INDEX

Explanations

the repeated use of the word "as"

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

as programming/data type

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

verwijspagina

-0.63

 '\\;'

-0.63

findpost

-0.59

ngrx

-0.58

новниш

-0.57

delwed

-0.56

transQ

-0.56

/**

-0.55

getColumnIndex

-0.54

GraphicsUnit

-0.54

POSITIVE LOGITS

GOTREF

0.59

 scientific

0.42

 AssemblyTitle

0.40

 scientifique

0.38

 matrix

0.38

MigrationBuilder

0.37

 omnipotent

0.34

amssymb

0.34

 científicas

0.34

 scientist

0.34

Activations Density 0.000%

the repeated use of the word "as"

as programming/data type

No Comments

No Known Activations