README.md

metadata

language:
  - en
  - fa

Model description|Example output|Banchmark results|How to use|Training and finetuning |

---

Model description

---

Example output:

Example 1:

• Input: "سلام، خوبی؟"
• Output: "سلام، خوشحالم که با شما صحبت می کنم. چطور می توانم به شما کمک کنم؟"

Example 2: - Input: "سلام، خوبی؟" - Output: "سلام، خوشحالم که با شما صحبت می کنم. چطور می توانم به شما کمک کنم؟"

Banchmark results

model	dataset	max_token	prompt	score
base-model-7b	ARC-easy-dev	2	en-1	0.41929
base-model-7b	ARC-easy-dev	80	en-2	0.39122
base-model-7b	ARC-easy-dev	300	en-1	0.34448
model	dataset	max_token	prompt	score
---------------	------------------	-----------	--------	---------
fa-model-7b	ARC-easy-dev	80	en-1	0.37894
fa-model-7b	ARC-easy-dev	80	en-2	0.33333
fa-model-7b	ARC-easy-dev	80	fa-2	0.28771
fa-model-7b	ARC-easy-dev	300	fa-1	0.25752
fa-model-7b	ARC-easy-dev	2	fa-1	0.24035

model	dataset	max_token	prompt	score
base-model-7b	ARC-challenge-dev	80	en-2	0.37123
base-model-7b	ARC-challenge-dev	2	en-2	0.36789
base-model-7b	ARC-challenge-dev	2	en-1	0.35451
base-model-7b	ARC-challenge-dev	80	en-1	0.33779
model	dataset	max_token	prompt	score
---------------	--------------------	-----------	--------	---------
fa-model-7b	ARC-challenge-dev	2	en-1	0.39298
fa-model-7b	ARC-challenge-dev	80	en-1	0.38421
fa-model-7b	ARC-challenge-dev	2	en-2	0.31929
fa-model-7b	ARC-challenge-dev	80	en-2	0.31754

---

How to use

from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("aidal/Persian-Mistral-7B")
model = AutoModelForCausalLM.from_pretrained("aidal/Persian-Mistral-7B")
input_text = "پایتخت ایران کجاست؟"
input_ids = tokenizer(input_text, return_tensors="pt")
outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))

---

Training and finetuning