INDEX

Explanations

references to Wikipedia and related wiki content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 يتيمه

-0.71

الإنجليزية

-0.64

⟬

-0.58

oporosis

-0.54

AsUp

-0.53

 Bewußt

-0.53

 mijne

-0.52

rungsseite

-0.52

majánló

-0.52

genodigd

-0.51

POSITIVE LOGITS

Đ

0.55

Tak

0.54

FX

0.50

Tak

0.49

GS

0.49

zed

0.48

ა

0.48

tak

0.48

Đ

0.47

GS

0.46

Activations Density 0.281%

references to Wikipedia and related wiki content

No Comments

No Known Activations