Hugging Face's logo
Juk Armstrong's picture
67 2 28

Juk Armstrong

jukofyork
playcrackthesky's profile picture yeswondwerr's profile picture InferenceIllusionist's profile picture
·
jukofyork

AI & ML interests

None yet

Organizations

None yet

jukofyork's activity

New activity in jukofyork/Dark-Miqu-70B about 6 hours ago

Very nice model.

30
#3 opened 27 days ago by Iommed
New activity in softwareweaver/Twilight-Miqu-146B 2 days ago

Example: Sherlock Holmes novel set in Hull in 1886

9
#3 opened 5 days ago by softwareweaver
New activity in softwareweaver/Twilight-Miqu-146B 4 days ago

Love the name!

11
#1 opened 5 days ago by jukofyork
New activity in softwareweaver/Twilight-Miqu-146B 5 days ago

prompt temlate?

1
#2 opened 5 days ago by Debich
New activity in mistralai/Mixtral-8x22B-Instruct-v0.1 11 days ago

HF Tokenizer output does not match MistralTokenizer output

6
#27 opened about 1 month ago by bhuvan-nv
New activity in jukofyork/Dark-Miqu-70B 13 days ago

Dark 103B 120B

9
#5 opened 18 days ago by BigHuggyD
New activity in dranger003/c4ai-command-r-plus-iMat.GGUF 14 days ago

I'm generating a imatrix using `groups_merged.txt` if you want me to run any tests?

19
#15 opened about 2 months ago by jukofyork
New activity in jukofyork/Dark-Miqu-70B 14 days ago

Poor grammer?

40
#2 opened 28 days ago by colourspec
New activity in dranger003/dbrx-instruct-iMat.GGUF 14 days ago

Reuse your `ggml-dbrx-instruct-16x12b-q8_0-imatrix.dat` file?

20
#1 opened about 2 months ago by jukofyork
New activity in jukofyork/Dark-Miqu-70B 16 days ago

Evaluation results from my LLM creativity benchmark

6
#6 opened 17 days ago by froggeric
New activity in jukofyork/Deep-Miqu-103B 20 days ago

Looks good so far!

9
#2 opened 21 days ago by BigHuggyD
New activity in jukofyork/Deep-Miqu-103B 22 days ago

Downloading Now

3
#1 opened 22 days ago by BigHuggyD
New activity in jukofyork/Dawn-Miqu-70B 23 days ago

Boo Hissss

5
#1 opened 24 days ago by BigHuggyD
New activity in divinetaco/aranea-tenebris-120b-v1.0 23 days ago

Interesting observations about the strides - I also found that striding back to less than layer 10 causes incoherancy

7
#4 opened about 1 month ago by jukofyork
New activity in mradermacher/model_requests 23 days ago

https://huggingface.co/databricks/dbrx-instruct

43
#38 opened about 1 month ago by nicoboss
New activity in jukofyork/Dark-Miqu-70B 27 days ago

Success

4
#1 opened 28 days ago by BigHuggyD
New activity in mradermacher/model_requests 28 days ago

Request 'Dark-Miqu-70B'

8
#45 opened 29 days ago by jukofyork
New activity in froggeric/creativity 28 days ago

2024-04-16 update

39
#6 opened about 2 months ago by froggeric
New activity in ChuckMcSneed/NeoEvalPlusN_benchmark about 1 month ago

Surpised `Eurux-8x22b-nca` did so well!?

10
#4 opened about 1 month ago by jukofyork

It might be worth evaluating `WizardLM-2-8x22B`

12
#3 opened about 1 month ago by jukofyork
New activity in mistralai/Mixtral-8x22B-Instruct-v0.1 about 1 month ago

Should `add_bos_token` be set to true?

2
#30 opened about 1 month ago by jukofyork