File size: 1,401 Bytes
591afb3
fccf749
591afb3
 
fccf749
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
---
library_name: peft
license: apache-2.0
---
# minihf_evaluator_openllama_7b

`minihf_evaluator_openllama_7b` is a LoRA instruct fine-tune of [OpenLLaMA 7B](https://huggingface.co/openlm-research/open_llama_7b).

The sequence `<|end|>` was used to separate the prompt and response. The correct way to prompt the model is: `Does 2 + 2 = 4?<|end|>`. The tokenizer will prepend a BOS token (`<s>`) by default. The response will end with an EOS token (`</s>`).

## Training procedure

`minihf_evaluator_openllama_7b` was fine-tuned for 100,000 examples on 90% [Muennighoff/flan](https://huggingface.co/datasets/Muennighoff/flan) / 10% [databricks/databricks-dolly-15k](https://huggingface.co/datasets/databricks/databricks-dolly-15k) using batch size 4 per GPU on 8 40GB A100 GPUs. Examples where the prompt and response would not fit into 2,048 tokens were dropped. The fine-tuning was done using the following command:

```bash
accelerate launch make_evaluator.py --output-dir minihf_evaluator_openllama_7b
```

The following `bitsandbytes` quantization config was used during training:
- load_in_8bit: False
- load_in_4bit: True
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: nf4
- bnb_4bit_use_double_quant: True
- bnb_4bit_compute_dtype: bfloat16

### Framework versions

- PEFT 0.4.0.dev0