Hugging Face's logo

yatsby
/
qlora-gemini-persona-qna-finetuned

PEFT
Model card Files Community
Chungyezun commited on
Commit
16838da
•
1 Parent(s): 98e9453

loss decreased

Browse files
Files changed (6) hide show
  1. adapter_model.bin +1 -1
  2. optimizer.pt +1 -1
  3. rng_state.pth +1 -1
  4. scheduler.pt +1 -1
  5. trainer_state.json +215 -35
  6. training_args.bin +1 -1
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d3b2b49fca5ef8bb3d5ea2db44159f973e5c3b4e4618cd9cf18de1da5f75234a
3
  size 26243937
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8acb0fbc98773f098e075e84f27ea9e58fc3219c739103550375409cc17a86e
3
  size 26243937
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:00d78df1d412a2050984de6ce82ad261a242d1b2f9a689bd1cdbf477de15923b
3
  size 3344517
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30d3709246c5902bea0dcf308ec3283e3b30df703419bce6545cb8f0c3f7ac9d
3
  size 3344517
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:279b02a9bf16df9152cfab5ba81e333a442181c6fedb218bfd3466771417286d
3
  size 14575
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9540e64b4b2c1b0831135b670b0e4ec129fdd0e6e6ee83fa19b063d1c57bca89
3
  size 14575
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f5300549c1360a61070d0f470484db3f343b09445bd39f11900cfcc98362b653
3
  size 627
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64b7240815379c413f6274e52aac65f3d6f2f24afff3dd72c81fc19a3d08e133
3
  size 627
trainer_state.json CHANGED
@@ -1,76 +1,256 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 1.2824623276691247,
5
- "global_step": 500,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
- "epoch": 0.13,
12
- "learning_rate": 9.357326478149101e-05,
13
- "loss": 1.7132,
14
  "step": 50
15
  },
16
  {
17
- "epoch": 0.26,
18
- "learning_rate": 8.7146529562982e-05,
19
- "loss": 1.3446,
20
  "step": 100
21
  },
22
  {
23
- "epoch": 0.38,
24
- "learning_rate": 8.071979434447301e-05,
25
- "loss": 1.2811,
26
  "step": 150
27
  },
28
  {
29
- "epoch": 0.51,
30
- "learning_rate": 7.429305912596401e-05,
31
- "loss": 1.2723,
32
  "step": 200
33
  },
34
  {
35
- "epoch": 0.64,
36
- "learning_rate": 6.786632390745502e-05,
37
- "loss": 1.2589,
38
  "step": 250
39
  },
40
  {
41
- "epoch": 0.77,
42
- "learning_rate": 6.143958868894601e-05,
43
- "loss": 1.2409,
44
  "step": 300
45
  },
46
  {
47
- "epoch": 0.9,
48
- "learning_rate": 5.501285347043702e-05,
49
- "loss": 1.2233,
50
  "step": 350
51
  },
52
  {
53
- "epoch": 1.03,
54
- "learning_rate": 4.8586118251928024e-05,
55
- "loss": 1.2315,
56
  "step": 400
57
  },
58
  {
59
- "epoch": 1.15,
60
- "learning_rate": 4.215938303341902e-05,
61
- "loss": 1.2169,
62
  "step": 450
63
  },
64
  {
65
- "epoch": 1.28,
66
- "learning_rate": 3.573264781491003e-05,
67
- "loss": 1.2081,
68
  "step": 500
