INDEX

Explanations

references to vertical and shade concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

resave

-0.72

ifrance

-0.71

nthetic

-0.57

 fieldNum

-0.54

 Vereins

-0.54

Rgds

-0.53

 endfor

-0.52

lexic

-0.52

Personensuche

-0.52

charAt

-0.51

POSITIVE LOGITS

 Hack

1.05

Hack

1.04

 hack

1.04

hack

1.02

 hacked

0.86

 HACK

0.83

 hackers

0.77

 hacking

0.77

hacks

0.76

 hacker

0.73

Activations Density 0.079%

references to vertical and shade concepts

No Comments

No Known Activations