Hugging Face's logo

zera09
/
gemma_FT

PEFT
Safetensors
trl
sft
Generated from Trainer
Model card Files
xet
Community
zera09 commited on
Commit
3135187
·
verified ·
1 Parent(s): 31c7c60

End of training

Browse files
Files changed (4) hide show
  1. README.md +1 -55
  2. adapter_config.json +4 -4
  3. adapter_model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -17,8 +17,6 @@ should probably proofread and complete it, then remove this comment. -->
17
  # gemma_FT
18
 
19
  This model is a fine-tuned version of [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) on the None dataset.
20
- It achieves the following results on the evaluation set:
21
- - Loss: 2.2294
22
 
23
  ## Model description
24
 
@@ -46,63 +44,11 @@ The following hyperparameters were used during training:
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 2
49
- - training_steps: 50
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
- | Training Loss | Epoch | Step | Validation Loss |
55
- |:-------------:|:-----:|:----:|:---------------:|
56
- | 3.0284 | 0.0 | 1 | 3.5564 |
57
- | 3.3693 | 0.0 | 2 | 3.4912 |
58
- | 3.2853 | 0.0 | 3 | 3.3325 |
59
- | 3.0263 | 0.0 | 4 | 3.1522 |
60
- | 3.2309 | 0.0 | 5 | 2.9678 |
61
- | 3.0045 | 0.0 | 6 | 2.8090 |
62
- | 2.3177 | 0.0 | 7 | 2.7007 |
63
- | 2.7699 | 0.0 | 8 | 2.6393 |
64
- | 2.593 | 0.0 | 9 | 2.6119 |
65
- | 2.6395 | 0.01 | 10 | 2.5837 |
66
- | 2.2016 | 0.01 | 11 | 2.5492 |
67
- | 2.4271 | 0.01 | 12 | 2.5134 |
68
- | 2.6307 | 0.01 | 13 | 2.4833 |
69
- | 2.5509 | 0.01 | 14 | 2.4618 |
70
- | 2.4261 | 0.01 | 15 | 2.4458 |
71
- | 2.468 | 0.01 | 16 | 2.4316 |
72
- | 2.2545 | 0.01 | 17 | 2.4174 |
73
- | 2.4923 | 0.01 | 18 | 2.4027 |
74
- | 2.8034 | 0.01 | 19 | 2.3862 |
75
- | 2.4174 | 0.01 | 20 | 2.3703 |
76
- | 2.0761 | 0.01 | 21 | 2.3553 |
77
- | 2.2617 | 0.01 | 22 | 2.3421 |
78
- | 2.582 | 0.01 | 23 | 2.3314 |
79
- | 1.9628 | 0.01 | 24 | 2.3239 |
80
- | 2.4096 | 0.01 | 25 | 2.3175 |
81
- | 2.1909 | 0.01 | 26 | 2.3110 |
82
- | 2.3109 | 0.01 | 27 | 2.3045 |
83
- | 2.4591 | 0.01 | 28 | 2.2984 |
84
- | 2.2765 | 0.02 | 29 | 2.2930 |
85
- | 2.0513 | 0.02 | 30 | 2.2875 |
86
- | 2.0974 | 0.02 | 31 | 2.2813 |
87
- | 2.5818 | 0.02 | 32 | 2.2755 |
88
- | 2.3872 | 0.02 | 33 | 2.2701 |
89
- | 2.1044 | 0.02 | 34 | 2.2654 |
90
- | 2.322 | 0.02 | 35 | 2.2611 |
91
- | 2.1259 | 0.02 | 36 | 2.2573 |
92
- | 2.3434 | 0.02 | 37 | 2.2538 |
93
- | 2.2724 | 0.02 | 38 | 2.2506 |
94
- | 2.3195 | 0.02 | 39 | 2.2476 |
95
- | 2.458 | 0.02 | 40 | 2.2448 |
96
- | 2.2682 | 0.02 | 41 | 2.2423 |
97
- | 2.0698 | 0.02 | 42 | 2.2397 |
98
- | 1.803 | 0.02 | 43 | 2.2374 |
99
- | 2.1955 | 0.02 | 44 | 2.2354 |
100
- | 1.9671 | 0.02 | 45 | 2.2336 |
101
- | 1.6402 | 0.02 | 46 | 2.2322 |
102
- | 1.9659 | 0.02 | 47 | 2.2312 |
103
- | 2.4895 | 0.03 | 48 | 2.2303 |
104
- | 2.2568 | 0.03 | 49 | 2.2297 |
105
- | 1.9725 | 0.03 | 50 | 2.2294 |
106
 
107
 
108
  ### Framework versions
 
17
  # gemma_FT
18
 
19
  This model is a fine-tuned version of [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) on the None dataset.
 
 
20
 
21
  ## Model description
22
 
 
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: linear
46
  - lr_scheduler_warmup_steps: 2
47
+ - training_steps: 12
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
51
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
52
 
53
 
54
  ### Framework versions
adapter_config.json CHANGED
@@ -19,13 +19,13 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
- "k_proj",
23
- "up_proj",
24
- "down_proj",
25
  "o_proj",
26
  "q_proj",
 
27
  "gate_proj",
28
- "v_proj"
 
 
29
  ],
30
  "task_type": "CAUSAL_LM",
31
  "use_rslora": false
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
 
 
 
22
  "o_proj",
23
  "q_proj",
24
+ "k_proj",
25
  "gate_proj",
26
+ "up_proj",
27
+ "v_proj",
28
+ "down_proj"
29
  ],
30
  "task_type": "CAUSAL_LM",
31
  "use_rslora": false
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:283ab3b7e827118ecf0a6e4fa02254b3e1827acc7c141997e462579ed9af7a2b
3
  size 29450584
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e8edcfe2a8de2682dacf0a9f10557569be7affe05ac359928a869b5a3063cf3
3
  size 29450584
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a758d8ba25c59ca7a5fd2936e4ad4bbb79770f68f89d741ace17befcba871356
3
  size 4856
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63d52faaacf58905f69df60603490dddb079e07179df233f9e34b467a3b80126
3
  size 4856