Tverous commited on
Commit
2477838
1 Parent(s): 8fe50d4

Tverous/sft-trl-claim-128-llama2-13b-chat-hf

Browse files
Files changed (2) hide show
  1. README.md +242 -0
  2. adapter_model.bin +1 -1
README.md ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: meta-llama/Llama-2-13b-chat-hf
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - anli
7
+ model-index:
8
+ - name: sft-trl-claim-128-llama2-13b-chat-hf
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # sft-trl-claim-128-llama2-13b-chat-hf
16
+
17
+ This model is a fine-tuned version of [meta-llama/Llama-2-13b-chat-hf](https://huggingface.co/meta-llama/Llama-2-13b-chat-hf) on the anli dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.4327
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 5e-05
39
+ - train_batch_size: 8
40
+ - eval_batch_size: 8
41
+ - seed: 42
42
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
+ - lr_scheduler_type: linear
44
+ - lr_scheduler_warmup_steps: 500
45
+ - num_epochs: 3.0
46
+
47
+ ### Training results
48
+
49
+ | Training Loss | Epoch | Step | Validation Loss |
50
+ |:-------------:|:-----:|:------:|:---------------:|
51
+ | 1.2823 | 0.02 | 1000 | 1.5153 |
52
+ | 1.1442 | 0.03 | 2000 | 1.4907 |
53
+ | 1.1578 | 0.05 | 3000 | 1.4674 |
54
+ | 1.1452 | 0.06 | 4000 | 1.4707 |
55
+ | 1.175 | 0.08 | 5000 | 1.4579 |
56
+ | 1.1516 | 0.1 | 6000 | 1.4382 |
57
+ | 1.2068 | 0.11 | 7000 | 1.4134 |
58
+ | 1.2737 | 0.13 | 8000 | 1.3936 |
59
+ | 1.3977 | 0.15 | 9000 | 1.4009 |
60
+ | 1.0338 | 0.16 | 10000 | 1.3899 |
61
+ | 1.0997 | 0.18 | 11000 | 1.3971 |
62
+ | 1.1361 | 0.19 | 12000 | 1.3488 |
63
+ | 1.012 | 0.21 | 13000 | 1.3628 |
64
+ | 1.0694 | 0.23 | 14000 | 1.3532 |
65
+ | 1.067 | 0.24 | 15000 | 1.3531 |
66
+ | 1.2775 | 0.26 | 16000 | 1.3492 |
67
+ | 1.1953 | 0.28 | 17000 | 1.3598 |
68
+ | 1.0807 | 0.29 | 18000 | 1.3294 |
69
+ | 1.2402 | 0.31 | 19000 | 1.2770 |
70
+ | 1.1454 | 0.32 | 20000 | 1.2803 |
71
+ | 1.0369 | 0.34 | 21000 | 1.2688 |
72
+ | 0.9862 | 0.36 | 22000 | 1.2651 |
73
+ | 1.151 | 0.37 | 23000 | 1.2371 |
74
+ | 1.1605 | 0.39 | 24000 | 1.2199 |
75
+ | 1.1367 | 0.41 | 25000 | 1.2268 |
76
+ | 1.011 | 0.42 | 26000 | 1.2123 |
77
+ | 1.1792 | 0.44 | 27000 | 1.1763 |
78
+ | 0.8839 | 0.45 | 28000 | 1.1784 |
79
+ | 1.166 | 0.47 | 29000 | 1.1999 |
