Angelectronic commited on
Commit
1b3c5fc
1 Parent(s): 60cab27

End of training

Browse files
Files changed (1) hide show
  1. README.md +48 -1
README.md CHANGED
@@ -2,9 +2,9 @@
2
  license: apache-2.0
3
  library_name: peft
4
  tags:
5
- - unsloth
6
  - trl
7
  - sft
 
8
  - generated_from_trainer
9
  base_model: unsloth/llama-3-8b-Instruct-bnb-4bit
10
  model-index:
@@ -18,6 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
18
  # llama3-chat_50000_500
19
 
20
  This model is a fine-tuned version of [unsloth/llama-3-8b-Instruct-bnb-4bit](https://huggingface.co/unsloth/llama-3-8b-Instruct-bnb-4bit) on an unknown dataset.
 
 
21
 
22
  ## Model description
23
 
@@ -47,6 +49,51 @@ The following hyperparameters were used during training:
47
  - lr_scheduler_warmup_steps: 5
48
  - num_epochs: 5
49
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
50
  ### Framework versions
51
 
52
  - PEFT 0.10.0
 
2
  license: apache-2.0
3
  library_name: peft
4
  tags:
 
5
  - trl
6
  - sft
7
+ - unsloth
8
  - generated_from_trainer
9
  base_model: unsloth/llama-3-8b-Instruct-bnb-4bit
10
  model-index:
 
18
  # llama3-chat_50000_500
19
 
20
  This model is a fine-tuned version of [unsloth/llama-3-8b-Instruct-bnb-4bit](https://huggingface.co/unsloth/llama-3-8b-Instruct-bnb-4bit) on an unknown dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 1.7378
23
 
24
  ## Model description
25
 
 
49
  - lr_scheduler_warmup_steps: 5
50
  - num_epochs: 5
51
 
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss |
55
+ |:-------------:|:------:|:----:|:---------------:|
56
+ | 1.7748 | 0.128 | 100 | 1.4928 |
57
+ | 1.5097 | 0.256 | 200 | 1.4743 |
58
+ | 1.5122 | 0.384 | 300 | 1.4623 |
59
+ | 1.4924 | 0.512 | 400 | 1.4609 |
60
+ | 1.485 | 0.64 | 500 | 1.4526 |
61
+ | 1.4779 | 0.768 | 600 | 1.4511 |
62
+ | 1.4728 | 0.896 | 700 | 1.4446 |
63
+ | 1.4476 | 1.024 | 800 | 1.4512 |
64
+ | 1.3725 | 1.152 | 900 | 1.4558 |
65
+ | 1.3747 | 1.28 | 1000 | 1.4560 |
66
+ | 1.3735 | 1.408 | 1100 | 1.4548 |
67
+ | 1.3717 | 1.536 | 1200 | 1.4499 |
68
+ | 1.3694 | 1.6640 | 1300 | 1.4526 |
69
+ | 1.3698 | 1.792 | 1400 | 1.4542 |
70
+ | 1.3701 | 1.92 | 1500 | 1.4512 |
71
+ | 1.3004 | 2.048 | 1600 | 1.4977 |
72
+ | 1.1904 | 2.176 | 1700 | 1.5075 |
73
+ | 1.1977 | 2.304 | 1800 | 1.5041 |
74
+ | 1.1888 | 2.432 | 1900 | 1.5094 |
75
+ | 1.1885 | 2.56 | 2000 | 1.5024 |
76
+ | 1.1989 | 2.6880 | 2100 | 1.5039 |
77
+ | 1.1905 | 2.816 | 2200 | 1.5046 |
78
+ | 1.1914 | 2.944 | 2300 | 1.5077 |
79
+ | 1.0764 | 3.072 | 2400 | 1.6027 |
80
+ | 0.9757 | 3.2 | 2500 | 1.6227 |
81
+ | 0.9768 | 3.328 | 2600 | 1.6228 |
82
+ | 0.9795 | 3.456 | 2700 | 1.6225 |
83
+ | 0.9775 | 3.584 | 2800 | 1.6190 |
84
+ | 0.9781 | 3.7120 | 2900 | 1.6164 |
85
+ | 0.981 | 3.84 | 3000 | 1.6199 |
86
+ | 0.9812 | 3.968 | 3100 | 1.6254 |
87
+ | 0.8731 | 4.096 | 3200 | 1.7307 |
88
+ | 0.8376 | 4.224 | 3300 | 1.7343 |
89
+ | 0.8352 | 4.352 | 3400 | 1.7398 |
90
+ | 0.8429 | 4.48 | 3500 | 1.7357 |
91
+ | 0.8431 | 4.608 | 3600 | 1.7386 |
92
+ | 0.8383 | 4.736 | 3700 | 1.7380 |
93
+ | 0.8375 | 4.864 | 3800 | 1.7376 |
94
+ | 0.842 | 4.992 | 3900 | 1.7378 |
95
+
96
+
97
  ### Framework versions
98
 
99
  - PEFT 0.10.0