Uploaded model

  • Developed by: bbcc88
  • License: apache-2.0
  • Finetuned from model : llm-jp/llm-jp-3-13b

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

llm-jp/llm-jp-3-13bを使ってqLoraしたモデルです。

松尾研大規模言語モデル講座2024のコンペ用の提出モデル作成の一環として作成・公開しています。

推論方法 松尾研大規模言語モデル講座2024のコンペのタスクの推論方法を以下に記載します。

from unsloth import FastLanguageModel
import torch
max_seq_length = 512 
dtype = None 
load_in_4bit = True
model_id = "bbcc88/llm-jp-3-13b-it_lora"

# FastLanguageModel インスタンスを作成
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name=model_id,
    dtype=dtype,
    load_in_4bit=load_in_4bit,
    trust_remote_code=True,
)

# データセットを準備
import json
datasets = []
with open("/content//elyza-tasks-100-TV_0.jsonl", "r") as f:
    item = ""
    for line in f:
      line = line.strip()
      item += line
      if item.endswith("}"):
        datasets.append(json.loads(item))
        item = ""


# 推論するためにモデルのモードを変更
FastLanguageModel.for_inference(model)

results = []
for dt in tqdm(datasets):
  input = dt["input"]

  prompt = f"""### 指示\n{input}\n### 回答\n"""

  inputs = tokenizer([prompt], return_tensors = "pt").to(model.device)

  outputs = model.generate(**inputs, max_new_tokens = 512, use_cache = True, do_sample=False, repetition_penalty=1.2)
  prediction = tokenizer.decode(outputs[0], skip_special_tokens=True).split('\n### 回答')[-1]

  results.append({"task_id": dt["task_id"], "input": input, "output": prediction})
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for bbcc88/llm-jp-3-13b-it_lora

Finetuned
(1083)
this model