# MoxoffSrL/Moxoff-Phi3Mini-PPO

This model is an updated version of a 128k instruct model aligned with PPO and trained on ultrafeedback binarized preferences cleaned. The card says it was evaluated on hellaswag, arc challenge, and mmlu test sets and warns that it was not aligned to human preferences for safety, so it can produce problematic outputs.

## Cite this

- Page: https://huggingbay.xyz/artifact/hf-model-moxoffsrl-moxoff-phi3mini-ppo
- Citation: MoxoffSrL/Moxoff-Phi3Mini-PPO. The Hugging Bay model record. https://huggingbay.xyz/artifact/hf-model-moxoffsrl-moxoff-phi3mini-ppo (retrieved 2026-11-16).
- Check a downloaded file: run `shasum -a 256 <file>` (macOS), `sha256sum <file>` (Linux) or `Get-FileHash -Algorithm SHA256 <file>` (Windows), then compare with that file's SHA-256 at https://huggingbay.xyz/api/artifacts/hf-model-moxoffsrl-moxoff-phi3mini-ppo/files?limit=100. How checks work: https://huggingbay.xyz/verification

## Facts

- Type: LLM
- Task: text-generation
- Framework: transformers
- License: mit: Permissive open license: commercial use, modification, and redistribution are permitted. (https://spdx.org/licenses/MIT.html)
- Source: https://huggingface.co/MoxoffSrL/Moxoff-Phi3Mini-PPO
- Hosting: hosted (9 files)
- Files: 9 hosted of 13 listed
- Needs its own code to run: Tagged custom_code on its source page. Plain llama.cpp, Ollama and LM Studio commands are not offered for this model unless a GGUF file is listed.
- File counts: "listed" is every file in the source repository; only hosted files download from Hugging Bay, and the download plan lists the files it selects for one download, so its count can differ.
- Hosted here: 9 hosted of 13 listed
- Full repo upstream: 7.64 GB
- Recommended file: 7.64 GB (3 files)
- Download: https://huggingbay.xyz/api/artifacts/hf-model-moxoffsrl-moxoff-phi3mini-ppo/download-plan (download plan: https://huggingbay.xyz/api/artifacts/hf-model-moxoffsrl-moxoff-phi3mini-ppo/download-plan)
- Per-file SHA-256: available for 9 of 9 hosted files via https://huggingbay.xyz/api/artifacts/hf-model-moxoffsrl-moxoff-phi3mini-ppo/files?limit=100
- Availability: Ready
- Downloadable: yes
- Run-ready: no (Hugging Bay has not certified a run path for this model; hosted bytes are downloadable and hash-checkable only)
- Peer-assisted fallbacks: 0 verified, 0 pending review
- Downloads: 2636
- Stars: 0
- Card completeness: 96%

## Description

This model is an updated version of a 128k instruct model aligned with PPO and trained on ultrafeedback binarized preferences cleaned. The card says it was evaluated on hellaswag, arc challenge, and mmlu test sets and warns that it was not aligned to human preferences for safety, so it can produce problematic outputs.

## Model Card Facts

- Declared license: mit
- Pipeline: text-generation
- Library: transformers
- Task categories: text-generation
- Languages: en
- Referenced datasets: argilla/ultrafeedback-binarized-preferences-cleaned
- Metrics: hellaswag, arc_challenge, m_mmlu 5 shot
- File hints: 13 upstream file samples

## Referenced Data

- Dataset: argilla/ultrafeedback-binarized-preferences-cleaned

## File Hints

- .gitattributes (1.5 KB)
- README.md (3.3 KB)
- added_tokens.json (293 B)
- config.json (3.6 KB)
- generation_config.json (90 B)
- model-00001-of-00002.safetensors (4.6 GB) LFS
- model-00002-of-00002.safetensors (2.5 GB) LFS
- model.safetensors.index.json (16 KB)
- special_tokens_map.json (569 B)
- tokenizer.json (1.8 MB)
- tokenizer.model (488 KB) LFS
- tokenizer_config.json (3.1 KB)

## Quality Checklist

- Summary: passed
- Description: passed
- Model card metadata: passed
- License: passed
- Provenance: passed
- File inventory: passed
- Hosted bytes: passed
- Scan result: passed
- Manifest signature: passed

## Safety Notes

- External catalog metadata imported from Hugging Face public API on 2026-09-25T08:38:41.193Z.
- Open license detected as mit.
- Model card references dataset context: argilla/ultrafeedback-binarized-preferences-cleaned.
- Not a run kit: hosted, hash-checked bytes only; run readiness is not certified for this model yet.

## Actions

- Web: https://huggingbay.xyz/artifact/hf-model-moxoffsrl-moxoff-phi3mini-ppo
- Files API: https://huggingbay.xyz/api/artifacts/hf-model-moxoffsrl-moxoff-phi3mini-ppo/files?limit=100
- Metadata: https://huggingbay.xyz/api/artifacts/hf-model-moxoffsrl-moxoff-phi3mini-ppo/metadata.json
- Distribution: https://huggingbay.xyz/api/artifacts/hf-model-moxoffsrl-moxoff-phi3mini-ppo/distribution
- Manifest: https://huggingbay.xyz/api/manifests/hf-model-moxoffsrl-moxoff-phi3mini-ppo
- Download plan (per-file SHA-256 + signed download URLs): https://huggingbay.xyz/api/artifacts/hf-model-moxoffsrl-moxoff-phi3mini-ppo/download-plan
- Hosted download rows: https://huggingbay.xyz/api/downloadable-files?artifactId=hf-model-moxoffsrl-moxoff-phi3mini-ppo&limit=100
- Request mirror: not applicable.
- Submit peer fallback for review: POST https://huggingbay.xyz/api/artifacts/hf-model-moxoffsrl-moxoff-phi3mini-ppo/decentralized-fallbacks
