pythia-160m-mlx

English | 日本語

Model Summary

This is an MLX conversion of EleutherAI/pythia-160m, part of EleutherAI's Pythia suite -- a set of models trained with identical data ordering and architecture across scales, designed for interpretability and scientific research into LLM training dynamics.

This is an unofficial, community-produced conversion. All credit for the base model goes to EleutherAI.

Usage

pip install -U mlx-lm
mlx_lm.generate --model masahiroid/pythia-160m-mlx --prompt "The history of artificial intelligence began" --max-tokens 200
from mlx_lm import load, generate

model, tokenizer = load("masahiroid/pythia-160m-mlx")
response = generate(model, tokenizer, prompt="The history of artificial intelligence began", max_tokens=200, verbose=True)
print(response)

Note: this is a base model (not instruction-tuned), so it continues text rather than following chat-style instructions.

Specs

Item Value
Base model EleutherAI/pythia-160m (160M parameters)
Precision float16
Framework MLX
Peak memory (tested) ~0.35 GB

Notes

  • This is a community conversion, not an official release from EleutherAI.

Security

Audited against its upstream with model-audit-lite: weight format, bundled code, and a machine-readable lineage (ML-BOM). Details, checksums and how to reproduce: SECURITY.md.


モデルの概要

本モデルは、EleutherAIが公開している解釈可能性・学習ダイナミクス研究向けモデル群 「Pythia」の一つ EleutherAI/pythia-160m のMLX変換版です(非公式)。元モデルの著作権はEleutherAIに帰属します。

使い方

pip install -U mlx-lm
mlx_lm.generate --model masahiroid/pythia-160m-mlx --prompt "The history of artificial intelligence began" --max-tokens 200
from mlx_lm import load, generate

model, tokenizer = load("masahiroid/pythia-160m-mlx")
response = generate(model, tokenizer, prompt="The history of artificial intelligence began", max_tokens=200, verbose=True)
print(response)

注: これはベースモデルです(instruction-tuning無し)。チャット形式の指示に従うのではなく、 文章の続きを生成します。

Specs

Item Value
ベースモデル EleutherAI/pythia-160m(160M params)
精度 float16
フレームワーク MLX
ピークメモリ(実測) 約0.35 GB

備考

  • 本変換は非公式のコミュニティ版です。EleutherAIによる公式リリースではありません。

セキュリティー

model-audit-lite で変換元と突き合わせて監査済みです(重みの形式、同梱コード、機械可読な系譜=ML-BOM)。詳細・チェックサム・再現方法は SECURITY.md をご覧ください。

Downloads last month
594
Safetensors
Model size
0.2B params
Tensor type
F16
·
MLX
Hardware compatibility
Log In to add your hardware

Quantized

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for masahiroid/pythia-160m-mlx

Finetuned
(379)
this model