Hugging Face's logo

Stern5497
/
org_modelorg_model

PEFT
Safetensors
Generated from Trainer
Model card Files Community
Stern5497 commited on
Commit
cb3126d
•
1 Parent(s): 7bd338b

mistral-lp2-org_org_a

Browse files
Files changed (3) hide show
  1. README.md +20 -20
  2. adapter_config.json +2 -2
  3. adapter_model.safetensors +2 -2
README.md CHANGED
@@ -16,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.9110
20
- - F1 Micro: 0.8214
21
- - F1 Macro: 0.8156
22
- - F1 Weighted: 0.8234
23
 
24
  ## Model description
25
 
@@ -50,22 +50,22 @@ The following hyperparameters were used during training:
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | F1 Micro | F1 Macro | F1 Weighted |
52
  |:-------------:|:------:|:----:|:---------------:|:--------:|:--------:|:-----------:|
53
- | 1.979 | 0.0154 | 25 | 1.4043 | 0.7220 | 0.7197 | 0.7259 |
54
- | 1.3006 | 0.0308 | 50 | 1.2184 | 0.7775 | 0.7754 | 0.7807 |
55
- | 1.1099 | 0.0462 | 75 | 1.1320 | 0.8010 | 0.7970 | 0.8040 |
56
- | 1.1383 | 0.0615 | 100 | 1.0762 | 0.8039 | 0.8007 | 0.8072 |
57
- | 1.0121 | 0.0769 | 125 | 1.0230 | 0.8010 | 0.7967 | 0.8037 |
58
- | 1.0296 | 0.0923 | 150 | 0.9966 | 0.8099 | 0.8056 | 0.8130 |
59
- | 1.0485 | 0.1077 | 175 | 0.9745 | 0.8111 | 0.8063 | 0.8139 |
60
- | 0.9996 | 0.1231 | 200 | 0.9647 | 0.8030 | 0.7984 | 0.8052 |
61
- | 0.9815 | 0.1385 | 225 | 0.9490 | 0.8160 | 0.8099 | 0.8178 |
62
- | 0.9456 | 0.1538 | 250 | 0.9378 | 0.8073 | 0.8033 | 0.8099 |
63
- | 0.8896 | 0.1692 | 275 | 0.9298 | 0.8143 | 0.8091 | 0.8164 |
64
- | 0.994 | 0.1846 | 300 | 0.9239 | 0.8064 | 0.8030 | 0.8094 |
65
- | 0.8588 | 0.2 | 325 | 0.9142 | 0.8119 | 0.8079 | 0.8145 |
66
- | 0.8971 | 0.2154 | 350 | 0.9139 | 0.8216 | 0.8158 | 0.8236 |
67
- | 0.9647 | 0.2308 | 375 | 0.9133 | 0.8223 | 0.8163 | 0.8242 |
68
- | 0.9352 | 0.2462 | 400 | 0.9110 | 0.8214 | 0.8156 | 0.8234 |
69
 
70
 
71
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.8643
20
+ - F1 Micro: 0.6606
21
+ - F1 Macro: 0.6479
22
+ - F1 Weighted: 0.6611
23
 
24
  ## Model description
25
 
 
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | F1 Micro | F1 Macro | F1 Weighted |
52
  |:-------------:|:------:|:----:|:---------------:|:--------:|:--------:|:-----------:|
53
+ | 1.6117 | 0.0154 | 25 | 1.3679 | 0.5541 | 0.4980 | 0.5313 |
54
+ | 1.2903 | 0.0308 | 50 | 1.2094 | 0.6156 | 0.5779 | 0.6029 |
55
+ | 1.164 | 0.0462 | 75 | 1.0987 | 0.6206 | 0.5931 | 0.6141 |
56
+ | 1.1168 | 0.0615 | 100 | 1.1057 | 0.6376 | 0.5883 | 0.6165 |
57
+ | 1.026 | 0.0769 | 125 | 0.9896 | 0.6314 | 0.6196 | 0.6328 |
58
+ | 0.9481 | 0.0923 | 150 | 0.9619 | 0.6438 | 0.6173 | 0.6373 |
59
+ | 0.9797 | 0.1077 | 175 | 0.9549 | 0.6514 | 0.6191 | 0.6411 |
60
+ | 1.045 | 0.1231 | 200 | 0.9121 | 0.6541 | 0.6403 | 0.6543 |
61
+ | 0.8954 | 0.1385 | 225 | 0.8991 | 0.6595 | 0.6418 | 0.6576 |
62
+ | 0.9245 | 0.1538 | 250 | 0.8887 | 0.6588 | 0.6433 | 0.6580 |
63
+ | 0.8636 | 0.1692 | 275 | 0.8824 | 0.6602 | 0.6458 | 0.6600 |
64
+ | 0.846 | 0.1846 | 300 | 0.8793 | 0.6672 | 0.6451 | 0.6627 |
65
+ | 0.8885 | 0.2 | 325 | 0.8820 | 0.6696 | 0.6431 | 0.6624 |
66
+ | 0.8323 | 0.2154 | 350 | 0.8652 | 0.6618 | 0.6474 | 0.6616 |
67
+ | 0.9313 | 0.2308 | 375 | 0.8654 | 0.6601 | 0.6477 | 0.6608 |
68
+ | 0.857 | 0.2462 | 400 | 0.8643 | 0.6606 | 0.6479 | 0.6611 |
69
 
70
 
71
  ### Framework versions
adapter_config.json CHANGED
@@ -20,9 +20,9 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "k_proj",
24
- "q_proj",
25
  "v_proj",
 
 
26
  "o_proj"
27
  ],
28
  "task_type": "SEQ_CLS",
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
 
23
  "v_proj",
24
+ "q_proj",
25
+ "k_proj",
26
  "o_proj"
27
  ],
28
  "task_type": "SEQ_CLS",
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d287f3e911660a53e7f747ea43f5714abadb39a0ac9ea29328dfffd8e531d587
3
- size 578898352
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0789a908ecfc7cdc60018ed5a135a610ddf9a4edd3e2dc3aebcfb909c83c39fe
3
+ size 578881968