INDEX

Explanations

personal pronouns, specifically focusing on the perspective of the speaker

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

CloseOperation

-0.90

檚

-0.79

Eſ

-0.78

Personensuche

-0.76

 serem

-0.75

 pinulongan

-0.74

Efq

-0.70

WithTag

-0.69

 terem

-0.66

localctx

-0.65

POSITIVE LOGITS

am

1.51

 myself

1.22

 have

1.10

can

1.08

don

0.96

 cannot

0.93

 will

0.90

myself

0.85

do

0.82

 believe

0.82

Activations Density 0.238%

personal pronouns, specifically focusing on the perspective of the speaker

No Comments

No Known Activations