GPT4chan-8B-FP8

GPT4chan 8B FP8

This model is v2ray/GPT4chan-8B quantized to fp8 using vllm-project/llm-compressor.

Trained using 8x H100 with global batch size 64, using 2e-4 learning rate, for 4000 steps, which is approximately 5 epochs.

Prompt Format

board<|start_header_id|>id<|end_header_id|>content<|start_header_id|>id<|end_header_id|>content...<|start_header_id|>id<|end_header_id|>

Example:

g<|start_header_id|>1<|end_header_id|>speculate thread\nwhat will ai land be like in 2025<|start_header_id|>2<|end_header_id|>

Terms of Service

By downloading and inferencing with this model, you (the users) agree to donate your soul to us (v2AI) for unholy purposes, also you will probably become a slave of us too! :3

You also agree that every output generated is only your own imagination and has nothing to do with this perfectly mentally sane and normal model, every bad output is made by you, not provided by us, so we take no responsibility of the bad outputs.

Usage Guidelines

You (the users) agree to use this model for:

• Mentally sane generations.
• Research purposes only.
• Sending L.O.V.E. to the world.

You (the users) agree NOT to use this model for:

• Dead internet theory.
• Doing inharmonious things.
• Saying gex.