INDEX

Explanations

references to fire incidents or related emergencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

pes

-0.57

elemField

-0.54

ta

-0.51

енча

-0.49

cef

-0.49

 vacuum

-0.49

 comuniques

-0.48

RegressionTest

-0.46

Ta

-0.46

penup

-0.45

POSITIVE LOGITS

 flames

1.36

 arson

1.26

 fires

1.25

 fire

1.24

 fuoco

1.18

 flame

1.15

 burning

1.11

 embers

1.09

fire

1.04

Fire

1.03

Activations Density 0.174%

references to fire incidents or related emergencies

No Comments

No Known Activations