INDEX

Explanations

references to specific entities or phrases, possibly related to legal or formal terminology

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 vectorielles

-0.86

Portail

-0.81

قایناق‌لار

-0.78

 scolaires

-0.74



-0.73

AlterField

-0.72

 reafon

-0.72

 Negroes

-0.71

AsUp

-0.71

 fédé

-0.71

POSITIVE LOGITS

0.78

0.69

0.55

AddHtmlAttribute

0.54

0.52

]^{-

0.50

vda

0.49

WithDefault

0.49

Activations Density 0.518%

references to specific entities or phrases, possibly related to legal or formal terminology

No Comments

No Known Activations