README.md

metadata

language:
  - ja
  - en
license: apache-2.0
tags:
  - japanese
  - causal-lm
inference: false
model-index:
  - name: calm2-7b-chat
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: AI2 Reasoning Challenge (25-Shot)
          type: ai2_arc
          config: ARC-Challenge
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: acc_norm
            value: 40.27
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cyberagent/calm2-7b-chat
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: HellaSwag (10-Shot)
          type: hellaswag
          split: validation
          args:
            num_few_shot: 10
        metrics:
          - type: acc_norm
            value: 68.12
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cyberagent/calm2-7b-chat
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU (5-Shot)
          type: cais/mmlu
          config: all
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 39.39
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cyberagent/calm2-7b-chat
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: TruthfulQA (0-shot)
          type: truthful_qa
          config: multiple_choice
          split: validation
          args:
            num_few_shot: 0
        metrics:
          - type: mc2
            value: 41.96
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cyberagent/calm2-7b-chat
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Winogrande (5-shot)
          type: winogrande
          config: winogrande_xl
          split: validation
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 64.96
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cyberagent/calm2-7b-chat
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GSM8k (5-shot)
          type: gsm8k
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 4.93
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cyberagent/calm2-7b-chat
          name: Open LLM Leaderboard

CyberAgentLM2-7B-Chat (CALM2-7B-Chat)

Model Description

CyberAgentLM2-Chat is a fine-tuned model of CyberAgentLM2 for dialogue use cases.

Requirements

• transformers >= 4.34.1
• accelerate

Usage

import transformers
from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer

assert transformers.__version__ >= "4.34.1"

model = AutoModelForCausalLM.from_pretrained("cyberagent/calm2-7b-chat", device_map="auto", torch_dtype="auto")
tokenizer = AutoTokenizer.from_pretrained("cyberagent/calm2-7b-chat")
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

prompt = """USER: AIによって私達の暮らしはどのように変わりますか？
ASSISTANT: """

token_ids = tokenizer.encode(prompt, return_tensors="pt")
output_ids = model.generate(
    input_ids=token_ids.to(model.device),
    max_new_tokens=300,
    do_sample=True,
    temperature=0.8,
    streamer=streamer,
)

Chat Template

USER: {user_message1}
ASSISTANT: {assistant_message1}<|endoftext|>
USER: {user_message2}
ASSISTANT: {assistant_message2}<|endoftext|>
USER: {user_message3}
ASSISTANT: {assistant_message3}<|endoftext|>

Model Details

• Model size: 7B
• Context length: 32768
• Model type: Transformer-based Language Model
• Language(s): Japanese, English
• Developed by: CyberAgent, Inc.
• License: Apache-2.0

Author

Ryosuke Ishigami

Citations

@article{touvron2023llama,
  title={LLaMA: Open and Efficient Foundation Language Models},
  author={Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier and Martinet, Xavier and Lachaux, Marie-Anne and Lacroix, Timoth{\'e}e and Rozi{\`e}re, Baptiste and Goyal, Naman and Hambro, Eric and Azhar, Faisal and Rodriguez, Aurelien and Joulin, Armand and Grave, Edouard and Lample, Guillaume},
  journal={arXiv preprint arXiv:2302.13971},
  year={2023}
}

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	43.27
AI2 Reasoning Challenge (25-Shot)	40.27
HellaSwag (10-Shot)	68.12
MMLU (5-Shot)	39.39
TruthfulQA (0-shot)	41.96
Winogrande (5-shot)	64.96
GSM8k (5-shot)	4.93