "epoch": 1.5294117647058822,
97
+ "grad_norm": 0.28688518177379235,
98
+ "learning_rate": 2.8119539115370218e-05,
99
+ "loss": 0.2001,
100
+ "step": 65
101
+ },
102
+ {
103
+ "epoch": 1.6470588235294117,
104
+ "grad_norm": 0.2521941986482173,
105
+ "learning_rate": 2.4652489880792128e-05,
106
+ "loss": 0.1928,
107
+ "step": 70
108
+ },
109
+ {
110
+ "epoch": 1.7647058823529411,
111
+ "grad_norm": 0.23146994181766112,
112
+ "learning_rate": 2.1192144906604876e-05,
113
+ "loss": 0.1901,
114
+ "step": 75
115
+ },
116
+ {
117
+ "epoch": 1.8823529411764706,
118
+ "grad_norm": 0.24231550036213442,
119
+ "learning_rate": 1.780526211572016e-05,
120
+ "loss": 0.1842,
121
+ "step": 80
122
+ },
123
+ {
124
+ "epoch": 2.0,
125
+ "grad_norm": 0.39003765031270565,
126
+ "learning_rate": 1.4557182178490636e-05,
127
+ "loss": 0.1977,
128
+ "step": 85
129
+ },
130
+ {
131
+ "epoch": 2.1176470588235294,
132
+ "grad_norm": 0.30306479354663846,
133
+ "learning_rate": 1.151056794260289e-05,
134
+ "loss": 0.134,
135
+ "step": 90
136
+ },
137
+ {
138
+ "epoch": 2.235294117647059,
139
+ "grad_norm": 0.28623172192127194,
140
+ "learning_rate": 8.724195524258688e-06,
141
+ "loss": 0.1285,
142
+ "step": 95
143
+ },
144
+ {
145
+ "epoch": 2.3529411764705883,
146
+ "grad_norm": 0.24645273464562403,
147
+ "learning_rate": 6.25182038324447e-06,
148
+ "loss": 0.1293,
149
+ "step": 100
150
+ },
151
+ {
152
+ "epoch": 2.4705882352941178,
153
+ "grad_norm": 0.23945572748242794,
154
+ "learning_rate": 4.14114025787932e-06,
155
+ "loss": 0.1239,
156
+ "step": 105
157
+ },
158
+ {
159
+ "epoch": 2.588235294117647,
160
+ "grad_norm": 0.2367978149293599,
161
+ "learning_rate": 2.4328749671846116e-06,
162
+ "loss": 0.1256,
163
+ "step": 110
164
+ },
165
+ {
166
+ "epoch": 2.7058823529411766,
167
+ "grad_norm": 0.24102969031047347,
168
+ "learning_rate": 1.1599808329836177e-06,
169
+ "loss": 0.1245,
170
+ "step": 115
171
+ },
172
+ {
173
+ "epoch": 2.8235294117647056,
174
+ "grad_norm": 0.22666386671902916,
175
+ "learning_rate": 3.470148775153448e-07,
176
+ "loss": 0.1242,
177
+ "step": 120
178
+ },
179
+ {
180
+ "epoch": 2.9411764705882355,
181
+ "grad_norm": 0.21907314464708774,
182
+ "learning_rate": 9.661062636148744e-09,
183
+ "loss": 0.121,
184
+ "step": 125
185
+ },
186
+ {
187
+ "epoch": 2.9647058823529413,
188
+ "step": 126,
189
+ "total_flos": 2.6054191277408256e+16,
190
+ "train_loss": 0.28340424479000154,
191
+ "train_runtime": 5349.4515,
192
+ "train_samples_per_second": 3.033,
193
+ "train_steps_per_second": 0.024
194
  }
195
  ],
196
  "logging_steps": 5,
197
+ "max_steps": 126,
198
  "num_input_tokens_seen": 0,
199
+ "num_train_epochs": 3,
200
  "save_steps": 500,
201
  "stateful_callbacks": {
202
  "TrainerControl": {
 
210
  "attributes": {}
211
  }
212
  },
213
+ "total_flos": 2.6054191277408256e+16,
214
  "train_batch_size": 8,
215
  "trial_name": null,
216
  "trial_params": null