INDEX

Explanations

programming-related structures and keywords

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

auto and automatic variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

SourceChecksum

-0.54

Personendaten

-0.53

цездатний

-0.50

󠁴

-0.50

AnimationsModule

-0.50

 jsPsych

-0.48

 tartalomajánló

-0.48

 referenties

-0.46

 createContext

-0.45

杓

-0.45

POSITIVE LOGITS

 auto

1.22

auto

1.14

 Auto

1.04

Auto

1.01

 AUTO

0.96

AUTO

0.94

setAuto

0.92

 automatic

0.91

 automático

0.84

 자동

0.82

Activations Density 0.015%

programming-related structures and keywords

auto and automatic variations

No Comments

No Known Activations