Hugging Face's logo

CallMeDaniel
/
Llama-2-7b-chat-hf_vn

Text Generation
Transformers
Safetensors
Vietnamese
llama
conversational
text-generation-inference
Inference Endpoints
8-bit precision
bitsandbytes
Model card Files Community
Llama-2-7b-chat-hf_vn
File size: 3,044 Bytes 
4570a8e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
 
{
  "best_metric": 0.5213077664375305,
  "best_model_checkpoint": "finetuned/Llama-2-7b-chat-hf_vn/checkpoint-200",
  "epoch": 6.042296072507553,
  "eval_steps": 200,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 1.2545,
      "step": 10
    },
    {
      "epoch": 0.6,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 1.122,
      "step": 20
    },
    {
      "epoch": 0.91,
      "learning_rate": 8.999999999999999e-05,
      "loss": 0.9456,
      "step": 30
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.795,
      "step": 40
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00015,
      "loss": 0.6801,
      "step": 50
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.6285,
      "step": 60
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.5934,
      "step": 70
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.5605,
      "step": 80
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.00027,
      "loss": 0.5303,
      "step": 90
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0003,
      "loss": 0.5363,
      "step": 100
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0002869565217391304,
      "loss": 0.5326,
      "step": 110
    },
    {
      "epoch": 3.63,
      "learning_rate": 0.00027391304347826085,
      "loss": 0.5291,
      "step": 120
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.0002608695652173913,
      "loss": 0.5139,
      "step": 130
    },
    {
      "epoch": 4.23,
      "learning_rate": 0.0002478260869565217,
      "loss": 0.4972,
      "step": 140
    },
    {
      "epoch": 4.53,
      "learning_rate": 0.00023478260869565215,
      "loss": 0.5095,
      "step": 150
    },
    {
      "epoch": 4.83,
      "learning_rate": 0.00022173913043478256,
      "loss": 0.5085,
      "step": 160
    },
    {
      "epoch": 5.14,
      "learning_rate": 0.00020869565217391303,
      "loss": 0.499,
      "step": 170
    },
    {
      "epoch": 5.44,
      "learning_rate": 0.00019565217391304346,
      "loss": 0.4926,
      "step": 180
    },
    {
      "epoch": 5.74,
      "learning_rate": 0.0001826086956521739,
      "loss": 0.4789,
      "step": 190
    },
    {
      "epoch": 6.04,
      "learning_rate": 0.00016956521739130433,
      "loss": 0.4882,
      "step": 200
    },
    {
      "epoch": 6.04,
      "eval_loss": 0.5213077664375305,
      "eval_runtime": 243.3322,
      "eval_samples_per_second": 8.219,
      "eval_steps_per_second": 0.686,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 330,
  "num_train_epochs": 10,
  "save_steps": 200,
  "total_flos": 6.452136280530616e+17,
  "trial_name": null,
  "trial_params": null
}