datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
rlhf-gemma3-indfood-1kgemma3-pythonic-function-tool-calling-v1gemma3-refusal-axis-data
Gemma 3 12B Refusal Axis: Activations and SAE Encodings
Mechanistic interpretability data for studying the refusal axis in Gemma 3 12B-IT.
This dataset contains the layer-41 residual-stream activations and Gemma Scope 2 SAE
encodings produced by running 280 contrastive prompt pairs through Gemma 3 12B, plus the
refusal direction vectors derived from those activations.
It is the data side of the gemma3-refusal-axis
project: an independent investigation of whether refusal in Gemma 3… See the full description on the dataset page: https://huggingface.co/datasets/abotresol/gemma3-refusal-axis-data.smoltalk-reasoning-gemma3-40kopenwebtext-gemma3-tokenized-1024-activations-layer23
OpenWebText — Gemma-3-1B Hidden State Activations (Layer 23)
Precomputed hidden state activations before layer 23 of Gemma-3-1B-IT for the OpenWebText dataset, tokenized with sequence length 1024.
Designed for training a Titans memory layer that replaces layer 23 of Gemma 3.
Dataset Structure
Each example contains the inputs to layer 23:
Field
Shape
Dtype
Description
activations
(1024, 1152)
float32
Hidden state activations (cast from bfloat16)
mask(1024… See the full description on the dataset page: https://huggingface.co/datasets/veriga/openwebtext-gemma3-tokenized-1024-activations-layer23.RULER-262144-gemma3-instruct2026_08_05_refinement_5env_gemma3_12b_gemma4_31b_tokgemma3
[!Note]
This repository corresponds to the launch version of Gemma 3n E2B, to be used with Hugging Face transformers,
supporting text, audio, and vision (image and video) inputs.
Gemma 3n models have multiple architecture innovations:
They are available in two sizes based on effective parameters. While the raw parameter count of this model is 6B, the architecture design allows the model to be run with a memory footprint comparable to a traditional 2B model by offloading low-utilization… See the full description on the dataset page: https://huggingface.co/datasets/aneeshm44/gemma3.gemma-3-taide-12b-chat-eval-logs-and-scoresgemma-3-4b-it-eval-logs-and-scoresgemma-3-4B-T1-it-eval-logs-and-scoresGemma-3-12b-it-eval-logs-and-scores2026_08_09_refinement_5env_gemma3_12b_gemma4_31b_flsft_tokgemma-3-27b-it-eval-logs-and-scorespython4-gemma3-27b-eft-v2-logsCoDIT-Gemma3
Dataset Name
🤖 Teacher Model
📂 Dataset Link
CoDIT-Gemma3 💎
google/gemma-3-27b-it
CoDIT-Gemma3 ↗
CoDIT-Qwen3-8B 🐉
Qwen/Qwen3-8B
CoDIT-Qwen3-8B ↗
CoDIT-Qwen3-30B 🚀
Qwen/Qwen3-30B-A3B
CoDIT-Qwen3-30B ↗
CoDIT-Gemma3
CoDIT-Gemma3 is a synthetic conversation dataset derived from LMSYS-Chat-1M [Zhang+, ICLR24].
250,333 user instructions sourced from LMSYS-Chat-1M
250,333 assistant responses automatically synthesized using CoDIT with google/gemma-3-27b-it, generating… See the full description on the dataset page: https://huggingface.co/datasets/Tatsuya-Ichinose/CoDIT-Gemma3.2026_07_19_collect_leandojo_gemma3_12b_gemma4_31b_flsft_toksmoltalk-gemma3-10242026_08_20_refinement_math_chess_gemma3_12b_gemma4_31b_transition_feedback_tokpython4-gemma3-27b-eft-v2-eval2026_07_19_collect_leandojo_gemma3_12b_gemma4_31bRULER-262144-gemma3-base2026_07_29_collect_mathnet_gemma3_12b_gemma4_31b_flsft_tokgemma3_1b_dpo_123_oracle_v1-training-datagemma-3-12b-longfact-jury-labels
Gemma-3-12B LongFact hallucination labels (cross-provider LLM jury)
6,471 entity-level factuality annotations over 300 long-form completions from
google/gemma-3-12b-it, produced by a three-judge cross-provider LLM jury voting
independently on shared, archived web-search evidence — with the jury's agreement
against human-derived public gold labels measured and reported below.
Gemma-3-12B has no public entity-level hallucination labels (the existing public sets —… See the full description on the dataset page: https://huggingface.co/datasets/praxagent-org/gemma-3-12b-longfact-jury-labels.subliminal10k-subliminal-gemma3-4b-itfinewebedu10B-gemma32026_07_20_collect_codeforces_gemma3_12b_gemma4_31bwildchat-1m-gpt-4-1-regenerated-english-unused-gemma32026_07_29_collect_mathnet_gemma3_12b_gemma4_31b