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
69
  }
70
  ],
71
- "max_steps": 778,
72
- "num_train_epochs": 2,
73
- "total_flos": 2.1034715008303104e+17,
74
  "trial_name": null,
75
  "trial_params": null
76
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 2.9125329935378175,
5
+ "global_step": 2000,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
+ "epoch": 0.07,
12
+ "learning_rate": 9.757045675413022e-05,
13
+ "loss": 1.7333,
14
  "step": 50
15
  },
16
  {
17
+ "epoch": 0.15,
18
+ "learning_rate": 9.514091350826045e-05,
19
+ "loss": 1.3819,
20
  "step": 100
21
  },
22
  {
23
+ "epoch": 0.22,
24
+ "learning_rate": 9.271137026239067e-05,
25
+ "loss": 1.3078,
26
  "step": 150
27
  },
28
  {
29
+ "epoch": 0.29,
30
+ "learning_rate": 9.02818270165209e-05,
31
+ "loss": 1.3028,
32
  "step": 200
33
  },
34
  {
35
+ "epoch": 0.36,
36
+ "learning_rate": 8.785228377065113e-05,
37
+ "loss": 1.2742,
38
  "step": 250
39
  },
40
  {
41
+ "epoch": 0.44,
42
+ "learning_rate": 8.542274052478134e-05,
43
+ "loss": 1.2594,
44
  "step": 300
45
  },
46
  {
47
+ "epoch": 0.51,
48
+ "learning_rate": 8.299319727891157e-05,
49
+ "loss": 1.2554,
50
  "step": 350
51
  },
52
  {
53
+ "epoch": 0.58,
54
+ "learning_rate": 8.056365403304179e-05,
55
+ "loss": 1.2412,
56
  "step": 400
57
  },
58
  {
59
+ "epoch": 0.66,
60
+ "learning_rate": 7.8134110787172e-05,
61
+ "loss": 1.2448,
62
  "step": 450
63
  },
64
  {
65
+ "epoch": 0.73,
66
+ "learning_rate": 7.570456754130224e-05,
67
+ "loss": 1.2325,
68
  "step": 500
69
+ },
70
+ {
71
+ "epoch": 0.8,
72
+ "learning_rate": 7.327502429543247e-05,
73
+ "loss": 1.2233,
74
+ "step": 550
75
+ },
76
+ {
77
+ "epoch": 0.87,
78
+ "learning_rate": 7.08454810495627e-05,
79
+ "loss": 1.2326,
80
+ "step": 600
81
+ },
82
+ {
83
+ "epoch": 0.95,
84
+ "learning_rate": 6.841593780369291e-05,
85
+ "loss": 1.2112,
86
+ "step": 650
87
+ },
88
+ {
89
+ "epoch": 1.02,
90
+ "learning_rate": 6.598639455782313e-05,
91
+ "loss": 1.2144,
92
+ "step": 700
93
+ },
94
+ {
95
+ "epoch": 1.09,
96
+ "learning_rate": 6.355685131195336e-05,
97
+ "loss": 1.1948,
98
+ "step": 750
99
+ },
100
+ {
101
+ "epoch": 1.17,
102
+ "learning_rate": 6.112730806608357e-05,
103
+ "loss": 1.2022,
104
+ "step": 800
105
+ },
106
+ {
107
+ "epoch": 1.24,
108
+ "learning_rate": 5.8697764820213804e-05,
109
+ "loss": 1.2028,
110
+ "step": 850
111
+ },
112
+ {
113
+ "epoch": 1.31,
114
+ "learning_rate": 5.626822157434403e-05,
115
+ "loss": 1.1978,
116
+ "step": 900
117
+ },
118
+ {
119
+ "epoch": 1.38,
120
+ "learning_rate": 5.3838678328474244e-05,
121
+ "loss": 1.1879,
122
+ "step": 950
123
+ },
124
+ {
125
+ "epoch": 1.46,
126
+ "learning_rate": 5.1409135082604474e-05,
127
+ "loss": 1.188,
128
+ "step": 1000
129
+ },
130
+ {
131
+ "epoch": 1.53,
132
+ "learning_rate": 4.89795918367347e-05,
133
+ "loss": 1.1836,
134
