Henk Poley
HenkPoley
·
AI & ML interests
None yet
Organizations
None yet
HenkPoley's activity
Maybe note in the readme if this uses the new 'June release' of Phi-3-mini-4k-instruct
#1 opened about 2 months ago
by
HenkPoley
Delcos/Velara is 'fine-tuned' not 'pretrained'
6
#440 opened 9 months ago
by
HenkPoley
NEW! OpenLLMLeaderboard 2023 fall update
20
#356 opened 10 months ago
by
clefourrier
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1644340617257-noauth.png)
Just pointing to this interesting paper (with code) on preparing models to be merged
1
#1 opened 9 months ago
by
HenkPoley
Meanwhile there is an improved version 😅
#1 opened 9 months ago
by
HenkPoley
Great work! Question about the merge specifics.
3
#4 opened 10 months ago
by
UMCU
![](https://cdn-avatars.huggingface.co/v1/production/uploads/618b9a79ba796dc2bf3f4412/66ustwEp1EDU_rxcXJBIO.jpeg)
PEBCAK: Others are split / Shouldn't Q2_K be the smallest GGUF?
1
#1 opened 10 months ago
by
HenkPoley
Sadly Q8_0 just gives gibberish with llama.cpp
2
#1 opened 10 months ago
by
HenkPoley
will there be a tinymistral?
1
#3 opened 11 months ago
by
LaferriereJC
Please add a license to config.json
#1 opened 10 months ago
by
HenkPoley
Will you try fine-tuning mistral with the dataset used here?
1
#1 opened 11 months ago
by
HenkPoley
Isn't it odd that these models compress pretty well?
3
#1 opened about 1 year ago
by
HenkPoley
tensor 'token_embd.weight' has wrong shape
11
#1 opened about 1 year ago
by
HenkPoley
GGML?
7
#2 opened over 1 year ago
by
creative420
q4_0, q4_1, q5_0
1
#1 opened about 1 year ago
by
venketh
uncensored version
8
#16 opened over 1 year ago
by
Feng7815