datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
qwen36-35b-a3b-fp8-two-blackhole-tt-cache
Qwen3.6-35B-A3B-FP8 two-Blackhole TT cache
This dataset contains the generated same-source compressed owner-bank cache used by a public Qwen/Qwen3.6-35B-A3B-FP8 two-Blackhole runtime project.
Project repo:
https://github.com/PMZFX/TT-qwen36-35b-a3b-fp8-two-blackhole
The GitHub repo contains the runtime code, TT-Lang spike, reliability harnesses, release notes, and helper scripts. This dataset supplies the generated TT cache that is too large for the GitHub repo.
Contents… See the full description on the dataset page: https://huggingface.co/datasets/katostrofik/qwen36-35b-a3b-fp8-two-blackhole-tt-cache.perfectblend-Qwen3-235B-A22B-Instruct-2507-FP8-generatedmagpie-qwen2.5-pro-1m-v0.1-Qwen3-235B-A22B-Instruct-2507-FP8-generatedGLM-5.2-FP8-nemotron-codealpaca
GLM-5.2-FP8-nemotron-codealpaca
Training data for UCloud-org/GLM-5.2-FP8-DFlash,
a DFlash speculative-decoding drafter for
zai-org/GLM-5.2-FP8.
A mix of code / math / chat prompts from two public instruction datasets
(see Composition); all assistant responses are regenerated by GLM-5.2-FP8 so the targets match the
verifier's own output distribution — the data recipe specified in the
DFlash paper (Appendix A.1).
800,022 single-turn conversations, English-dominant
Generation:… See the full description on the dataset page: https://huggingface.co/datasets/JessieWei/GLM-5.2-FP8-nemotron-codealpaca.NVIDIA-Nemotron-3-Super-120B-A12B-FP8-eval-logs-and-scoresglm53-flash-fidelity-fp8-v1
fidelity--glm53flash.malaiwah.quant.official-fp8
A quant fidelity dataset in hidden form, produced by engines/tools/hf_capture.py from zai-org/GLM-5.3-Flash.
The cut
the final hidden state handed to lm_head -- after the text model's final norm and immediately before the head matmul -- captured as the head module's input via torch.nn.Module.register_forward_pre_hook; replay applies the head ONLY (no final norm at replay time: the capture already sits after it).… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/glm53-flash-fidelity-fp8-v1.perfectblend-Qwen3-235B-A22B-Instruct-2507-FP8-generatedGLM-5.2-FP8-nemotron-codealpaca-thinking
GLM-5.2-FP8 Nemotron-CodeAlpaca Thinking Dataset
820,790 single-turn conversations generated by zai-org/GLM-5.2-FP8
with thinking enabled.
Prompt source
Rows (public)
Nemotron-Post-Training-Dataset-v2
800,944
CodeAlpaca-20k (corrected prompts, instruction + "\n\n" + input)
19,846
Total
820,790
Generation: temperature=1.0, top_p=0.95, max_tokens=24576, thinking
enabled. The CodeAlpaca prompts here include the input field.
Relationship to… See the full description on the dataset page: https://huggingface.co/datasets/JessieWei/GLM-5.2-FP8-nemotron-codealpaca-thinking.GLM-5.2-FP8-magpie-ultrachat
GLM-5.2-FP8 Regenerated Responses (Magpie + UltraChat mix)
A combined instruction-response dataset of 507,864 single-turn conversations. The
prompts are drawn from two public instruction datasets; the responses were freshly
regenerated with zai-org/GLM-5.2-FP8.
It was built as on-policy distillation data for training speculative-decoding drafts
(DFlash / DSpark) for GLM-5.2 — i.e. so the draft learns from GLM-5.2's own output
distribution — but it is a general-purpose GLM-5.2… See the full description on the dataset page: https://huggingface.co/datasets/mgoin/GLM-5.2-FP8-magpie-ultrachat.fruit-fidelity-fp8-v1
fidelity--fruit.malaiwah.quant.fp8
A quant fidelity dataset in hidden form, produced by engines/tools/hf_capture.py from malaiwah/GLM-5.2-SIQ-Fruit-fp8.
The cut
the final hidden state handed to lm_head -- after the text model's final norm and immediately before the head matmul -- captured as the head module's input via torch.nn.Module.register_forward_pre_hook; replay applies the head ONLY (no final norm at replay time: the capture already sits after it). Same cut… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/fruit-fidelity-fp8-v1.glm53-fidelity-fp8-v1
fidelity--glm53.malaiwah.quant.fp8
A quant fidelity dataset in hidden form, produced by engines/tools/hf_capture.py from zai-org/GLM-5.3.
The cut
the final hidden state handed to lm_head -- after the text model's final norm and immediately before the head matmul -- captured as the head module's input via torch.nn.Module.register_forward_pre_hook; replay applies the head ONLY (no final norm at replay time: the capture already sits after it). Same cut as… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/glm53-fidelity-fp8-v1.glm52-fidelity-fp8-v1
fidelity--glm52.malaiwah.quant.fp8
A quant fidelity dataset in hidden form, produced by engines/tools/hf_capture.py from zai-org/GLM-5.2-FP8.
