INDEX

Explanations

strong and assertive expressions of belief or conviction

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

firm believer, faith, grasp, or established state

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

 DataAccess

-0.50

 getItemCount

-0.49

 collage

-0.48

 totalSupply

-0.48

 millionaires

-0.47

Билгалдахарш

-0.47

 шля

-0.46

 Mayo

-0.46

 Gewer

-0.46

 anthology

-0.45

POSITIVE LOGITS

 firmly

0.89

 solid

0.77

solid

0.75

 sólidos

0.74

 sólido

0.73

strong

0.71

 strong

0.70

Solid

0.69

 sólida

0.69

firm

0.68

Activations Density 0.006%

strong and assertive expressions of belief or conviction

firm believer, faith, grasp, or established state

No Comments

No Known Activations