README.md

metadata

base_model: unsloth/Qwen2.5-14B-Instruct-bnb-4bit
tags:
  - text-generation-inference
  - transformers
  - unsloth
  - qwen2
  - trl
license: apache-2.0
language:
  - en
datasets:
  - HuggingFaceTB/smoltalk
  - HuggingFaceTB/finemath
  - lightblue/tagengo-gpt4
  - LDJnr/Capybara
  - argilla/ifeval-like-data
  - AIRRC/Eudaimonic

A Fishy Model

This model was trained with SFT using Unsloth on the ChatML format with 8k context. Carp models are trained with a combination of pretrain, instruct, and chat datasets.

Changes

• Training dataset had some more "slop" removed.
• Some datasets were added and deleted from training.
• Datasets were reorganized.

Uploaded model

• Developed by: TheTsar1209
• License: apache-2.0
• Finetuned from model : unsloth/Qwen2.5-14B-Instruct-bnb-4bit

This qwen2 model was trained 2x faster with Unsloth and Huggingface's TRL library.