18 2 93

Markus Heiervang

marksverdhei

marksverdhei

AI & ML interests

NLP

Recent Activity

Reacted to MonsterMMORPG's post with 🔥 3 days ago

FLUX Redux is a hidden Gem I am still doing huge research to publish an amazing fully Public - no paywalled Tutorial, but this is generated via SwarmUI Style Model Merge Strength : 0.5 FLUX Guidance Scale is : 6 Used base model is my FLUX fine tuned model with 256 images via Kohya SS GUI as shown in tutorial ( https://youtu.be/FvpWy1x5etM ) - 70 epoch Prompt : anime ohwx man walking in a jungle <segment:yolo-face_yolov9c.pt-1,0.7,0.5> ohwx man, anime

updated a dataset 7 days ago

marksverdhei/data-by-countries

updated a dataset 8 days ago

marksverdhei/hdi-ihdi-democracy-by-country

View all activity

Organizations

marksverdhei's activity

Reacted to MonsterMMORPG's post with 🔥 3 days ago

Post

1742

FLUX Redux is a hidden Gem

I am still doing huge research to publish an amazing fully Public - no paywalled Tutorial, but this is generated via SwarmUI

Style Model Merge Strength : 0.5

FLUX Guidance Scale is : 6

Used base model is my FLUX fine tuned model with 256 images via Kohya SS GUI as shown in tutorial ( https://youtu.be/FvpWy1x5etM ) - 70 epoch

Prompt : anime ohwx man walking in a jungle <segment:yolo-face_yolov9c.pt-1,0.7,0.5> ohwx man, anime

2 replies

updated a dataset 7 days ago

marksverdhei/data-by-countries

Viewer • Updated 7 days ago • 201 • 59

updated a dataset 8 days ago

marksverdhei/hdi-ihdi-democracy-by-country

Viewer • Updated 8 days ago • 194 • 43

updated a collection 21 days ago

Norwegian GPTs

Collection

A collection of GPT-style decoder only models trained on Norwegian • 5 items • Updated 21 days ago

liked a Space 22 days ago

Running

462

📢

UGI Leaderboard

updated a Space about 1 month ago

Sleeping

↗️🔎

Explore Embedding Inversion

This repository lets you explore inverted embeddings

liked a model 4 months ago

mistralai/Mistral-Large-Instruct-2407

Updated Oct 16 • 13.6k • 809

New activity in mistralai/Mistral-Large-Instruct-2407 4 months ago

License for tokenizer

#22 opened 4 months ago by

marksverdhei

liked a model 5 months ago

hf-internal-testing/tiny-random-gpt2

Updated Apr 8 • 1.36M • 6

New activity in norallm/normistral-7b-warm-instruct 6 months ago

GGUF 8bit model not recommended.

#11 opened 6 months ago by

marksverdhei

liked a model 6 months ago

meta-llama/Meta-Llama-3-70B-Instruct

Text Generation • Updated Sep 27 • 102k • • 1.44k

Reacted to mlabonne's post with 🔥 6 months ago

Post

16664

✂️ Uncensor any LLM with abliteration

I wrote an article about abliteration and how NeuralDaredevil-8B was created. Beyond removing alignment, I believe it's an interesting technique with a lot of potential. It's basically fine-tuning without retraining.

In this article, we see how it works, implement it in Google Colab, and heal the abliterated model to recover the performance drop due to this technique. The final model is an uncensored and high-quality model with the highest MMLU score on the Open LLM Leaderboard (8B category).

https://huggingface.co/blog/mlabonne/abliteration