Evaluation datasets maintained by EleutherAI
AI & ML interests
Large language models, scaling laws, AI Alignment, democratization of DL
Recent Activity
View all activity
Papers
Agent Memory Is a Surface for Endogenous Authorization Laundering
Soohak: A Mathematician-Curated Benchmark for Evaluating Research-level Math Capabilities of LLMs
Organization Card
Welcome to EleutherAI's HuggingFace page. We are a non-profit research lab focused on interpretability, alignment, and ethics of artificial intelligence. Our open source models are hosted here on HuggingFace.
You may also be interested in our GitHub, website, or Discord server.
This collection contains the model and data artifacts from O'Brien et al. (2025). https://deepignorance.ai
-
Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs
Paper • 2508.06601 • Published • 7 -
EleutherAI/deep-ignorance-unfiltered
Text Generation • 7B • Updated • 850 • 7 -
EleutherAI/deep-ignorance-e2e-strong-filter
Text Generation • 7B • Updated • 764 • 1 -
EleutherAI/deep-ignorance-strong-filter-pt-weak-filter-anneal
Text Generation • 7B • Updated • 62 • 1
Evaluation datasets maintained by EleutherAI
This collection contains the model and data artifacts from O'Brien et al. (2025). https://deepignorance.ai
-
Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs
Paper • 2508.06601 • Published • 7 -
EleutherAI/deep-ignorance-unfiltered
Text Generation • 7B • Updated • 850 • 7 -
EleutherAI/deep-ignorance-e2e-strong-filter
Text Generation • 7B • Updated • 764 • 1 -
EleutherAI/deep-ignorance-strong-filter-pt-weak-filter-anneal
Text Generation • 7B • Updated • 62 • 1
models 981
EleutherAI/pythia-160m-weight-seed2
Text Generation • Updated • 2.04k
EleutherAI/pythia-160m-weight-seed1
Text Generation • Updated • 2.16k
EleutherAI/pythia-160m-weight-seed2-corrected
Text Generation • Updated • 1.15k
EleutherAI/pythia-160m-weight-seed1-corrected
Text Generation • Updated • 1.35k
EleutherAI/pythia-160m-weight-seed3
Text Generation • Updated • 1.41k
EleutherAI/tiny-Phi3ForCausalLM
Text Generation • 514k • Updated • 945
EleutherAI/olmo3-7b-sdf-sft-clean150
Text Generation • 7B • Updated • 149
EleutherAI/olmo3-7b-sdf-sft-scrub-b1reset150
Text Generation • 7B • Updated • 139
EleutherAI/bergson-wikitext-gpt2-leaderboard
Updated • 24
EleutherAI/qwen3-8b-djinnsdf-dolci
Text Generation • 8B • Updated • 1.78k
datasets 297
EleutherAI/bergson-wikitext-gpt2-leaderboard-bank
Preview • Updated • 867
EleutherAI/hack-ignition-benchmark
Viewer • Updated • 521k • 593 • 2
EleutherAI/reward-hacking-sdf-djinn
Viewer • Updated • 2.97k • 45
EleutherAI/fineweb-heldout-queries-2048
Updated • 68
EleutherAI/pile-heldout-queries-2048
Updated • 38
EleutherAI/djinn-problems-v1.0
Viewer • Updated • 1.22k • 60
EleutherAI/PARTIAL_LDS-retrain-bank-gpt2medium-16k-bs32
Viewer • Updated • 1.3k • 620
EleutherAI/PARTIAL_LDS-retrain-bank-muon-N64k-bs256
Updated • 727
EleutherAI/PARTIAL_LDS-retrain-bank-london16k-bs256-adamw
Viewer • Updated • 1.48k • 394
EleutherAI/LDS-retrain-bank-london16k-bs256-muon
Viewer • Updated • 2k • 489