Juk Armstrong

jukofyork's activity

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

New activity in zai-org/GLM-5 8 days ago

Please release a model with native 4-bit quantization

👀 ➕ 10

4

#4 opened 15 days ago by

calycekr

New activity in jukofyork/creative-writing-control-vectors-v3.0 10 days ago

“The doom lies in yourself, not in your name.”

👀 7

345

#15 opened 6 months ago by

jukofyork

New activity in jukofyork/GLM-4.5-DRAFT-0.6B-v3.0 27 days ago

GLM 4.7

1

#1 opened about 1 month ago by

jpsequeira

New activity in moonshotai/Kimi-K2.5 27 days ago

Should `rope_scaling.beta_fast` be `1.0`?

4

#6 opened about 1 month ago by

jukofyork

New activity in unsloth/Kimi-K2.5-GGUF 28 days ago

Thinking block parsing issue for Q3_K_XL

5

#1 opened 29 days ago by

EvgeniyIV

New activity in AesSedai/Kimi-K2.5-GGUF 28 days ago

Fix for thinking block not showing up

👍 4

5

#1 opened 28 days ago by

jukofyork

New activity in moonshotai/Kimi-K2-Thinking about 1 month ago

I think `eos_token_id` should be `163585`in `config.json`?

#51 opened about 1 month ago by

jukofyork

New activity in jukofyork/command-a-03-2025-DRAFT-0.8B-v3.0-GGUF about 2 months ago

Draft/Speculative decoding

3

#1 opened 2 months ago by

yano2mch

New activity in jukofyork/Qwen3-Coder-Instruct-DRAFT-0.75B-GGUF about 2 months ago

Draft model with ik_llama.cpp

👍 1

2

#1 opened 6 months ago by

Chico70

New activity in jukofyork/creative-writing-control-vectors-v3.0 about 2 months ago

Create Boy in love

#17 opened 2 months ago by

Imma9907

New activity in jukofyork/merge-lora 3 months ago

Amazing It worked, Thanks.

1

#1 opened 3 months ago by

pankajmathur

New activity in jukofyork/creative-writing-control-vectors-v3.0 3 months ago

“The doom lies in yourself, not in your name.”

👀 7

345

#15 opened 6 months ago by

jukofyork

New activity in daavidhauser/chess-bot-3000-250m 3 months ago

The "Softmax Bottleneck" might be interesting to look at

2

#1 opened 3 months ago by

jukofyork

New activity in moonshotai/Kimi-K2-Thinking 4 months ago

Was this really trained during QAT using a symmetric 4bit quant with only 15/16 values used?

👀 9

3

#26 opened 4 months ago by

jukofyork

New activity in jukofyork/Mistral-Large-Instruct-2411-DRAFT-0.4B-v3.0 4 months ago

These don't really work.

5

#1 opened 4 months ago by

Lockout

New activity in jukofyork/creative-writing-control-vectors-v3.0 4 months ago

“The doom lies in yourself, not in your name.”

👀 7

345

#15 opened 6 months ago by

jukofyork

New activity in jukofyork/Mistral-Large-Instruct-2411-DRAFT-0.4B-v3.0 4 months ago

These don't really work.

5

#1 opened 4 months ago by

Lockout

New activity in future-architect/Llama-3.3-Future-Code-Instructions 4 months ago

The data seems to have lost the `task_language` field?

2

#2 opened 4 months ago by

jukofyork