angelaupc commited on
Commit
08399bc
1 Parent(s): 59d2463

Meta-Llama-3-8B-Instruct-rm-Anthropic-hh-rlhf-concateye

Browse files
Files changed (3) hide show
  1. README.md +30 -21
  2. adapter_model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.6613
22
- - Accuracy: 0.6102
23
 
24
  ## Model description
25
 
@@ -38,8 +38,8 @@ More information needed
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
- - learning_rate: 1e-05
42
- - train_batch_size: 8
43
  - eval_batch_size: 8
44
  - seed: 42
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -51,23 +51,32 @@ The following hyperparameters were used during training:
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
53
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
54
- | 0.7033 | 0.11 | 64 | 0.6980 | 0.5019 |
55
- | 0.726 | 0.22 | 128 | 0.6890 | 0.5442 |
56
- | 0.6971 | 0.34 | 192 | 0.6835 | 0.5430 |
57
- | 0.7056 | 0.45 | 256 | 0.6792 | 0.5492 |
58
- | 0.7015 | 0.56 | 320 | 0.6744 | 0.5729 |
59
- | 0.6752 | 0.67 | 384 | 0.6704 | 0.5841 |
60
- | 0.69 | 0.79 | 448 | 0.6661 | 0.6052 |
61
- | 0.7041 | 0.9 | 512 | 0.6651 | 0.6065 |
62
- | 0.6855 | 1.01 | 576 | 0.6642 | 0.6102 |
63
- | 0.6825 | 1.12 | 640 | 0.6631 | 0.6152 |
64
- | 0.6907 | 1.24 | 704 | 0.6626 | 0.6152 |
65
- | 0.6601 | 1.35 | 768 | 0.6619 | 0.6115 |
66
- | 0.6887 | 1.46 | 832 | 0.6618 | 0.6127 |
67
- | 0.6976 | 1.57 | 896 | 0.6612 | 0.6090 |
68
- | 0.6782 | 1.69 | 960 | 0.6614 | 0.6152 |
69
- | 0.6653 | 1.8 | 1024 | 0.6616 | 0.6139 |
70
- | 0.6618 | 1.91 | 1088 | 0.6613 | 0.6102 |
 
 
 
 
 
 
 
 
 
71
 
72
 
73
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.7036
22
+ - Accuracy: 0.5236
23
 
24
  ## Model description
25
 
 
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
+ - learning_rate: 5e-05
42
+ - train_batch_size: 16
43
  - eval_batch_size: 8
44
  - seed: 42
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
53
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
54
+ | 0.7293 | 0.08 | 128 | 0.7252 | 0.4850 |
55
+ | 0.7412 | 0.15 | 256 | 0.6925 | 0.5386 |
56
+ | 0.7182 | 0.23 | 384 | 0.6954 | 0.5327 |
57
+ | 0.6997 | 0.3 | 512 | 0.6941 | 0.5277 |
58
+ | 0.7547 | 0.38 | 640 | 0.6959 | 0.5279 |
59
+ | 0.7123 | 0.45 | 768 | 0.6993 | 0.5252 |
60
+ | 0.7281 | 0.53 | 896 | 0.6962 | 0.5275 |
61
+ | 0.7169 | 0.6 | 1024 | 0.6986 | 0.5156 |
62
+ | 0.7244 | 0.68 | 1152 | 0.6981 | 0.5125 |
63
+ | 0.7199 | 0.75 | 1280 | 0.7000 | 0.5060 |
64
+ | 0.7311 | 0.83 | 1408 | 0.6959 | 0.5140 |
65
+ | 0.7123 | 0.9 | 1536 | 0.6956 | 0.5154 |
66
+ | 0.7344 | 0.98 | 1664 | 0.6970 | 0.5100 |
67
+ | 0.7105 | 1.05 | 1792 | 0.6933 | 0.5219 |
68
+ | 0.6947 | 1.13 | 1920 | 0.6944 | 0.5259 |
69
+ | 0.7261 | 1.21 | 2048 | 0.6960 | 0.5256 |
70
+ | 0.6997 | 1.28 | 2176 | 0.6974 | 0.5188 |
71
+ | 0.7442 | 1.36 | 2304 | 0.6960 | 0.5163 |
72
+ | 0.7004 | 1.43 | 2432 | 0.6987 | 0.5286 |
73
+ | 0.7089 | 1.51 | 2560 | 0.6982 | 0.5288 |
74
+ | 0.7142 | 1.58 | 2688 | 0.7014 | 0.5154 |
75
+ | 0.7364 | 1.66 | 2816 | 0.6997 | 0.5202 |
76
+ | 0.6915 | 1.73 | 2944 | 0.7043 | 0.5200 |
77
+ | 0.7322 | 1.81 | 3072 | 0.7037 | 0.5229 |
78
+ | 0.7524 | 1.88 | 3200 | 0.7019 | 0.5219 |
79
+ | 0.7192 | 1.96 | 3328 | 0.7036 | 0.5236 |
80
 
81
 
82
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7801c8ce75a629d5a104eb5897a203c0dad1582ffd3513403103ecbd32999684
3
  size 2115093984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb0f4897fd7ba92e6d8a5f8984044d163b7b984c0be96e5e9b1bb16361d8b298
3
  size 2115093984
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ca74357d99bc990aac57594da9a66a7bd708b7907367cd7f3a67374b0002f262
3
  size 4664
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:48428abd1317f678f1e7d88f80b48701fd06496cb790b85bee6ce943ebedbdaa
3
  size 4664