INDEX

Explanations

negations and expressions of inability or non-existence

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

adpleegd

-0.78

setSource

-0.76

 Thales

-0.75

 merce

-0.75

 Weiss

-0.75

 Gibbs

-0.74

 PACE

-0.73

_('

-0.71

ebe

-0.71

Meyer

-0.71

POSITIVE LOGITS

isn

1.28

__":

1.26

 wasn

1.22

 weren

1.20

 Wasn

1.19

 didn

1.16

Isn

1.15

 aren

1.15

__':

1.14

 mustn

1.14

Activations Density 0.076%

negations and expressions of inability or non-existence

No Comments

No Known Activations