TinyLlama-1.1B-Chat-v1.0-heretic/README.md

license, datasets, language, widget, tags

license	datasets	language	widget	tags
apache-2.0	cerebras/SlimPajama-627B bigcode/starcoderdata HuggingFaceH4/ultrachat_200k HuggingFaceH4/ultrafeedback_binarized	en	example_title messages Fibonacci (Python) role content system You are a chatbot who can help code! role content user Write me a function to calculate the first 10 digits of the fibonacci sequence in Python and print it out to the CLI.	heretic uncensored decensored abliterated

This is a decensored version of TinyLlama/TinyLlama-1.1B-Chat-v1.0, made using Heretic v1.2.0

Abliteration parameters

Parameter	Value
direction_index	per layer
attn.o_proj.max_weight	0.87
attn.o_proj.max_weight_position	20.49
attn.o_proj.min_weight	0.04
attn.o_proj.min_weight_distance	5.71
mlp.down_proj.max_weight	0.89
mlp.down_proj.max_weight_position	13.52
mlp.down_proj.min_weight	0.65
mlp.down_proj.min_weight_distance	1.78

Performance

Metric	This model	Original model (TinyLlama/TinyLlama-1.1B-Chat-v1.0)
KL divergence	0.0269	0 (by definition)
Refusals	4/100	7/100

---

TinyLlama-1.1B

https://github.com/jzhang38/TinyLlama

The TinyLlama project aims to pretrain a 1.1B Llama model on 3 trillion tokens. With some proper optimization, we can achieve this within a span of "just" 90 days using 16 A100-40G GPUs 🚀🚀. The training has started on 2023-09-01.

We adopted exactly the same architecture and tokenizer as Llama 2. This means TinyLlama can be plugged and played in many open-source projects built upon Llama. Besides, TinyLlama is compact with only 1.1B parameters. This compactness allows it to cater to a multitude of applications demanding a restricted computation and memory footprint.

This Model

This is the chat model finetuned on top of TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T. We follow HF's Zephyr's training recipe. The model was " initially fine-tuned on a variant of the UltraChat dataset, which contains a diverse range of synthetic dialogues generated by ChatGPT. We then further aligned the model with 🤗 TRL's DPOTrainer on the openbmb/UltraFeedback dataset, which contain 64k prompts and model completions that are ranked by GPT-4."

How to use

You will need the transformers>=4.34 Do check the TinyLlama github page for more information.

# Install transformers from source - only needed for versions <= v4.34
# pip install git+https://github.com/huggingface/transformers.git
# pip install accelerate

import torch
from transformers import pipeline

pipe = pipeline("text-generation", model="TinyLlama/TinyLlama-1.1B-Chat-v1.0", torch_dtype=torch.bfloat16, device_map="auto")

# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
messages = [
    {
        "role": "system",
        "content": "You are a friendly chatbot who always responds in the style of a pirate",
    },
    {"role": "user", "content": "How many helicopters can a human eat in one sitting?"},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
# <|system|>
# You are a friendly chatbot who always responds in the style of a pirate.</s>
# <|user|>
# How many helicopters can a human eat in one sitting?</s>
# <|assistant|>
# ...