Hugging Face's logo

yinmingzhang
/
qwen2.5-0.5b-zhihu-sft

Text Generation
Transformers
Safetensors
qwen2
Generated from Trainer
trl
sft
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
yinmingzhang commited on
Commit
09f143b
·
verified ·
1 Parent(s): fab5bb2

Model save

Browse files
Files changed (7) hide show
  1. README.md +1 -1
  2. adapter_config.json +2 -2
  3. adapter_model.safetensors +1 -1
  4. all_results.json +6 -6
  5. train_results.json +6 -6
  6. trainer_state.json +114 -84
  7. training_args.bin +1 -1
README.md CHANGED
@@ -27,7 +27,7 @@ print(output["generated_text"])
27
 
28
  ## Training procedure
29
 
30
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/mzyin-university-of-florida/huggingface/runs/d7nzn042)
31
 
32
 
33
  This model was trained with SFT.
 
27
 
28
  ## Training procedure
29
 
30
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/mzyin-university-of-florida/huggingface/runs/t93s4ryo)
31
 
32
 
33
  This model was trained with SFT.
adapter_config.json CHANGED
@@ -23,8 +23,8 @@
23
  "rank_pattern": {},
24
  "revision": null,
25
  "target_modules": [
26
- "q_proj",
27
- "v_proj"
28
  ],
29
  "task_type": "CAUSAL_LM",
30
  "use_dora": false,
 
23
  "rank_pattern": {},
24
  "revision": null,
25
  "target_modules": [
26
+ "v_proj",
27
+ "q_proj"
28
  ],
29
  "task_type": "CAUSAL_LM",
30
  "use_dora": false,
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bfb69b113065b3ebbae4fc959507a704194f2173e04ccab9645085911d126d5f
3
  size 545743328
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8ee017b466ac8e1af621f225814f55dc2c20824ed0966a3bdb375ba1006db6a
3
  size 545743328
all_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 9.666666666666666,
3
- "total_flos": 1376786000642048.0,
4
- "train_loss": 6.840779709815979,
5
- "train_runtime": 46.6654,
6
  "train_samples": 568,
7
- "train_samples_per_second": 19.072,
8
- "train_steps_per_second": 0.214
9
  }
 
1
  {
2
+ "epoch": 5.0,
3
+ "total_flos": 18203285913600.0,
4
+ "train_loss": 3.477997573216756,
5
+ "train_runtime": 36.1944,
6
  "train_samples": 568,
7
+ "train_samples_per_second": 12.295,
8
+ "train_steps_per_second": 1.658
9
  }
train_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 9.666666666666666,
3
- "total_flos": 1376786000642048.0,
4
- "train_loss": 6.840779709815979,
5
- "train_runtime": 46.6654,
6
  "train_samples": 568,
7
- "train_samples_per_second": 19.072,
8
- "train_steps_per_second": 0.214
9
  }
 
1
  {
2
+ "epoch": 5.0,
3
+ "total_flos": 18203285913600.0,
4
+ "train_loss": 3.477997573216756,
5
+ "train_runtime": 36.1944,
6
  "train_samples": 568,
7
+ "train_samples_per_second": 12.295,
8
+ "train_steps_per_second": 1.658
9
  }
trainer_state.json CHANGED
@@ -1,128 +1,158 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 9.666666666666666,
5
  "eval_steps": 500,
6
- "global_step": 10,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.6666666666666666,
13
- "grad_norm": 0.31581476137364617,
14
- "learning_rate": 2e-06,
15
- "loss": 3.566,
16
  "step": 1
17
  },
18
  {
19
- "epoch": 0.6666666666666666,
20
- "eval_loss": 3.655458688735962,
21
- "eval_runtime": 0.3709,
22
- "eval_samples_per_second": 48.534,
23
- "eval_steps_per_second": 8.089,
24
- "step": 1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25
  },
26
  {
27
  "epoch": 1.6666666666666665,
28
- "eval_loss": 3.6550912857055664,
29
- "eval_runtime": 0.3817,
30
- "eval_samples_per_second": 47.155,
31
- "eval_steps_per_second": 7.859,
32
- "step": 2
33
  },
34
  {
35
- "epoch": 2.6666666666666665,
36
- "eval_loss": 3.6545002460479736,
37
- "eval_runtime": 0.374,
38
- "eval_samples_per_second": 48.126,
39
- "eval_steps_per_second": 8.021,
40
- "step": 3
41
  },
42
  {
43
- "epoch": 3.6666666666666665,
44
- "eval_loss": 3.6540424823760986,
45
- "eval_runtime": 0.3751,
46
- "eval_samples_per_second": 47.981,
47
- "eval_steps_per_second": 7.997,
48
- "step": 4
49
  },
50
  {
51
- "epoch": 4.666666666666667,
52
- "grad_norm": 0.6240134572232815,
53
- "learning_rate": 1.1736481776669305e-06,
54
- "loss": 7.2366,
55
- "step": 5
56
  },
57
  {
58
- "epoch": 4.666666666666667,
59
- "eval_loss": 3.6533453464508057,
60
- "eval_runtime": 0.3708,
61
- "eval_samples_per_second": 48.543,
62
- "eval_steps_per_second": 8.091,
63
- "step": 5
 
