INDEX

Explanations

phrases that indicate purpose or reason for actions

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_16/width_16k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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



-0.33

 socios

-0.32

Wearing

-0.32

 presently

-0.30

Deaths

-0.30

 Niedersch

-0.30

++];

-0.30

Studies

-0.30

 retrait

-0.30

 Theile

-0.29

POSITIVE LOGITS

تقاوى

0.70

 يتيمه

0.60

 reasons

0.59

 suốt

0.59

pentru

0.56

addPreferredGap

0.55

 VOOR

0.55

SequentialGroup

0.54

For

0.53

 sake

0.53

Activations Density 0.281%

phrases that indicate purpose or reason for actions

No Comments

No Known Activations