nvidia/NVIDIA-Nemotron-Nano-12B-v2
Text Generation
•
12B
•
Updated
•
101k
•
96
Joseph C Steward
jsteward2930
AI & ML interests
Genomics, proteomics, biochemical language models, AI in oncology and cancer care
Recent Activity
liked
a model
about 1 month ago
nvidia/NVIDIA-Nemotron-Nano-12B-v2
liked
a model
about 1 year ago
black-forest-labs/FLUX.1-dev
upvoted
an
article
over 1 year ago
AI Apps in a Flash with Gradio's Reload Mode
Organizations
None yet
liked
a
model
about 1 month ago
liked
a
model
about 1 year ago
upvoted
an
article
over 1 year ago
Article
AI Apps in a Flash with Gradio's Reload Mode
•
27
liked
a
model
over 1 year ago
upvoted
a
paper
over 1 year ago
liked
4
models
over 1 year ago
upvoted
3
papers
over 1 year ago
Vision-Flan: Scaling Human-Labeled Tasks in Visual Instruction Tuning
Paper
•
2402.11690
•
Published
•
10
SPAR: Personalized Content-Based Recommendation via Long Engagement Attention
Paper
•
2402.10555
•
Published
•
35
GhostWriter: Augmenting Collaborative Human-AI Writing Experiences Through Personalization and Agency
Paper
•
2402.08855
•
Published
•
14
reacted to
gsarti's
post with đź‘Ť
over 1 year ago
Post
🔍 Today's pick in Interpretability & Analysis of LMs: Model Editing with Canonical Examples by
@johnhew
@sachen
@lora-x
E. Adams P. Jiang
@manning
This works introduces a model editing approach using individual “canonical” examples to showcase desired/unwanted behavior. An evaluation is then conducted on out-of-distribution samples spanning six datasets (3 introduced in this work) covering settings of interest in bias mitigation, hard syntactic constructions and knowledge-based predictions, while limiting the degradation of the original model’s loss.
Authors experiment with Pythia LMs, finding that LoRa fine-tuning on canonical examples outperforms other established editing methods such as MEMIT.
Then, the approach is tested on Backpack LMs, using a linear combination of sense vectors to disentangle semantic information in the input texts. In particular, authors introduce “sense fine-tuning” where only a handful of sense vectors is updated per example, which is shown to be more efficient yet more effective than regular fine-tuning.
Finally, the relation between the predictions of pre- and post-sense fine-tuning backpack LMs is used to successfully transfer the desired adaptation to a larger standard LM, at no performance cost.
đź“„ Paper: Model Editing with Canonical Examples (2402.06155)
🔍 All daily picks in LM interpretability: https://huggingface.co/collections/gsarti/daily-picks-in-interpretability-and-analysis-of-lms-65ae3339949c5675d25de2f9
This works introduces a model editing approach using individual “canonical” examples to showcase desired/unwanted behavior. An evaluation is then conducted on out-of-distribution samples spanning six datasets (3 introduced in this work) covering settings of interest in bias mitigation, hard syntactic constructions and knowledge-based predictions, while limiting the degradation of the original model’s loss.
Authors experiment with Pythia LMs, finding that LoRa fine-tuning on canonical examples outperforms other established editing methods such as MEMIT.
Then, the approach is tested on Backpack LMs, using a linear combination of sense vectors to disentangle semantic information in the input texts. In particular, authors introduce “sense fine-tuning” where only a handful of sense vectors is updated per example, which is shown to be more efficient yet more effective than regular fine-tuning.
Finally, the relation between the predictions of pre- and post-sense fine-tuning backpack LMs is used to successfully transfer the desired adaptation to a larger standard LM, at no performance cost.
đź“„ Paper: Model Editing with Canonical Examples (2402.06155)
🔍 All daily picks in LM interpretability: https://huggingface.co/collections/gsarti/daily-picks-in-interpretability-and-analysis-of-lms-65ae3339949c5675d25de2f9
updated
a
collection
over 1 year ago
liked
a
Space
over 1 year ago
updated
a
collection
almost 2 years ago
liked
a
model
almost 2 years ago
liked
a
Space
almost 2 years ago