Hugging Face's logo

QomSSLab
/
Legal-gemma3-12b-it

Safetensors
gemma3
Model card Files
xet
 Community
QomSSLab commited on
Commit
00d0a98
·
verified ·
1 Parent(s): 1ccb709

Upload model and tokenizer

Browse files
Files changed (1) hide show
  1. checkpoint-4937/trainer_state.json +167 -0
checkpoint-4937/trainer_state.json ADDED
@@ -0,0 +1,167 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.9998987341772152,
6
+ "eval_steps": 500,
7
+ "global_step": 4937,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.05063291139240506,
14
+ "grad_norm": 43.75,
15
+ "learning_rate": 9.960000000000001e-06,
16
+ "loss": 1.6749,
17
+ "step": 250
18
+ },
19
+ {
20
+ "epoch": 0.10126582278481013,
21
+ "grad_norm": 34.0,
22
+ "learning_rate": 9.930523139001199e-06,
23
+ "loss": 2.2096,
24
+ "step": 500
25
+ },
26
+ {
27
+ "epoch": 0.1518987341772152,
28
+ "grad_norm": 52.5,
29
+ "learning_rate": 9.722924278929705e-06,
30
+ "loss": 2.0945,
31
+ "step": 750
32
+ },
33
+ {
34
+ "epoch": 0.20253164556962025,
35
+ "grad_norm": 44.5,
36
+ "learning_rate": 9.383017947787972e-06,
37
+ "loss": 2.2143,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 0.25316455696202533,
42
+ "grad_norm": 44.25,
43
+ "learning_rate": 8.920326243415839e-06,
44
+ "loss": 2.244,
45
+ "step": 1250
46
+ },
47
+ {
48
+ "epoch": 0.3037974683544304,
49
+ "grad_norm": 60.0,
50
+ "learning_rate": 8.347810959328346e-06,
51
+ "loss": 2.1818,
52
+ "step": 1500
53
+ },
54
+ {
55
+ "epoch": 0.35443037974683544,
56
+ "grad_norm": 39.0,
57
+ "learning_rate": 7.68151047448588e-06,
58
+ "loss": 2.0913,
59
+ "step": 1750
60
+ },
61
+ {
62
+ "epoch": 0.4050632911392405,
63
+ "grad_norm": 33.0,
64
+ "learning_rate": 6.940090455944688e-06,
65
+ "loss": 2.0647,
66
+ "step": 2000
67
+ },
68
+ {
69
+ "epoch": 0.45569620253164556,
70
+ "grad_norm": 47.75,
71
+ "learning_rate": 6.144320960953343e-06,
72
+ "loss": 1.9985,
73
+ "step": 2250
74
+ },
75
+ {
76
+ "epoch": 0.5063291139240507,
77
+ "grad_norm": 32.25,
78
+ "learning_rate": 5.316494586898351e-06,
79
+ "loss": 2.037,
80
+ "step": 2500
81
+ },
82
+ {
83
+ "epoch": 0.5569620253164557,
84
+ "grad_norm": 30.5,
85
+ "learning_rate": 4.479801968981108e-06,
86
+ "loss": 1.917,
87
+ "step": 2750
88
+ },
89
+ {
90
+ "epoch": 0.6075949367088608,
91
+ "grad_norm": 23.0,
92
+ "learning_rate": 3.6576821203642086e-06,
93
+ "loss": 1.8975,
94
+ "step": 3000
95
+ },
96
+ {
97
+ "epoch": 0.6582278481012658,
98
+ "grad_norm": 51.5,
99
+ "learning_rate": 2.8731658142854126e-06,
100
+ "loss": 1.8655,
101
+ "step": 3250
102
+ },
103
+ {
104
+ "epoch": 0.7088607594936709,
105
+ "grad_norm": 24.5,
106
+ "learning_rate": 2.1482304025590883e-06,
107
+ "loss": 1.9389,
108
+ "step": 3500
109
+ },
110
+ {
111
+ "epoch": 0.759493670886076,
112
+ "grad_norm": 27.625,
113
+ "learning_rate": 1.5031841445072532e-06,
114
+ "loss": 1.8563,
115
+ "step": 3750
116
+ },
117
+ {
118
+ "epoch": 0.810126582278481,
119
+ "grad_norm": 28.75,
120
+ "learning_rate": 9.56097293660526e-07,
121
+ "loss": 1.8943,
122
+ "step": 4000
123
+ },
124
+ {
125
+ "epoch": 0.8607594936708861,
126
+ "grad_norm": 34.5,
127
+ "learning_rate": 5.222958797023036e-07,
128
+ "loss": 1.8757,
129
+ "step": 4250
130
+ },
131
+ {
132
+ "epoch": 0.9113924050632911,
133
+ "grad_norm": 60.25,
134
+ "learning_rate": 2.1393236679203845e-07,
135
+ "loss": 1.917,
136
+ "step": 4500
137
+ },
138
+ {
139
+ "epoch": 0.9620253164556962,
140
+ "grad_norm": 21.0,
141
+ "learning_rate": 3.964521579718605e-08,
142
+ "loss": 1.8739,
143
+ "step": 4750
144
+ }
145
+ ],
146
+ "logging_steps": 250,
147
+ "max_steps": 4937,
148
+ "num_input_tokens_seen": 0,
149
+ "num_train_epochs": 1,
150
+ "save_steps": 250,
151
+ "stateful_callbacks": {
152
+ "TrainerControl": {
153
+ "args": {
154
+ "should_epoch_stop": false,
155
+ "should_evaluate": false,
156
+ "should_log": false,
157
+ "should_save": true,
158
+ "should_training_stop": true
159
+ },
160
+ "attributes": {}
161
+ }
162
+ },
163
+ "total_flos": 8.26174299954206e+17,
164
+ "train_batch_size": 1,
165
+ "trial_name": null,
166
+ "trial_params": null
167
+ }