INDEX

Explanations

phrases related to research findings and their implications

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

after certain words

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

 surla

-0.75

 <<<<<<<<<<<<<<

-0.65

httphttps

-0.60

CloseOperation

-0.59

MigrationBuilder

-0.55

__(/*!

-0.48

 Comprometido

-0.48

ViewFeatures

-0.46

MessageTagHelper

-0.46

HasColumnName

-0.45

POSITIVE LOGITS

BeginContext

0.49

ftance

0.43

 prud

0.36

ftances

0.36

ably

0.35

iosi

0.34

adena

0.34

|};

0.34

 könnt

0.33

ilités

0.33

Activations Density 0.074%

phrases related to research findings and their implications

after certain words

No Comments

No Known Activations