Kkk

Michalea's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in Qwen/Qwen3-4B-Thinking-2507 2 days ago

RoPE theta 5mln instead of 1mln

#15 opened 2 days ago by

Michalea

New activity in lmsys/SGLang-EAGLE3-Qwen3-Next-80B-A3B-Instruct-FP8-SpecForge-Meituan 10 days ago

The comparison with the original MTP

👍 1

1

#2 opened 10 days ago by

Michalea

New activity in nvidia/Qwen3-235B-A22B-Eagle3 16 days ago

context length/number of generated tokens during training

#4 opened 16 days ago by

Michalea

New activity in AngelSlim/Qwen3-8B_eagle3 20 days ago

Datasets used to create this head

#5 opened 20 days ago by

Michalea

New activity in unsloth/GLM-4.7-Flash-REAP-23B-A3B-GGUF 21 days ago

A low number of evaluation benchmarks

2

#5 opened 28 days ago by

Michalea

New activity in RedHatAI/Qwen3-235B-A22B-speculator.eagle3 21 days ago

Context length and regeneration

4

#1 opened 22 days ago by

Michalea

New activity in GadflyII/GLM-4.7-Flash-NVFP4 26 days ago

MTP quality, 47 layer

3

#7 opened 27 days ago by

Michalea

New activity in nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4 26 days ago

Efficiency of NVFP4 vs FP16/8

➕ 2

#4 opened 26 days ago by

Michalea

New activity in speakleash/Bielik-11B-v3.0-Instruct 27 days ago

opis wersji 3.0

2

#1 opened about 2 months ago by

jacek2024

New activity in cerebras/GLM-4.7-Flash-REAP-23B-A3B 28 days ago

Evaluation

👍 3

#3 opened 28 days ago by

Michalea

New activity in unsloth/GLM-4.7-Flash-FP8-Dynamic 28 days ago

Severe Looping/Repetitive Output when using --kv-cache-dtype fp8 with GLM-4.7-Flash-FP8-Dynamic on vLLM

4

#2 opened about 1 month ago by

ShelterW

New activity in nvidia/gpt-oss-120b-Eagle3-throughput about 2 months ago

Inconsistent description with the evaluation results

1

#3 opened about 2 months ago by

Michalea

New activity in AngelSlim/Qwen3-1.7B_eagle3 about 2 months ago

Data used to train the EAGLE head

#5 opened about 2 months ago by

Michalea

New activity in nex-agi/SGLANG-EAGLE3-Qwen3-32B-Nex-N1 about 2 months ago

Training details - Question about Reasoning | /think

1

#1 opened about 2 months ago by

Michalea

New activity in Qwen/Qwen3-Reranker-4B 7 months ago

does the tokenizer need to be updated for this model?

1

#5 opened 8 months ago by

electroglyph