|
--- |
|
datasets: |
|
- allenai/tulu-3-sft-personas-instruction-following |
|
- simplescaling/s1K-1.1 |
|
- simplescaling/s1K-claude-3-7-sonnet |
|
- FreedomIntelligence/Medical-R1-Distill-Data |
|
- OpenCoder-LLM/opc-sft-stage1 |
|
- cognitivecomputations/SystemChat-2.0 |
|
- anthracite-org/kalo-opus-instruct-22k-no-refusal |
|
- allura-org/scienceqa_sharegpt |
|
- KodCode/KodCode-V1-SFT-R1 |
|
license: apache-2.0 |
|
language: |
|
- en |
|
- zh |
|
base_model: |
|
- mistralai/Mistral-Small-24B-Base-2501 |
|
library_name: transformers |
|
tags: |
|
- instruct |
|
- conversational |
|
--- |
|
|
|
# Sertraline 24b |
|
|
|
 |
|
|
|
## About |
|
|
|
An *actually decent* instruct SFT tune of Mistral Small 3. |
|
|
|
## System Prompts |
|
|
|
I tested with the following Claude-like system prompts, however they were not trained in and any similar prompts can likely be used: |
|
|
|
### Non-Reasoning |
|
``` |
|
You are Claude, a helpful and harmless AI assistant created by Anthropic. |
|
``` |
|
|
|
### Reasoning |
|
``` |
|
You are Claude, a helpful and harmless AI assistant created by Anthropic. Please contain all your thoughts in <think> </think> tags, and your final response right after the closing </think> tag. |
|
``` |
|
|
|
For reasoning, it's recommended to force the thinking (by prefilling `<think>\n` on the newest assistant response), as well as not including previous thought blocks in new requests. |
|
|
|
## Instruct Template |
|
|
|
v7-Tekken, same as the original instruct model. |
|
|
|
## Dataset |
|
|
|
This model was trained on [allura-org/inkstructmix-v0.2.1](https://hf.co/datasets/allura-org/inkstructmix-v0.2.1). |
