Hugging Face's logo

zelalt
/
paper_imp_new_format_1

PEFT
Safetensors
trl
sft
Generated from Trainer
Model card Files
xet
 Community
zelalt commited on
Commit
ddd91e5
·
verified ·
1 Parent(s): d93accf

End of training

Browse files
Files changed (4) hide show
  1. README.md +69 -14
  2. adapter_config.json +2 -2
  3. adapter_model.safetensors +1 -1
  4. training_args.bin +2 -2
README.md CHANGED
@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [teknium/OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.5958
22
 
23
  ## Model description
24
 
@@ -44,24 +44,79 @@ The following hyperparameters were used during training:
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: reduce_lr_on_plateau
46
  - lr_scheduler_warmup_steps: 200
47
- - num_epochs: 4
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss |
52
  |:-------------:|:-----:|:----:|:---------------:|
53
- | 0.9184 | 0.33 | 9 | 0.9121 |
54
- | 0.8543 | 0.67 | 18 | 0.7775 |
55
- | 0.712 | 1.0 | 27 | 0.6491 |
56
- | 0.6182 | 1.33 | 36 | 0.6139 |
57
- | 0.5267 | 1.67 | 45 | 0.5951 |
58
- | 0.5141 | 2.0 | 54 | 0.5840 |
59
- | 0.454 | 2.33 | 63 | 0.5961 |
60
- | 0.4618 | 2.67 | 72 | 0.5944 |
61
- | 0.4989 | 3.0 | 81 | 0.5944 |
62
- | 0.3703 | 3.33 | 90 | 0.6079 |
63
- | 0.4162 | 3.67 | 99 | 0.5984 |
64
- | 0.4135 | 4.0 | 108 | 0.5958 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
65
 
66
 
67
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [teknium/OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.6201
22
 
