sszymczyk/DeepSeek-V3.2-4Layers-GGUF
15B • Updated • 225
Stanisław Szymczyk
sszymczyk
AI & ML interests
None yet
Recent Activity
new activity 7 days ago
fraunhofer-iis/elmod-2.7b-it:A problem with pre-tokenizer regex new activity 22 days ago
AngelSlim/Hy4-preview-GGUF:upstreaming llama.cpp Hy4-preview support new activity about 1 month ago
MiniMaxAI/MiniMax-Text-01:Zero-valued token embeddings in model.embed_tokens.weight may disrupt the token sampling processOrganizations
None yet
A problem with pre-tokenizer regex
❤️ 1
3
#2 opened 10 days ago
by sszymczyk
upstreaming llama.cpp Hy4-preview support
3
#1 opened 24 days ago
by sszymczyk
Zero-valued token embeddings in model.embed_tokens.weight may disrupt the token sampling process
#43 opened about 1 month ago
by sszymczyk
Add NVFP4 scale tensors to model.safetensors.index.json
1
#5 opened 4 months ago
by sszymczyk
updated a
model 6 months ago
published a
model 6 months ago
updated a
model 6 months ago
published a
model 6 months ago
updated a
model 6 months ago
published a
model 6 months ago
updated a
model 6 months ago
published a
model 6 months ago
updated 2
models 6 months ago
published 2
models 6 months ago
Running the model with a dense attention
4
#35 opened 9 months ago
by sszymczyk
Quick Start section in README.md is a bit misleading
👍 1
2
#10 opened 7 months ago
by sszymczyk
Temperature and top_p values are swapped in the example code:
👍 1
1
#2 opened 8 months ago
by sszymczyk
Recommended sampling parameters?
👍 1
2
#3 opened 8 months ago
by sszymczyk