INDEX

Explanations

mentions of the word "Brown" and its variations in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

transQ

-0.81

 betweenstory

-0.73

<unused21>

-0.68

<unused28>

-0.68

<unused74>

-0.68

[@BOS@]

-0.68

<unused41>

-0.68

<unused52>

-0.68

<unused3>

-0.68

<unused8>

-0.68

POSITIVE LOGITS

 Brown

0.79

Brown

0.71

 BROWN

0.66

 brown

0.60

brow

0.59

brown

0.57

BROWN

0.56

 Browne

0.56

Brow

0.55

 Browns

0.54

Activations Density 0.070%

mentions of the word "Brown" and its variations in various contexts

No Comments

No Known Activations