+ "step": 1050
135
+ },
136
+ {
137
+ "epoch": 1.6,
138
+ "learning_rate": 4.655004859086492e-05,
139
+ "loss": 1.1794,
140
+ "step": 1100
141
+ },
142
+ {
143
+ "epoch": 1.67,
144
+ "learning_rate": 4.412050534499514e-05,
145
+ "loss": 1.1815,
146
+ "step": 1150
147
+ },
148
+ {
149
+ "epoch": 1.75,
150
+ "learning_rate": 4.1690962099125366e-05,
151
+ "loss": 1.1856,
152
+ "step": 1200
153
+ },
154
+ {
155
+ "epoch": 1.82,
156
+ "learning_rate": 3.926141885325559e-05,
157
+ "loss": 1.1747,
158
+ "step": 1250
159
+ },
160
+ {
161
+ "epoch": 1.89,
162
+ "learning_rate": 3.683187560738581e-05,
163
+ "loss": 1.1736,
164
+ "step": 1300
165
+ },
166
+ {
167
+ "epoch": 1.97,
168
+ "learning_rate": 3.4402332361516035e-05,
169
+ "loss": 1.1899,
170
+ "step": 1350
171
+ },
172
+ {
173
+ "epoch": 2.04,
174
+ "learning_rate": 3.1972789115646265e-05,
175
+ "loss": 1.1726,
176
+ "step": 1400
177
+ },
178
+ {
179
+ "epoch": 2.11,
180
+ "learning_rate": 2.954324586977648e-05,
181
+ "loss": 1.1544,
182
+ "step": 1450
183
+ },
184
+ {
185
+ "epoch": 2.18,
186
+ "learning_rate": 2.7113702623906705e-05,
187
+ "loss": 1.1568,
188
+ "step": 1500
189
+ },
190
+ {
191
+ "epoch": 2.26,
192
+ "learning_rate": 2.468415937803693e-05,
193
+ "loss": 1.1684,
194
+ "step": 1550
195
+ },
196
+ {
197
+ "epoch": 2.33,
198
+ "learning_rate": 2.225461613216715e-05,
199
+ "loss": 1.1629,
200
+ "step": 1600
201
+ },
202
+ {
203
+ "epoch": 2.4,
204
+ "learning_rate": 1.9825072886297377e-05,
205
+ "loss": 1.1558,
206
+ "step": 1650
207
+ },
208
+ {
209
+ "epoch": 2.48,
210
+ "learning_rate": 1.73955296404276e-05,
211
+ "loss": 1.1608,
212
+ "step": 1700
213
+ },
214
+ {
215
+ "epoch": 2.55,
216
+ "learning_rate": 1.4965986394557824e-05,
217
+ "loss": 1.155,
218
+ "step": 1750
219
+ },
220
+ {
221
+ "epoch": 2.62,
222
+ "learning_rate": 1.2536443148688048e-05,
223
+ "loss": 1.153,
224
+ "step": 1800
225
+ },
226
+ {
227
+ "epoch": 2.69,
228
+ "learning_rate": 1.010689990281827e-05,
229
+ "loss": 1.1597,
230
+ "step": 1850
231
+ },
232
+ {
233
+ "epoch": 2.77,
234
+ "learning_rate": 7.677356656948493e-06,
235
+ "loss": 1.1517,
236
+ "step": 1900
237
+ },
238
+ {
239
+ "epoch": 2.84,
240
+ "learning_rate": 5.247813411078718e-06,
241
+ "loss": 1.1488,
242
+ "step": 1950
243
+ },
244
+ {
245
+ "epoch": 2.91,
246
+ "learning_rate": 2.818270165208941e-06,
247
+ "loss": 1.1602,
248
+ "step": 2000
249
  }
250
  ],
251
+ "max_steps": 2058,
252
+ "num_train_epochs": 3,
253
+ "total_flos": 7.611953410899763e+17,
254
  "trial_name": null,
255
  "trial_params": null
256
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:408f2578543c0e9b80661c350bda96d3da931d8ccaff1f76b65bfbf5678d3ca0
3
  size 3899
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13903edbd77125fb4397a5222ac19b908a79246400ab53eba39aed9cc173f9aa
3
  size 3899