23
  ## Model description
24
 
 
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: reduce_lr_on_plateau
46
  - lr_scheduler_warmup_steps: 200
47
+ - num_epochs: 10
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss |
52
  |:-------------:|:-----:|:----:|:---------------:|
53
+ | 0.99 | 0.15 | 4 | 1.0009 |
54
+ | 1.0287 | 0.3 | 8 | 0.9440 |
55
+ | 0.9432 | 0.44 | 12 | 0.8927 |
56
+ | 0.829 | 0.59 | 16 | 0.8395 |
57
+ | 0.7702 | 0.74 | 20 | 0.7773 |
58
+ | 0.6886 | 0.89 | 24 | 0.7180 |
59
+ | 0.5844 | 1.04 | 28 | 0.6774 |
60
+ | 0.6016 | 1.19 | 32 | 0.6593 |
61
+ | 0.5237 | 1.33 | 36 | 0.6484 |
62
+ | 0.5505 | 1.48 | 40 | 0.6418 |
63
+ | 0.669 | 1.63 | 44 | 0.6350 |
64
+ | 0.4595 | 1.78 | 48 | 0.6320 |
65
+ | 0.5486 | 1.93 | 52 | 0.6290 |
66
+ | 0.539 | 2.07 | 56 | 0.6242 |
67
+ | 0.5089 | 2.22 | 60 | 0.6188 |
68
+ | 0.6012 | 2.37 | 64 | 0.6195 |
69
+ | 0.4286 | 2.52 | 68 | 0.6147 |
70
+ | 0.4335 | 2.67 | 72 | 0.6154 |
71
+ | 0.4584 | 2.81 | 76 | 0.6154 |
72
+ | 0.4597 | 2.96 | 80 | 0.6115 |
73
+ | 0.421 | 3.11 | 84 | 0.6076 |
74
+ | 0.4541 | 3.26 | 88 | 0.6131 |
75
+ | 0.3533 | 3.41 | 92 | 0.6199 |
76
+ | 0.4007 | 3.56 | 96 | 0.6142 |
77
+ | 0.3876 | 3.7 | 100 | 0.6104 |
78
+ | 0.4258 | 3.85 | 104 | 0.6073 |
79
+ | 0.4537 | 4.0 | 108 | 0.6059 |
80
+ | 0.3486 | 4.15 | 112 | 0.6098 |
81
+ | 0.3399 | 4.3 | 116 | 0.6172 |
82
+ | 0.3826 | 4.44 | 120 | 0.6215 |
83
+ | 0.3813 | 4.59 | 124 | 0.6210 |
84
+ | 0.3835 | 4.74 | 128 | 0.6192 |
85
+ | 0.3864 | 4.89 | 132 | 0.6173 |
86
+ | 0.3216 | 5.04 | 136 | 0.6147 |
87
+ | 0.3923 | 5.19 | 140 | 0.6151 |
88
+ | 0.3757 | 5.33 | 144 | 0.6158 |
89
+ | 0.2861 | 5.48 | 148 | 0.6164 |
90
+ | 0.3838 | 5.63 | 152 | 0.6174 |
91
+ | 0.3669 | 5.78 | 156 | 0.6180 |
92
+ | 0.3405 | 5.93 | 160 | 0.6188 |
93
+ | 0.3355 | 6.07 | 164 | 0.6188 |
94
+ | 0.3533 | 6.22 | 168 | 0.6193 |
95
+ | 0.4245 | 6.37 | 172 | 0.6191 |
96
+ | 0.3257 | 6.52 | 176 | 0.6196 |
97
+ | 0.3361 | 6.67 | 180 | 0.6198 |
98
+ | 0.3595 | 6.81 | 184 | 0.6198 |
99
+ | 0.3075 | 6.96 | 188 | 0.6198 |
100
+ | 0.4133 | 7.11 | 192 | 0.6199 |
101
+ | 0.3341 | 7.26 | 196 | 0.6197 |
102
+ | 0.385 | 7.41 | 200 | 0.6197 |
103
+ | 0.3254 | 7.56 | 204 | 0.6198 |
104
+ | 0.385 | 7.7 | 208 | 0.6199 |
105
+ | 0.3332 | 7.85 | 212 | 0.6197 |
106
+ | 0.2648 | 8.0 | 216 | 0.6198 |
107
+ | 0.3602 | 8.15 | 220 | 0.6203 |
108
+ | 0.4129 | 8.3 | 224 | 0.6200 |
109
+ | 0.3716 | 8.44 | 228 | 0.6196 |
110
+ | 0.3152 | 8.59 | 232 | 0.6200 |
111
+ | 0.3463 | 8.74 | 236 | 0.6199 |
112
+ | 0.3357 | 8.89 | 240 | 0.6198 |
113
+ | 0.3029 | 9.04 | 244 | 0.6199 |
114
+ | 0.3679 | 9.19 | 248 | 0.6200 |
115
+ | 0.3911 | 9.33 | 252 | 0.6201 |
116
+ | 0.2844 | 9.48 | 256 | 0.6199 |
117
+ | 0.2584 | 9.63 | 260 | 0.6199 |
118
+ | 0.3966 | 9.78 | 264 | 0.6200 |
119
+ | 0.3423 | 9.93 | 268 | 0.6201 |
120
 
121
 
122
  ### Framework versions
adapter_config.json CHANGED
@@ -19,10 +19,10 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
- "q_proj",
23
  "k_proj",
 
24
  "o_proj",
25
- "v_proj"
26
  ],
27
  "task_type": "CAUSAL_LM"
28
  }
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
 
22
  "k_proj",
23
+ "v_proj",
24
  "o_proj",
25
+ "q_proj"
26
  ],
27
  "task_type": "CAUSAL_LM"
28
  }
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:85184fa9f94b7b7dc26fe233479247c3838e968bc3cb7d6018068aceba964cc4
3
  size 13665592
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f8ec8a93dbdb23e41f74a343e2b50b9b5662ea45b7f3f3a6845e048a0ed71cd
3
  size 13665592
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:91b70e9eb938f894f98dacb706ece473da02042e963f4972526e8a1d0dbd1d25
3
- size 4792
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b0fc550998bc585bb9445b1c0cd9d2a71b69b59a80be4ae1bf0b884b839912a
3
+ size 4856