80
+ | 1.2947 | 0.49 | 30000 | 1.2064 |
81
+ | 0.7324 | 0.5 | 31000 | 1.2174 |
82
+ | 0.9919 | 0.52 | 32000 | 1.1837 |
83
+ | 1.3371 | 0.54 | 33000 | 1.1761 |
84
+ | 0.9874 | 0.55 | 34000 | 1.1966 |
85
+ | 1.0411 | 0.57 | 35000 | 1.1885 |
86
+ | 1.1669 | 0.58 | 36000 | 1.1789 |
87
+ | 1.231 | 0.6 | 37000 | 1.1616 |
88
+ | 0.9799 | 0.62 | 38000 | 1.1003 |
89
+ | 1.0801 | 0.63 | 39000 | 1.0844 |
90
+ | 1.0984 | 0.65 | 40000 | 1.1037 |
91
+ | 1.1046 | 0.67 | 41000 | 1.0683 |
92
+ | 1.1537 | 0.68 | 42000 | 1.0569 |
93
+ | 1.0744 | 0.7 | 43000 | 1.0661 |
94
+ | 1.1451 | 0.71 | 44000 | 1.0460 |
95
+ | 0.9822 | 0.73 | 45000 | 1.0559 |
96
+ | 1.0439 | 0.75 | 46000 | 1.0473 |
97
+ | 0.8704 | 0.76 | 47000 | 1.0721 |
98
+ | 0.9934 | 0.78 | 48000 | 0.9789 |
99
+ | 1.134 | 0.79 | 49000 | 0.9800 |
100
+ | 1.0722 | 0.81 | 50000 | 1.0122 |
101
+ | 1.0894 | 0.83 | 51000 | 0.9872 |
102
+ | 0.822 | 0.84 | 52000 | 1.0067 |
103
+ | 1.0764 | 0.86 | 53000 | 0.9941 |
104
+ | 1.1809 | 0.88 | 54000 | 0.9966 |
105
+ | 1.0228 | 0.89 | 55000 | 0.9657 |
106
+ | 0.9944 | 0.91 | 56000 | 0.9778 |
107
+ | 0.7453 | 0.92 | 57000 | 0.9442 |
108
+ | 1.1951 | 0.94 | 58000 | 0.9621 |
109
+ | 0.7633 | 0.96 | 59000 | 0.9271 |
110
+ | 0.5546 | 0.97 | 60000 | 0.9493 |
111
+ | 1.2136 | 0.99 | 61000 | 0.9211 |
112
+ | 0.7377 | 1.01 | 62000 | 0.8821 |
113
+ | 0.8892 | 1.02 | 63000 | 0.8635 |
114
+ | 0.9491 | 1.04 | 64000 | 0.8695 |
115
+ | 0.9627 | 1.05 | 65000 | 0.8808 |
116
+ | 0.7405 | 1.07 | 66000 | 0.8450 |
117
+ | 1.1451 | 1.09 | 67000 | 0.8496 |
118
+ | 0.9935 | 1.1 | 68000 | 0.8708 |
119
+ | 1.2591 | 1.12 | 69000 | 0.8246 |
120
+ | 1.1467 | 1.14 | 70000 | 0.8376 |
121
+ | 0.8296 | 1.15 | 71000 | 0.8411 |
122
+ | 0.9733 | 1.17 | 72000 | 0.8356 |
123
+ | 1.1116 | 1.18 | 73000 | 0.8564 |
124
+ | 0.9909 | 1.2 | 74000 | 0.8420 |
125
+ | 1.0602 | 1.22 | 75000 | 0.8398 |
126
+ | 1.0284 | 1.23 | 76000 | 0.8429 |
127
+ | 0.9611 | 1.25 | 77000 | 0.8288 |
128
+ | 0.9866 | 1.27 | 78000 | 0.8432 |
129
+ | 0.6751 | 1.28 | 79000 | 0.8109 |
130
+ | 0.9637 | 1.3 | 80000 | 0.8039 |
131
+ | 1.2506 | 1.31 | 81000 | 0.8088 |
132
+ | 1.1821 | 1.33 | 82000 | 0.8080 |
133
+ | 0.9813 | 1.35 | 83000 | 0.8003 |
134
+ | 1.175 | 1.36 | 84000 | 0.7962 |
135
+ | 0.8377 | 1.38 | 85000 | 0.7913 |
