File size: 2,167 Bytes
3691bbb 437fbd2 6ee2e20 7aae1f4 6ee2e20 437fbd2 |
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 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 |
---
library_name: peft
---
## Inference
```python
from transformers import AutoTokenizer
import transformers
import torch
model = "qanastek/LLaMa-2-FrenchMedMCQA"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
prompt = "Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request. ### Instruction: We are giving you a scientific question (easy level) and five answers options (associated to « A », « B », « C », « D », « E »). Your task is to find the correct(s) answer(s) based on scientific facts, knowledge and reasoning. Don't generate anything other than one of the following characters : 'A B C D E'. ### Input: Parmi les propositions suivantes, quelle est celle qui est exacte? Lorsqu'on ajoute un acide fort à une solution tampon: (A) Le pH reste constant (B) Le pH diminue légèrement (C) Le constituant basique du tampon reste constant (D) Le constituant acide du tampon réagit (E) Le rapport acide/base reste inchangé ### Response: "
seq = pipeline(
prompt,
do_sample=True,
top_k=10,
num_return_sequences=1,
eos_token_id=tokenizer.eos_token_id,
max_length=200,
)[0]
print(seq['generated_text'][len(prompt):])
```
## Training procedure
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: False
- bnb_4bit_compute_dtype: float16
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: False
- bnb_4bit_compute_dtype: float16
### Framework versions
- PEFT 0.4.0
- PEFT 0.4.0
|