uploaded readme
Browse files
README.md
ADDED
|
@@ -0,0 +1,90 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Quantization made by Richard Erkhov.
|
| 2 |
+
|
| 3 |
+
[Github](https://github.com/RichardErkhov)
|
| 4 |
+
|
| 5 |
+
[Discord](https://discord.gg/pvy7H8DZMG)
|
| 6 |
+
|
| 7 |
+
[Request more models](https://github.com/RichardErkhov/quant_request)
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
starchat2-15b-sft-v0.1 - bnb 8bits
|
| 11 |
+
- Model creator: https://huggingface.co/HuggingFaceH4/
|
| 12 |
+
- Original model: https://huggingface.co/HuggingFaceH4/starchat2-15b-sft-v0.1/
|
| 13 |
+
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
Original model description:
|
| 18 |
+
---
|
| 19 |
+
license: bigcode-openrail-m
|
| 20 |
+
base_model: bigcode/starcoder2-15b
|
| 21 |
+
tags:
|
| 22 |
+
- alignment-handbook
|
| 23 |
+
- generated_from_trainer
|
| 24 |
+
datasets:
|
| 25 |
+
- HuggingFaceH4/airoboros-3.2
|
| 26 |
+
- HuggingFaceH4/Code-Feedback
|
| 27 |
+
- HuggingFaceH4/orca-math-word-problems-200k
|
| 28 |
+
- HuggingFaceH4/SystemChat
|
| 29 |
+
- HuggingFaceH4/capybara
|
| 30 |
+
model-index:
|
| 31 |
+
- name: starcoder2-15b-sft-v5.0
|
| 32 |
+
results: []
|
| 33 |
+
---
|
| 34 |
+
|
| 35 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 36 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 37 |
+
|
| 38 |
+
# Model Card for starchat2-15b-sft-v0.1
|
| 39 |
+
|
| 40 |
+
This model is a fine-tuned version of [bigcode/starcoder2-15b](https://huggingface.co/bigcode/starcoder2-15b) on the HuggingFaceH4/airoboros-3.2, the HuggingFaceH4/Code-Feedback, the HuggingFaceH4/orca-math-word-problems-200k, the HuggingFaceH4/SystemChat and the HuggingFaceH4/capybara datasets.
|
| 41 |
+
It achieves the following results on the evaluation set:
|
| 42 |
+
- Loss: 0.6614
|
| 43 |
+
|
| 44 |
+
## Model description
|
| 45 |
+
|
| 46 |
+
More information needed
|
| 47 |
+
|
| 48 |
+
## Intended uses & limitations
|
| 49 |
+
|
| 50 |
+
More information needed
|
| 51 |
+
|
| 52 |
+
## Training and evaluation data
|
| 53 |
+
|
| 54 |
+
More information needed
|
| 55 |
+
|
| 56 |
+
## Training procedure
|
| 57 |
+
|
| 58 |
+
### Training hyperparameters
|
| 59 |
+
|
| 60 |
+
The following hyperparameters were used during training:
|
| 61 |
+
- learning_rate: 2e-05
|
| 62 |
+
- train_batch_size: 8
|
| 63 |
+
- eval_batch_size: 8
|
| 64 |
+
- seed: 42
|
| 65 |
+
- distributed_type: multi-GPU
|
| 66 |
+
- num_devices: 16
|
| 67 |
+
- total_train_batch_size: 128
|
| 68 |
+
- total_eval_batch_size: 128
|
| 69 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 70 |
+
- lr_scheduler_type: cosine
|
| 71 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 72 |
+
- num_epochs: 3
|
| 73 |
+
|
| 74 |
+
### Training results
|
| 75 |
+
|
| 76 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 77 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 78 |
+
| 0.6422 | 1.0 | 910 | 0.6910 |
|
| 79 |
+
| 0.5701 | 2.0 | 1820 | 0.6639 |
|
| 80 |
+
| 0.5227 | 3.0 | 2730 | 0.6614 |
|
| 81 |
+
|
| 82 |
+
|
| 83 |
+
### Framework versions
|
| 84 |
+
|
| 85 |
+
- Transformers 4.39.0.dev0
|
| 86 |
+
- Pytorch 2.1.2+cu121
|
| 87 |
+
- Datasets 2.16.1
|
| 88 |
+
- Tokenizers 0.15.1
|
| 89 |
+
|
| 90 |
+
|