136
+ | 1.114 | 1.4 | 86000 | 0.7977 |
137
+ | 0.9089 | 1.41 | 87000 | 0.7779 |
138
+ | 0.9896 | 1.43 | 88000 | 0.7665 |
139
+ | 0.7499 | 1.44 | 89000 | 0.7693 |
140
+ | 1.0132 | 1.46 | 90000 | 0.7420 |
141
+ | 0.6964 | 1.48 | 91000 | 0.7405 |
142
+ | 0.9243 | 1.49 | 92000 | 0.7396 |
143
+ | 0.8555 | 1.51 | 93000 | 0.7448 |
144
+ | 0.9978 | 1.52 | 94000 | 0.7449 |
145
+ | 1.2293 | 1.54 | 95000 | 0.7324 |
146
+ | 0.8886 | 1.56 | 96000 | 0.7519 |
147
+ | 0.8325 | 1.57 | 97000 | 0.7625 |
148
+ | 0.5212 | 1.59 | 98000 | 0.7554 |
149
+ | 0.8564 | 1.61 | 99000 | 0.7211 |
150
+ | 0.7141 | 1.62 | 100000 | 0.7227 |
151
+ | 0.5702 | 1.64 | 101000 | 0.6725 |
152
+ | 1.0332 | 1.65 | 102000 | 0.6611 |
153
+ | 1.0074 | 1.67 | 103000 | 0.6633 |
154
+ | 0.8829 | 1.69 | 104000 | 0.6682 |
155
+ | 0.5767 | 1.7 | 105000 | 0.6641 |
156
+ | 0.6375 | 1.72 | 106000 | 0.6652 |
157
+ | 0.7965 | 1.74 | 107000 | 0.6728 |
158
+ | 1.1104 | 1.75 | 108000 | 0.6564 |
159
+ | 0.5071 | 1.77 | 109000 | 0.6273 |
160
+ | 0.9045 | 1.78 | 110000 | 0.6298 |
161
+ | 0.7056 | 1.8 | 111000 | 0.6223 |
162
+ | 1.013 | 1.82 | 112000 | 0.6290 |
163
+ | 0.943 | 1.83 | 113000 | 0.6281 |
164
+ | 0.6309 | 1.85 | 114000 | 0.6231 |
165
+ | 0.7388 | 1.87 | 115000 | 0.6069 |
166
+ | 0.5067 | 1.88 | 116000 | 0.5786 |
167
+ | 1.144 | 1.9 | 117000 | 0.5723 |
168
+ | 1.1245 | 1.91 | 118000 | 0.5761 |
169
+ | 0.9093 | 1.93 | 119000 | 0.5832 |
170
+ | 1.1755 | 1.95 | 120000 | 0.5798 |
171
+ | 0.4415 | 1.96 | 121000 | 0.5789 |
172
+ | 0.8623 | 1.98 | 122000 | 0.5793 |
173
+ | 0.7605 | 2.0 | 123000 | 0.5529 |
174
+ | 0.9743 | 2.01 | 124000 | 0.5542 |
175
+ | 1.0476 | 2.03 | 125000 | 0.5585 |
176
+ | 1.0112 | 2.04 | 126000 | 0.5673 |
177
+ | 0.9344 | 2.06 | 127000 | 0.5644 |
178
+ | 0.976 | 2.08 | 128000 | 0.5771 |
179
+ | 1.0471 | 2.09 | 129000 | 0.5615 |
180
+ | 0.626 | 2.11 | 130000 | 0.5597 |
181
+ | 0.6866 | 2.13 | 131000 | 0.5520 |
182
+ | 0.7723 | 2.14 | 132000 | 0.5416 |
183
+ | 0.7605 | 2.16 | 133000 | 0.5407 |
184
+ | 0.8413 | 2.17 | 134000 | 0.5452 |
185
+ | 1.1015 | 2.19 | 135000 | 0.5506 |
186
+ | 0.7203 | 2.21 | 136000 | 0.5470 |
187
+ | 0.7008 | 2.22 | 137000 | 0.5535 |
188
+ | 1.035 | 2.24 | 138000 | 0.5404 |
189
+ | 0.8432 | 2.25 | 139000 | 0.5459 |
190
+ | 0.7886 | 2.27 | 140000 | 0.5403 |