 
 
 
 
 
 
64
  },
65
  {
66
- "epoch": 5.666666666666667,
67
- "eval_loss": 3.6530237197875977,
68
- "eval_runtime": 0.3753,
69
- "eval_samples_per_second": 47.956,
70
- "eval_steps_per_second": 7.993,
71
- "step": 6
72
  },
73
  {
74
- "epoch": 6.666666666666667,
75
- "eval_loss": 3.6526899337768555,
76
- "eval_runtime": 0.3681,
77
- "eval_samples_per_second": 48.895,
78
- "eval_steps_per_second": 8.149,
79
- "step": 7
80
  },
81
  {
82
- "epoch": 7.666666666666667,
83
- "eval_loss": 3.6524767875671387,
84
- "eval_runtime": 0.3812,
85
- "eval_samples_per_second": 47.218,
86
- "eval_steps_per_second": 7.87,
87
- "step": 8
88
  },
89
  {
90
- "epoch": 8.666666666666666,
91
- "eval_loss": 3.6523280143737793,
92
- "eval_runtime": 0.3692,
93
- "eval_samples_per_second": 48.752,
94
- "eval_steps_per_second": 8.125,
95
- "step": 9
96
  },
97
  {
98
- "epoch": 9.666666666666666,
99
- "grad_norm": 0.5211683877512812,
 
 
 
 
 
 
 
100
  "learning_rate": 0.0,
101
- "loss": 7.1791,
102
- "step": 10
103
  },
104
  {
105
- "epoch": 9.666666666666666,
106
- "eval_loss": 3.652298927307129,
107
- "eval_runtime": 0.3728,
108
- "eval_samples_per_second": 48.284,
109
- "eval_steps_per_second": 8.047,
110
- "step": 10
111
  },
112
  {
113
- "epoch": 9.666666666666666,
114
- "step": 10,
115
- "total_flos": 1376786000642048.0,
116
- "train_loss": 6.840779709815979,
117
- "train_runtime": 46.6654,
118
- "train_samples_per_second": 19.072,
119
- "train_steps_per_second": 0.214
120
  }
121
  ],
122
  "logging_steps": 5,
123
- "max_steps": 10,
124
  "num_input_tokens_seen": 0,
125
- "num_train_epochs": 10,
126
  "save_steps": 500,
127
  "stateful_callbacks": {
128
  "TrainerControl": {
@@ -136,7 +166,7 @@
136
  "attributes": {}
137
  }
138
  },
139
- "total_flos": 1376786000642048.0,
140
  "train_batch_size": 8,
141
  "trial_name": null,
142
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 5.0,
5
  "eval_steps": 500,
6
+ "global_step": 60,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.08333333333333333,
13
+ "grad_norm": 1.9183009817804988,
14
+ "learning_rate": 3.3333333333333333e-06,
15
+ "loss": 3.4904,
16
  "step": 1
17
  },
18
  {
19
+ "epoch": 0.4166666666666667,
20
+ "grad_norm": 1.4669764740494382,
21
+ "learning_rate": 1.6666666666666667e-05,
22
+ "loss": 3.6113,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.8333333333333334,
27
+ "grad_norm": 2.3102441270892657,
28
+ "learning_rate": 1.973044870579824e-05,
29
+ "loss": 3.5363,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 1.0,
34
+ "eval_loss": 3.6004714965820312,
35
+ "eval_runtime": 0.3916,
36
+ "eval_samples_per_second": 45.968,
37
+ "eval_steps_per_second": 7.661,
38
+ "step": 12
39
+ },
40
+ {
41
+ "epoch": 1.25,
42
+ "grad_norm": 1.2536552391599605,
43
+ "learning_rate": 1.866025403784439e-05,
44
+ "loss": 3.4578,
45
+ "step": 15
46
  },
47
  {
48
  "epoch": 1.6666666666666665,
49
+ "grad_norm": 0.8379504173528296,
50
+ "learning_rate": 1.686241637868734e-05,
51
+ "loss": 3.516,
52
+ "step": 20
 
