INDEX

Explanations

references to individuals making claims or statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 blames

-0.49

amentlich

-0.47

 blaming

-0.47

 považ

-0.47

 blame

-0.45

 мәкал

-0.43

 betrekking

-0.43

 atribu

-0.43

 wonder

-0.43

 wondering

-0.43

POSITIVE LOGITS

ValueStyle

0.79

OGND

0.58

 intimated

0.56

 implied

0.51

 indicated

0.51

MigrationBuilder

0.51

indicated

0.49

SequentialGroup

0.49

 صوتيه

0.48

 IBOutlet

0.48

Activations Density 0.488%

references to individuals making claims or statements

No Comments

No Known Activations