INDEX

Explanations

references to hay and positive descriptors

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

 NSCoder

-0.64

 Meksi

-0.55

 שוליים

-0.54

 adquis

-0.54

 Erzb

-0.52

 poveznice

-0.52

 Küsten

-0.50

 parha

-0.49

 olvido

-0.49

 Italij

-0.49

POSITIVE LOGITS

 Weather

0.55

 nice

0.54

 factor

0.53

hay

0.52

misc

0.50

 misc

0.47

ip

0.47

factor

0.46

 weather

0.46

Dia

0.46

Activations Density 0.329%

references to hay and positive descriptors

No Comments

No Known Activations