INDEX

Explanations

phrases indicating inclusivity or reference to various entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

seamnă

-0.55

 ſtand

-0.51

 ſein

-0.51

ſcher

-0.50

ésultats

-0.50

 ſind

-0.50

iſten

-0.49

 laſſen

-0.49

ientras

-0.49

 indígen

-0.48

POSITIVE LOGITS

0.53

of

0.48

 betweenstory

0.45

 given

0.44

 involved

0.43

 known

0.43

BeginContext

0.41

ArrowToggle

0.40

↵

0.39

Activations Density 1.133%

phrases indicating inclusivity or reference to various entities

No Comments

No Known Activations