End of training
Browse files
README.md
ADDED
@@ -0,0 +1,100 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- unsloth
|
6 |
+
- generated_from_trainer
|
7 |
+
base_model: unsloth/llama-2-13b-bnb-4bit
|
8 |
+
model-index:
|
9 |
+
- name: llama_2_13b_Magiccoder_evol_10k_qlora_ortho
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# llama_2_13b_Magiccoder_evol_10k_qlora_ortho
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [unsloth/llama-2-13b-bnb-4bit](https://huggingface.co/unsloth/llama-2-13b-bnb-4bit) on an unknown dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 1.0950
|
21 |
+
|
22 |
+
## Model description
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Intended uses & limitations
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training and evaluation data
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training procedure
|
35 |
+
|
36 |
+
### Training hyperparameters
|
37 |
+
|
38 |
+
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 0.0001
|
40 |
+
- train_batch_size: 8
|
41 |
+
- eval_batch_size: 8
|
42 |
+
- seed: 42
|
43 |
+
- gradient_accumulation_steps: 8
|
44 |
+
- total_train_batch_size: 64
|
45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
46 |
+
- lr_scheduler_type: cosine
|
47 |
+
- lr_scheduler_warmup_steps: 0.02
|
48 |
+
- num_epochs: 1
|
49 |
+
|
50 |
+
### Training results
|
51 |
+
|
52 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
53 |
+
|:-------------:|:------:|:----:|:---------------:|
|
54 |
+
| 1.2067 | 0.0262 | 4 | 1.1823 |
|
55 |
+
| 1.1675 | 0.0523 | 8 | 1.1498 |
|
56 |
+
| 1.1004 | 0.0785 | 12 | 1.1349 |
|
57 |
+
| 1.0531 | 0.1047 | 16 | 1.1288 |
|
58 |
+
| 1.0946 | 0.1308 | 20 | 1.1246 |
|
59 |
+
| 1.0602 | 0.1570 | 24 | 1.1215 |
|
60 |
+
| 1.0636 | 0.1832 | 28 | 1.1175 |
|
61 |
+
| 1.1078 | 0.2093 | 32 | 1.1151 |
|
62 |
+
| 1.04 | 0.2355 | 36 | 1.1125 |
|
63 |
+
| 1.115 | 0.2617 | 40 | 1.1123 |
|
64 |
+
| 1.0994 | 0.2878 | 44 | 1.1102 |
|
65 |
+
| 1.1379 | 0.3140 | 48 | 1.1098 |
|
66 |
+
| 1.1145 | 0.3401 | 52 | 1.1064 |
|
67 |
+
| 1.0849 | 0.3663 | 56 | 1.1088 |
|
68 |
+
| 1.1317 | 0.3925 | 60 | 1.1087 |
|
69 |
+
| 1.134 | 0.4186 | 64 | 1.1056 |
|
70 |
+
| 1.0856 | 0.4448 | 68 | 1.1038 |
|
71 |
+
| 1.0972 | 0.4710 | 72 | 1.1004 |
|
72 |
+
| 1.044 | 0.4971 | 76 | 1.1005 |
|
73 |
+
| 1.1311 | 0.5233 | 80 | 1.1004 |
|
74 |
+
| 1.1474 | 0.5495 | 84 | 1.1002 |
|
75 |
+
| 1.0886 | 0.5756 | 88 | 1.0999 |
|
76 |
+
| 1.0372 | 0.6018 | 92 | 1.0973 |
|
77 |
+
| 1.0376 | 0.6280 | 96 | 1.0968 |
|
78 |
+
| 1.1006 | 0.6541 | 100 | 1.0965 |
|
79 |
+
| 1.09 | 0.6803 | 104 | 1.0964 |
|
80 |
+
| 1.0786 | 0.7065 | 108 | 1.0969 |
|
81 |
+
| 1.111 | 0.7326 | 112 | 1.0970 |
|
82 |
+
| 1.053 | 0.7588 | 116 | 1.0961 |
|
83 |
+
| 1.0764 | 0.7850 | 120 | 1.0948 |
|
84 |
+
| 1.0971 | 0.8111 | 124 | 1.0944 |
|
85 |
+
| 1.0572 | 0.8373 | 128 | 1.0948 |
|
86 |
+
| 0.999 | 0.8635 | 132 | 1.0949 |
|
87 |
+
| 1.1098 | 0.8896 | 136 | 1.0951 |
|
88 |
+
| 1.0215 | 0.9158 | 140 | 1.0951 |
|
89 |
+
| 1.0759 | 0.9419 | 144 | 1.0951 |
|
90 |
+
| 1.096 | 0.9681 | 148 | 1.0950 |
|
91 |
+
| 1.08 | 0.9943 | 152 | 1.0950 |
|
92 |
+
|
93 |
+
|
94 |
+
### Framework versions
|
95 |
+
|
96 |
+
- PEFT 0.7.1
|
97 |
+
- Transformers 4.40.2
|
98 |
+
- Pytorch 2.3.0+cu121
|
99 |
+
- Datasets 2.19.1
|
100 |
+
- Tokenizers 0.19.1
|