Locutusque's TinyMistral-248M trained on chat datasets

Base model: Locutusque/TinyMistral-248M with two additional special tokens (<|im_start|> and <|im_end|>)
Datasets:
License: Apache License 2.0
Availability in other ML formats:
- GGUF: Felladrin/gguf-TinyMistral-248M-Chat-v2
- ONNX: Felladrin/onnx-TinyMistral-248M-Chat-v2

Recommended Prompt Format

<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{user_message}<|im_end|>
<|im_start|>assistant

Recommended Inference Parameters

penalty_alpha: 0.5
top_k: 5

Usage Example

from transformers import pipeline

generate = pipeline("text-generation", "Felladrin/TinyMistral-248M-Chat-v2")

messages = [
    {
        "role": "system",
        "content": "You are a highly knowledgeable and friendly assistant. Your goal is to understand and respond to user inquiries with clarity. Your interactions are always respectful, helpful, and focused on delivering the most accurate information to the user.",
    },
    {
        "role": "user",
        "content": "Hey! Got a question for you!",
    },
    {
        "role": "assistant",
        "content": "Sure! What's it?",
    },
    {
        "role": "user",
        "content": "What are some potential applications for quantum computing?",
    },
]

prompt = generate.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)

output = generate(
    prompt,
    max_new_tokens=256,
    penalty_alpha=0.5,
    top_k=5,
)

print(output[0]["generated_text"])

How it was trained

This model was trained with SFTTrainer using the following settings:

Hyperparameter	Value
Learning rate	2e-5
Total train batch size	32
Max. sequence length	2048
Weight decay	0.01
Warmup ratio	0.1
NEFTune Noise Alpha	5
Optimizer	Adam with betas=(0.9,0.999) and epsilon=1e-08
Scheduler	cosine
Seed	42

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	27.42
AI2 Reasoning Challenge (25-Shot)	23.29
HellaSwag (10-Shot)	27.39
MMLU (5-Shot)	23.52
TruthfulQA (0-shot)	41.32
Winogrande (5-shot)	49.01
GSM8k (5-shot)	0.00

Downloads last month: 774

Safetensors

Model size

248M params

Tensor type

F32

Inference Examples

Text Generation

This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.