INDEX

Explanations

references to quantities or numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 čierna

-0.72

 greateſt

-0.72

 ſeveral

-0.71

ngiliz

-0.70

 feveral

-0.70

 rootReducer

-0.69

plass

-0.69

 hinweg

-0.68

 Monfieur

-0.68

 superiori

-0.68

POSITIVE LOGITS

ten

0.73

rungsseite

0.69

TestingModule

0.65

 jScrollPane

0.63

Ten

0.62

libft

0.60

 disagree

0.59

tro

0.57

glie

0.57

latch

0.56

Activations Density 0.176%

references to quantities or numerical values

No Comments

No Known Activations