sri-lasya commited on
Commit
5894e44
1 Parent(s): e3b7c04

sri-lasya/gst-taxing-llm

Browse files
README.md CHANGED
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 1.7859
21
 
22
  ## Model description
23
 
@@ -43,37 +43,22 @@ The following hyperparameters were used during training:
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: constant
45
  - lr_scheduler_warmup_steps: 0.03
46
- - training_steps: 250
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
- | 3.1512 | 0.04 | 10 | 2.7003 |
53
- | 2.5126 | 0.08 | 20 | 2.1770 |
54
- | 1.696 | 0.12 | 30 | 2.0002 |
55
- | 1.7231 | 0.16 | 40 | 1.9348 |
56
- | 1.7336 | 0.2 | 50 | 1.9126 |
57
- | 1.4891 | 0.24 | 60 | 1.8458 |
58
- | 1.7379 | 0.28 | 70 | 1.7978 |
59
- | 1.6231 | 0.32 | 80 | 1.7494 |
60
- | 1.6705 | 0.36 | 90 | 1.7712 |
61
- | 1.5771 | 0.4 | 100 | 1.7794 |
62
- | 1.7315 | 0.44 | 110 | 1.7449 |
63
- | 1.2828 | 0.48 | 120 | 1.7161 |
64
- | 1.5608 | 0.52 | 130 | 1.6869 |
65
- | 1.5608 | 0.56 | 140 | 1.6915 |
66
- | 1.4046 | 0.6 | 150 | 1.6782 |
67
- | 1.3853 | 0.64 | 160 | 1.6715 |
68
- | 1.4398 | 0.68 | 170 | 1.6472 |
69
- | 1.5917 | 0.72 | 180 | 1.6333 |
70
- | 1.7207 | 0.76 | 190 | 1.6295 |
71
- | 1.5135 | 0.8 | 200 | 1.6901 |
72
- | 1.3951 | 0.84 | 210 | 1.7478 |
73
- | 1.6068 | 0.88 | 220 | 1.7256 |
74
- | 1.7094 | 0.92 | 230 | 1.7206 |
75
- | 1.8219 | 0.96 | 240 | 1.7528 |
76
- | 1.7888 | 1.0 | 250 | 1.7859 |
77
 
78
 
79
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 1.7777
21
 
22
  ## Model description
23
 
 
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: constant
45
  - lr_scheduler_warmup_steps: 0.03
46
+ - training_steps: 100
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
+ | 3.1448 | 0.1 | 10 | 2.6930 |
53
+ | 2.5114 | 0.2 | 20 | 2.1804 |
54
+ | 1.6951 | 0.3 | 30 | 2.0022 |
55
+ | 1.7202 | 0.4 | 40 | 1.9292 |
56
+ | 1.7303 | 0.5 | 50 | 1.9083 |
57
+ | 1.4861 | 0.6 | 60 | 1.8414 |
58
+ | 1.7349 | 0.7 | 70 | 1.7941 |
59
+ | 1.6208 | 0.8 | 80 | 1.7469 |
60
+ | 1.6669 | 0.9 | 90 | 1.7685 |
61
+ | 1.5728 | 1.0 | 100 | 1.7777 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
62
 
63
 
64
  ### Framework versions
adapter_config.json CHANGED
@@ -21,11 +21,11 @@
21
  "revision": null,
22
  "target_modules": [
23
  "k_proj",
24
- "down_proj",
25
- "gate_proj",
26
  "v_proj",
27
  "q_proj",
28
- "up_proj",
 
29
  "o_proj",
30
  "lm_head"
31
  ],
 
21
  "revision": null,
22
  "target_modules": [
23
  "k_proj",
24
+ "up_proj",
 
25
  "v_proj",
26
  "q_proj",
27
+ "down_proj",
28
+ "gate_proj",
29
  "o_proj",
30
  "lm_head"
31
  ],
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c1a922bdd50f194559d95b2dc6dbc3dea965dde7ddce5072bc9e336d3773945b
3
  size 1217458040
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:882aa51c16706785857813f589e45856756f4e694354b248a3bbd8616a6d5a7d
3
  size 1217458040
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5690a186ae8a6f85c974bd1f2517acfc089f339519fa9acbe05f884bb56f47c6
3
  size 5048
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c4bcac94e7d8a069e668bc9663b1ba2feffdb9f1ba69641a598006ebedc3773
3
  size 5048