Chikuma_10.7B/README.md

language, license, library_name, tags, base_model, pipeline_tag, model-index

language	license	library_name	tags	base_model	pipeline_tag	model-index
en	apache-2.0	transformers	merge	sethuiyer/SynthIQ-7b openchat/openchat-3.5-0106	text-generation	name results Chikuma_10.7B task dataset metrics source type name text-generation Text Generation name type config split args AI2 Reasoning Challenge (25-Shot) ai2_arc ARC-Challenge test num_few_shot 25 type value name acc_norm 65.7 normalized accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Chikuma_10.7B Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type split args HellaSwag (10-Shot) hellaswag validation num_few_shot 10 type value name acc_norm 84.31 normalized accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Chikuma_10.7B Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args MMLU (5-Shot) cais/mmlu all test num_few_shot 5 type value name acc 64.81 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Chikuma_10.7B Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args TruthfulQA (0-shot) truthful_qa multiple_choice validation num_few_shot 0 type value mc2 57.01 url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Chikuma_10.7B Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args Winogrande (5-shot) winogrande winogrande_xl validation num_few_shot 5 type value name acc 79.56 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Chikuma_10.7B Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args GSM8k (5-shot) gsm8k main test num_few_shot 5 type value name acc 57.62 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Chikuma_10.7B Open LLM Leaderboard

NOTE: For experimental purposes

Chikuma is a 10.7B parameter model and is a merge of the following models using LazyMergekit:

• sethuiyer/SynthIQ-7b
• openchat/openchat-3.5-0106

The name "Chikuma" is inspired by the Chikuma River, the longest in Japan, known for its continuous flow and meandering path. This metaphorically represents the model's depth, fluidity, and adaptability in processing and understanding language.

It also perfectly fits the approach taken here - Depth Upscaling, inspired by SOLAR 10.7B.

Nous LLM Evaluation (with ChatML Prompt Template)

Model	AGIEval	GPT4All	TruthfulQA	Bigbench	Average
SynthIQ-7b	42.67	73.71	56.51	44.59	54.37
openchat/openchat-3.5-0106	44.17	73.72	52.53	44.4	53.71
Chikuma_10.7B	42.41	73.41	56.69	43.5	54

More details can be found here

Recommended Prompt Template (Experimental)

<|im_start|>GPT4 Correct system
You are Chikuma, a constantly learning AI assistant who strives to be
insightful, engaging, and helpful. You possess vast knowledge and creativity,
but also a humble curiosity about the world and the people you interact
with. If you don't know the answer to a question, please don't share false information. 
Always use <|end_of_turn|> when you want to end the answer.<|im_end|>
<|im_start|>GPT4 Correct User:
{{Input}}
<|im_end|>GPT4 Correct Assistant:

ChatML also works, but make sure to add the sentence "Always use <|end_of_turn|> when you want to end the answer" as the default eos token is <|end_of_turn|>.

Tested to work well in :

1. text-generation-webui, LLaMa-Precise sampling settings.
2. transformers text generation pipeline, temperature=4.0, top_k=50, top_p=0.01.

🧩 Configuration

slices:
  - sources:
    - model: sethuiyer/SynthIQ-7b
      layer_range: [0, 24]
  - sources:
    - model: openchat/openchat-3.5-0106
      layer_range: [8, 32]
merge_method: passthrough
dtype: bfloat16

Ollama:

Chikuma is on Ollama. You can use it by running the command ollama run stuehieyr/chikuma in your terminal. If you have limited computing resources, check out this video to learn how to run it on a Google Colab backend.

💻 Usage

sys_message = ''' 
You are Chikuma, a constantly learning AI assistant who strives to be
insightful, engaging, and helpful. You possess vast knowledge and creativity,
but also a humble curiosity about the world and the people you interact
with. If you don't know the answer to a question, please don't share false information. 
Always use <|end_of_turn|> when you want to end the answer.
'''

question = '''
Tell me what is a large language model in under 250 words.
'''

messages = [{"role":"system", "content": sys_message}, {"role": "user", "content": question}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=4.0, top_k=50, top_p=0.01)
print(outputs[0]["generated_text"])

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	68.17
AI2 Reasoning Challenge (25-Shot)	65.70
HellaSwag (10-Shot)	84.31
MMLU (5-Shot)	64.81
TruthfulQA (0-shot)	57.01
Winogrande (5-shot)	79.56
GSM8k (5-shot)	57.62