The cut
the final hidden state handed to lm_head -- after the text model's final norm and immediately before the head matmul -- captured as the head module's input via torch.nn.Module.register_forward_pre_hook; replay applies the head ONLY (no final norm at replay time: the capture already sits after it). Same cut as… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/glm52-fidelity-fp8-v1.SPEED-Bench-Qualitative-Qwen3.6-35B-A3B-FP8-torchspec
SPEED-Bench Qualitative Qwen3.6 TorchSpec
TorchSpec-compatible chat dataset generated from the 880 fully materialized SPEED-Bench qualitative prompts.
Responses were generated on Doubleword with Qwen/Qwen3.6-35B-A3B-FP8 using /v1/chat/completions and max_tokens=4096.
Files
data/train.jsonl: 880 rows in TorchSpec chat format.
Schema
Each row contains:
{
"id": "<speedbench_question_id>",
"conversations": [
{"role": "user", "content":… See the full description on the dataset page: https://huggingface.co/datasets/jamesdborin/SPEED-Bench-Qualitative-Qwen3.6-35B-A3B-FP8-torchspec.fp8-quantizationClimateMBERT-syn-qwen3-30b-a3b-fp8-10k-seed42
ClimateMBERT Synthetic Qwen3 30B A3B FP8 10K Seed42
Synthetic continuation dataset generated from WxChat/ClimateMBERT_syn train split.
Source dataset: WxChat/ClimateMBERT_syn
Source split: train
Sampling: shuffled with random seed 42, ranks 0..9999
Rows: 10,000
Generator: Qwen/Qwen3-30B-A3B-Instruct-2507-FP8
Inference: vLLM on Clariden GH200 GPUs, tensor parallel size 2, non-eager mode
Max tokens: 4096
Generation config: temperature 0.7, top_p 0.8, top_k 20, min_p 0.0… See the full description on the dataset page: https://huggingface.co/datasets/JingweiNi/ClimateMBERT-syn-qwen3-30b-a3b-fp8-10k-seed42.ClimateMBERT-syn-qwen35-122b-fp8-10k-seed42
ClimateMBERT Synthetic Qwen3.5 FP8 10K Seed42
Synthetic continuation dataset generated from WxChat/ClimateMBERT_syn train split.
Source dataset: WxChat/ClimateMBERT_syn
Source split: train
Sampling: shuffled with random seed 42, ranks 0..9999
Rows: 10,000
Generator: Qwen/Qwen3.5-122B-A10B-FP8
Inference: vLLM on Clariden GH200 GPUs, tensor parallel size 4, non-eager mode
Max tokens: 4096
No-thinking mode: chat_template_kwargs={"enable_thinking": false}
Generation config: temperature… See the full description on the dataset page: https://huggingface.co/datasets/JingweiNi/ClimateMBERT-syn-qwen35-122b-fp8-10k-seed42.experiment_real_fp8
/hub_data4/seohyun/saves/ecva_instruct/full/sft/checkpoint-350 · happy8825/valid_ecva_clean results
Model: /hub_data4/seohyun/saves/ecva_instruct/full/sft/checkpoint-350
Dataset: happy8825/valid_ecva_clean
Generated: 2026-01-08 12:12:25Z
Metrics
Metric
Value
Total samples
924
With GT
0
Parsed answers
0
Top-1 accuracy
0
Recall@5
0
MRR
0
The uploaded JSON contains full per-sample predictions produced via t3_infer_with_vllm.bash.… See the full description on the dataset page: https://huggingface.co/datasets/happy8825/experiment_real_fp8.ctest-subset-Qwen3.5-397B-A17B-FP8-dynamic-speculator-datasetfp8_quantized
/hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-350-fp8 · happy8825/valid_ecva_clean results
Model: /hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-350-fp8
Dataset: happy8825/valid_ecva_clean
Generated: 2025-12-24 05:49:28Z
Metrics
Metric
Value
Total samples
924
With GT
0
Parsed answers
0
Top-1 accuracy
0
Recall@5
0
MRR
0
The uploaded JSON contains full per-sample predictions produced via t3_infer_with_vllm.bash.… See the full description on the dataset page: https://huggingface.co/datasets/happy8825/fp8_quantized.experiment_fp8
/hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-350 · happy8825/valid_ecva_clean results
Model: /hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-350
Dataset: happy8825/valid_ecva_clean
Generated: 2026-01-08 11:33:45Z
Metrics
Metric
Value
Total samples
924
With GT
0
Parsed answers
0
Top-1 accuracy
0
Recall@5
0
MRR
0
The uploaded JSON contains full per-sample predictions produced via t3_infer_with_vllm.bash.… See the full description on the dataset page: https://huggingface.co/datasets/happy8825/experiment_fp8.magpie-llama-405b-fp8-ja-18kfp8-wfp8a16kvfp8
FP8 W-FP8 / A-FP16 / KV-FP8 (Llama-3.1-8B-Instruct)
w_of_wfp8a16kvfp8_llama_31_8b/ — FP8 weights
The per-tensor scaled FP8 (E4M3) weights of Llama-3.1-8B-Instruct — a type-conversion variant (weights cast to FP8).
Stored per layer: layer_0.safetensors … layer_31.safetensors + embeddings.safetensors.
The 7 linears per layer are cast to FP8 (fp8 weight + fp32 per-tensor scale); everything else stays fp16:
key
dtype
shape
self_attn.q_proj.weight… See the full description on the dataset page: https://huggingface.co/datasets/jsyeom/fp8-wfp8a16kvfp8.