INDEX

Explanations

mentions of the character Andre

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

Andre's actions and statements

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

Chham

-0.45

 microfilm

-0.41

الحياه

-0.41

Publication

-0.41

LabelTagHelper

-0.41

chtes

-0.40

gence

-0.40

PyExc

-0.40

Household

-0.40

♥♥

-0.40

POSITIVE LOGITS

 Andre

0.62

ViewInit

0.58

AutoScaleMode

0.53

Andre

0.51

 André

0.44

TagMode

0.44

Dre

0.43

André

0.43

YOND

0.43

Dre

0.42

Activations Density 0.004%

mentions of the character Andre

Andre's actions and statements

No Comments

No Known Activations