53
  },
54
  {
55
+ "epoch": 2.0,
56
+ "eval_loss": 3.560063362121582,
57
+ "eval_runtime": 0.395,
58
+ "eval_samples_per_second": 45.564,
59
+ "eval_steps_per_second": 7.594,
60
+ "step": 24
61
  },
62
  {
63
+ "epoch": 2.0833333333333335,
64
+ "grad_norm": 0.8826743196650887,
65
+ "learning_rate": 1.4487991802004625e-05,
66
+ "loss": 3.429,
67
+ "step": 25
 
68
  },
69
  {
70
+ "epoch": 2.5,
71
+ "grad_norm": 0.9643551070955534,
72
+ "learning_rate": 1.1736481776669307e-05,
73
+ "loss": 3.4933,
74
+ "step": 30
75
  },
76
  {
77
+ "epoch": 2.9166666666666665,
78
+ "grad_norm": 0.800770231101502,
79
+ "learning_rate": 8.839070858747697e-06,
80
+ "loss": 3.4836,
81
+ "step": 35
82
+ },
83
+ {
84
+ "epoch": 3.0,
85
+ "eval_loss": 3.534764289855957,
86
+ "eval_runtime": 0.5846,
87
+ "eval_samples_per_second": 30.789,
88
+ "eval_steps_per_second": 5.131,
89
+ "step": 36
90
  },
91
  {
92
+ "epoch": 3.3333333333333335,
93
+ "grad_norm": 0.8243232107653643,
94
+ "learning_rate": 6.039202339608432e-06,
95
+ "loss": 3.4287,
96
+ "step": 40
 
97
  },
98
  {
99
+ "epoch": 3.75,
100
+ "grad_norm": 1.0751974758407474,
101
+ "learning_rate": 3.5721239031346067e-06,
102
+ "loss": 3.5007,
103
+ "step": 45
 
104
  },
105
  {
106
+ "epoch": 4.0,
107
+ "eval_loss": 3.523860454559326,
108
+ "eval_runtime": 0.3897,
109
+ "eval_samples_per_second": 46.188,
110
+ "eval_steps_per_second": 7.698,
111
+ "step": 48
112
  },
113
  {
114
+ "epoch": 4.166666666666667,
115
+ "grad_norm": 0.7780096021946287,
116
+ "learning_rate": 1.6451218858706374e-06,
117
+ "loss": 3.4041,
118
+ "step": 50
 
119
  },
120
  {
121
+ "epoch": 4.583333333333333,
122
+ "grad_norm": 0.9901654284407495,
123
+ "learning_rate": 4.2010487684511105e-07,
124
+ "loss": 3.4807,
125
+ "step": 55
126
+ },
127
+ {
128
+ "epoch": 5.0,
129
+ "grad_norm": 1.3948344306556273,
130
  "learning_rate": 0.0,
131
+ "loss": 3.4187,
132
+ "step": 60
133
  },
134
  {
135
+ "epoch": 5.0,
136
+ "eval_loss": 3.5221145153045654,
137
+ "eval_runtime": 0.3973,
138
+ "eval_samples_per_second": 45.3,
139
+ "eval_steps_per_second": 7.55,
140
+ "step": 60
141
  },
142
  {
143
+ "epoch": 5.0,
144
+ "step": 60,
145
+ "total_flos": 18203285913600.0,
146
+ "train_loss": 3.477997573216756,
147
+ "train_runtime": 36.1944,
148
+ "train_samples_per_second": 12.295,
149
+ "train_steps_per_second": 1.658
150
  }
151
  ],
152
  "logging_steps": 5,
153
+ "max_steps": 60,
154
  "num_input_tokens_seen": 0,
155
+ "num_train_epochs": 5,
156
  "save_steps": 500,
157
  "stateful_callbacks": {
158
  "TrainerControl": {
 
166
  "attributes": {}
167
  }
168
  },
169
+ "total_flos": 18203285913600.0,
170
  "train_batch_size": 8,
171
  "trial_name": null,
172
  "trial_params": null
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:40cae7246274451657f506f8583541c02934b9be8d7b704351eadd8e3a7d5526
3
  size 7544
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef5f8a3833a329a348e58e3e8dcca749f3c66651ed96679d686727e3e3c31c4c
3
  size 7544