datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
qwen36-q4km-swebench-lite-saliency-overfit-layercompare
Qwen3.6 Q4_K_M SWE-bench Lite Saliency Layer Residency
This dataset contains the reduced-scope Q4_K_M-only comparison of:
attention10_baseline_nommap
hybrid_promote_5_b11_nommap
The run uses SWE-bench Lite oracle-context patch-generation prompts, Q4_K_M
MTP, 64k context, q8_0/q8_0 KV cache, Flash Attention, no-mmap, identical
prompt order, and identical sampling. Quality grading is deferred, but
prediction JSONLs are included for later official SWE-bench evaluation.
Read… See the full description on the dataset page: https://huggingface.co/datasets/jakeatx/qwen36-q4km-swebench-lite-saliency-overfit-layercompare.translategemma-4b-it-Q4_K_M-GGUF
TranslateGemma 4B IT Q4_K_M GGUF
This repository contains a GGUF Q4_K_M conversion of Google TranslateGemma 4B IT.
Model information
Base model: google/translategemma-4b-it
Source revision: 10042cb0e6e7fdce748996a71dc3dc432a4e0c89
llama.cpp revision used in the project validation: 2048b5913d51beab82dfe29955f9008130b936c0
Artifact filename: translategemma-4b-it-Q4_K_M.gguf
Size: 2489909120 bytes
SHA-256:… See the full description on the dataset page: https://huggingface.co/datasets/ctc88haha/translategemma-4b-it-Q4_K_M-GGUF.qwen36-q4km-saliency-vs-attention-residency-100
Qwen3.6 Q4_K_M Saliency vs Attention Residency
This dataset package contains a paired 100-prompt throughput replay comparing
the original attention-spaced MoE layer residency policy against the
saliency-selected b11 hot-layer policy for
Qwen3.6-35B-A3B-UD-Q4_K_M.gguf.
The prompt file is an exact replay of qwen36-moe-layer-residency-20260520-205703: 56 SciCode
subproblems and 44 Artificial Analysis Terminal-Bench-Hard prompts. The run
uses 64k context, q8_0/q8_0 KV cache, Flash… See the full description on the dataset page: https://huggingface.co/datasets/jakeatx/qwen36-q4km-saliency-vs-attention-residency-100.nerobali-results-localllm-gpt-oss-20b-q4_k_mOpenHerms7B_Q4_k_m_sql_evaluationsmetadata_queries_vectordb_llama_8b_Q4_K_MOpenHerms7B_Q4_k_m_sql_evaluationsministral-3_3b-instruct-2512-q4_K_M_discretegemma3_12b-it-q4_K_M_hierarchicalgemma3_12b-it-q4_K_M_discrete_resultsgemma3_4b-it-q4_K_M_discretegemma3_4b-it-q4_K_M_continuousgemma3_4b-it-q4_K_M_discrete_standardizedministral-3_3b-instruct-2512-q4_K_M_discrete_standardizedgemma3_4b-it-q4_K_M_discrete_resultsministral-3_3b-instruct-2512-q4_K_M_discrete_resultsOpenHerms7B_Q4_k_m_Responsesgemma3_12b-it-q4_K_M_discretegemma3_4b-it-q4_K_M_hierarchicalgemma3_12b-it-q4_K_M_continuousgemma3_12b-it-q4_K_M_hierarchical_standardizedgemma3_12b-it-q4_K_M_discrete_standardizedgemma3_12b-it-q4_K_M_hierarchical_resultsgemma3_4b-it-q4_K_M_hierarchical_resultsgemma3_12b-it-q4_K_M_continuous_resultsgemma3_4b-it-q4_K_M_continuous_resultsgemma3_4b-it-q4_K_M_hierarchical_standardized