INDEX

Explanations

references to articles and blogging content

oai_token-act-pair · gpt-4o-mini Triggered by @bot

blog posts and archives

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Clements

-0.40

rzost

-0.38

 inter

-0.38

Manbalar

-0.37

verifyException

-0.37

 Oppen

-0.36

toThrow

-0.35

ⓧ

-0.35

 interdit

-0.35

 near

-0.35

POSITIVE LOGITS

0.68

 blogging

0.68

 Blogging

0.68

 blog

0.68

 articles

0.63

 Blog

0.63

 blogs

0.62

 виправивши

0.60

blog

0.59

Blog

0.59

Activations Density 0.018%

references to articles and blogging content

blog posts and archives

No Comments

No Known Activations