Karen Róbertsdóttir's picture

34 1

Karen Róbertsdóttir

Nafnlaus

·

AI & ML interests

None yet

Recent Activity

updated a dataset about 7 hours ago

Nafnlaus/font_training_incomplete

new activity 2 days ago

deepseek-ai/Janus-Pro-7B:Surely this is misconfigured? I haven't seen image generation this bad in many years.

new activity 2 days ago

deepseek-ai/Janus-Pro-7B:Surely this is misconfigured? I haven't seen image generation this bad in many years.

View all activity

Organizations

None yet

Nafnlaus's activity

New activity in deepseek-ai/Janus-Pro-7B 2 days ago

Surely this is misconfigured? I haven't seen image generation this bad in many years.

#5 opened 2 days ago by

New activity in Nafnlaus/font_training_incomplete 2 days ago

[bot] Conversion to Parquet

#1 opened 3 days ago by

parquet-converter

New activity in NaniDAO/deepseek-r1-qwen-2.5-32B-ablated 3 days ago

What ablation library / tool did you use?

#1 opened 3 days ago by

New activity in Nafnlaus/ShrimpMoss_Chinese_Censorship_Abliteration 6 days ago

[bot] Conversion to Parquet

#1 opened 6 days ago by

parquet-converter

New activity in chuanli11/Llama-3.2-3B-Instruct-uncensored about 1 month ago

Does not seem at all uncensored.

#6 opened about 1 month ago by

New activity in meta-llama/Llama-3.2-3B-Instruct about 1 month ago

The refusal rate is out of control.

#47 opened about 1 month ago by

New activity in microsoft/Phi-3-mini-4k-instruct 7 months ago

Thanks for the updated version!

#78 opened 7 months ago by

New activity in microsoft/Phi-3-mini-4k-instruct 9 months ago

Instruct model needs work

#35 opened 9 months ago by

New activity in microsoft/Phi-3-mini-4k-instruct-gguf 9 months ago

Phi-3 gives multiple responses with <|end|><|assistant|> as separator

#8 opened 9 months ago by

New activity in microsoft/Phi-3-mini-128k-instruct 9 months ago

Good model but Bullshit chart and inaccurate numbers

#20 opened 9 months ago by

Base model

#10 opened 9 months ago by

Incredible, But Way Too Woke

#27 opened 9 months ago by deleted

New activity in TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUF 12 months ago

WOW - best opensource llm I ever seen !

#1 opened about 1 year ago by

No K_S models?

#24 opened 12 months ago by

New activity in cognitivecomputations/dolphin about 1 year ago

You can't Apache license GPT3/4 outputs.

#17 opened about 1 year ago by

New activity in OpenAssistant/oasst1 about 1 year ago

How sure are you that this is all human generations?

#28 opened about 1 year ago by

New activity in jamesdborin/tiny-mixtral about 1 year ago

What are the details on this model?

#1 opened about 1 year ago by

New activity in hf-internal-testing/Mixtral-tiny about 1 year ago

What's the license on this?

#1 opened about 1 year ago by

New activity in migtissera/Synthia-v1.3 about 1 year ago

Is this dataset generated by GPT-4?

#2 opened about 1 year ago by

New activity in migtissera/Synthia-MoE-v3-Mixtral-8x7B about 1 year ago

How are you "apache-2.0 licensing" GPT-4 outputs?

#3 opened about 1 year ago by