INDEX

Explanations

terms related to edges, borders, and boundaries in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

texttt

-0.57

Aiheesta

-0.54

bonucleic

-0.49

 frutti

-0.49

 frutos

-0.49

 costumi

-0.48

LOAT

-0.48

 poids

-0.48

rucks

-0.48

 darunter

-0.48

POSITIVE LOGITS

 edge

1.22

 Edge

1.13

 edges

1.09

edge

1.08

 border

1.02

Edge

1.01

EDGE

0.98

 EDGE

0.98

 Edges

0.96

 fringe

0.94

Activations Density 0.458%

terms related to edges, borders, and boundaries in various contexts

No Comments

No Known Activations