license: apache-2.0
datasets:
- cerebras/SlimPajama-627B
- bigcode/starcoderdata
- timdettmers/openassistant-guanaco
language:
- en
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 PY007/TinyLlama-1.1B-intermediate-step-240k-503b. The dataset used is openassistant-guananco.
How to use
You will need the transformers>=4.31 Do check the TinyLlama github page for more information.
from transformers import AutoTokenizer
import transformers
import torch
model = "PY007/TinyLlama-1.1B-Chat-v0.1"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
prompt = "What are the values in open source projects?"
formatted_prompt = (
f"### Human: {prompt}### Assistant:"
)
sequences = pipeline(
formatted_prompt,
do_sample=True,
top_k=50,
top_p = 0.7,
num_return_sequences=1,
repetition_penalty=1.1,
max_new_tokens=500,
)
for seq in sequences:
print(f"Result: {seq['generated_text']}")
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 30.31 |
| ARC (25-shot) | 32.0 |
| HellaSwag (10-shot) | 54.21 |
| MMLU (5-shot) | 26.71 |
| TruthfulQA (0-shot) | 39.03 |
| Winogrande (5-shot) | 54.93 |
| GSM8K (5-shot) | 0.53 |
| DROP (3-shot) | 4.75 |