191
+ | 1.1197 | 2.29 | 141000 | 0.5533 |
192
+ | 0.8474 | 2.3 | 142000 | 0.5237 |
193
+ | 0.8785 | 2.32 | 143000 | 0.5325 |
194
+ | 1.1119 | 2.34 | 144000 | 0.5105 |
195
+ | 1.2089 | 2.35 | 145000 | 0.5133 |
196
+ | 0.8626 | 2.37 | 146000 | 0.5097 |
197
+ | 1.106 | 2.38 | 147000 | 0.5120 |
198
+ | 0.9681 | 2.4 | 148000 | 0.5116 |
199
+ | 1.0139 | 2.42 | 149000 | 0.5102 |
200
+ | 0.6389 | 2.43 | 150000 | 0.5152 |
201
+ | 0.86 | 2.45 | 151000 | 0.5141 |
202
+ | 1.041 | 2.47 | 152000 | 0.5102 |
203
+ | 0.984 | 2.48 | 153000 | 0.4967 |
204
+ | 0.7444 | 2.5 | 154000 | 0.4792 |
205
+ | 0.6543 | 2.51 | 155000 | 0.4693 |
206
+ | 0.6596 | 2.53 | 156000 | 0.4511 |
207
+ | 1.2077 | 2.55 | 157000 | 0.4506 |
208
+ | 0.9068 | 2.56 | 158000 | 0.4580 |
209
+ | 0.9328 | 2.58 | 159000 | 0.4602 |
210
+ | 0.8111 | 2.6 | 160000 | 0.4607 |
211
+ | 0.8047 | 2.61 | 161000 | 0.4475 |
212
+ | 0.7448 | 2.63 | 162000 | 0.4532 |
213
+ | 1.0685 | 2.64 | 163000 | 0.4628 |
214
+ | 0.9309 | 2.66 | 164000 | 0.4569 |
215
+ | 0.8024 | 2.68 | 165000 | 0.4571 |
216
+ | 0.9119 | 2.69 | 166000 | 0.4572 |
217
+ | 0.5736 | 2.71 | 167000 | 0.4567 |
218
+ | 0.7446 | 2.73 | 168000 | 0.4481 |
219
+ | 1.1721 | 2.74 | 169000 | 0.4449 |
220
+ | 0.7552 | 2.76 | 170000 | 0.4429 |
221
+ | 0.7927 | 2.77 | 171000 | 0.4358 |
222
+ | 1.2197 | 2.79 | 172000 | 0.4301 |
223
+ | 0.771 | 2.81 | 173000 | 0.4388 |
224
+ | 0.7656 | 2.82 | 174000 | 0.4385 |
225
+ | 1.0673 | 2.84 | 175000 | 0.4432 |
226
+ | 1.0228 | 2.86 | 176000 | 0.4361 |
227
+ | 0.7585 | 2.87 | 177000 | 0.4380 |
228
+ | 0.63 | 2.89 | 178000 | 0.4371 |
229
+ | 1.0722 | 2.9 | 179000 | 0.4314 |
230
+ | 0.7093 | 2.92 | 180000 | 0.4337 |
231
+ | 0.468 | 2.94 | 181000 | 0.4341 |
232
+ | 0.7841 | 2.95 | 182000 | 0.4313 |
233
+ | 0.779 | 2.97 | 183000 | 0.4312 |
234
+ | 0.4624 | 2.98 | 184000 | 0.4327 |
235
+
236
+
237
+ ### Framework versions
238
+
239
+ - Transformers 4.33.0.dev0
240
+ - Pytorch 2.0.1+cu117
241
+ - Datasets 2.13.1
242
+ - Tokenizers 0.13.3
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f14181c63d46c69c820e3dff0db84a99c1eb7bd6ff93b11ef4eec627197d3833
3
  size 52486477
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85d1de092d2269e32cd84cdccc2c9f85db8aa766596cabaa6d52e3591c91e7b9
3
  size 52486477