INDEX

Explanations

terms related to specific roles, titles, and academic or professional distinctions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

Hochspringen

-0.62

omiast

-0.57

талья

-0.54

our

-0.54

期刊论文

-0.53

出版年

-0.53

°;

-0.52

 arşivlendi

-0.52

}}}}

-0.52

hány

-0.49

POSITIVE LOGITS

 named

0.81

 extraordinaire

0.80

0.79

 olan

0.71

們

0.66

 namens

0.66

 called

0.66

cyklopedia

0.66

ThemeOverlay

0.65

rungsseite

0.64

Activations Density 0.517%

terms related to specific roles, titles, and academic or professional distinctions

No Comments

No Known Activations