INDEX

Explanations

programming syntax and structure elements

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

code punctuation endings

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

 itſelf

-0.88

RenderAtEndOf

-0.86

MLLoader

-0.81

wapV

-0.75

 myſelf

-0.75

 erſt

-0.73

ScopeManager

-0.73

ſicht

-0.71

ſelf

-0.71

ſelben

-0.69

POSITIVE LOGITS

 kids

0.41

 tourney

0.35

kids

0.33

 stats

0.33

 hubby

0.33

感じで

0.31

 meds

0.30

Japanese

0.29

 guys

0.29

canestro

0.29

Activations Density 0.128%

programming syntax and structure elements

code punctuation endings

No Comments

No Known Activations