File size: 4,451 Bytes
7051cbe
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
{
  "best_metric": 2.358070135116577,
  "best_model_checkpoint": "./Qwen2-4B-Chat-hinglish-sft/checkpoint-275",
  "epoch": 1.0923535253227408,
  "eval_steps": 25,
  "global_step": 275,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 1.1830133199691772,
      "learning_rate": 0.00019999629591162656,
      "loss": 4.7862,
      "step": 25
    },
    {
      "epoch": 0.1,
      "eval_loss": 4.098670482635498,
      "eval_runtime": 4.1063,
      "eval_samples_per_second": 2.679,
      "eval_steps_per_second": 2.679,
      "step": 25
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.0619295835494995,
      "learning_rate": 0.00019932568492674144,
      "loss": 2.9161,
      "step": 50
    },
    {
      "epoch": 0.2,
      "eval_loss": 3.1406357288360596,
      "eval_runtime": 4.0777,
      "eval_samples_per_second": 2.698,
      "eval_steps_per_second": 2.698,
      "step": 50
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.1028335094451904,
      "learning_rate": 0.0001975064532257195,
      "loss": 2.4282,
      "step": 75
    },
    {
      "epoch": 0.3,
      "eval_loss": 2.8003995418548584,
      "eval_runtime": 4.076,
      "eval_samples_per_second": 2.699,
      "eval_steps_per_second": 2.699,
      "step": 75
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.9855464100837708,
      "learning_rate": 0.00019455963874271426,
      "loss": 2.2815,
      "step": 100
    },
    {
      "epoch": 0.4,
      "eval_loss": 2.6393470764160156,
      "eval_runtime": 4.084,
      "eval_samples_per_second": 2.693,
      "eval_steps_per_second": 2.693,
      "step": 100
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.154259204864502,
      "learning_rate": 0.00019051931898913976,
      "loss": 2.1485,
      "step": 125
    },
    {
      "epoch": 0.5,
      "eval_loss": 2.5862808227539062,
      "eval_runtime": 4.0324,
      "eval_samples_per_second": 2.728,
      "eval_steps_per_second": 2.728,
      "step": 125
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.2115691900253296,
      "learning_rate": 0.0001854322169749827,
      "loss": 2.0884,
      "step": 150
    },
    {
      "epoch": 0.6,
      "eval_loss": 2.491671323776245,
      "eval_runtime": 4.0837,
      "eval_samples_per_second": 2.694,
      "eval_steps_per_second": 2.694,
      "step": 150
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.0713191032409668,
      "learning_rate": 0.00017935716089521474,
      "loss": 2.1055,
      "step": 175
    },
    {
      "epoch": 0.7,
      "eval_loss": 2.4541499614715576,
      "eval_runtime": 4.0493,
      "eval_samples_per_second": 2.717,
      "eval_steps_per_second": 2.717,
      "step": 175
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.1404445171356201,
      "learning_rate": 0.00017236440382959126,
      "loss": 2.1012,
      "step": 200
    },
    {
      "epoch": 0.79,
      "eval_loss": 2.42814040184021,
      "eval_runtime": 4.0309,
      "eval_samples_per_second": 2.729,
      "eval_steps_per_second": 2.729,
      "step": 200
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.1801480054855347,
      "learning_rate": 0.00016453481132295506,
      "loss": 1.9956,
      "step": 225
    },
    {
      "epoch": 0.89,
      "eval_loss": 2.3967058658599854,
      "eval_runtime": 4.0145,
      "eval_samples_per_second": 2.74,
      "eval_steps_per_second": 2.74,
      "step": 225
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.2651640176773071,
      "learning_rate": 0.00015595892624101765,
      "loss": 1.9931,
      "step": 250
    },
    {
      "epoch": 0.99,
      "eval_loss": 2.3804991245269775,
      "eval_runtime": 4.0509,
      "eval_samples_per_second": 2.715,
      "eval_steps_per_second": 2.715,
      "step": 250
    },
    {
      "epoch": 1.09,
      "grad_norm": 1.2018358707427979,
      "learning_rate": 0.00014673592171580025,
      "loss": 1.7605,
      "step": 275
    },
    {
      "epoch": 1.09,
      "eval_loss": 2.358070135116577,
      "eval_runtime": 4.0302,
      "eval_samples_per_second": 2.729,
      "eval_steps_per_second": 2.729,
      "step": 275
    }
  ],
  "logging_steps": 25,
  "max_steps": 753,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 25,
  "total_flos": 1972985693091840.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}