INDEX

Explanations

references to methodologies or strategies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/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.21.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

 muebles

-0.48

 zelve

-0.47

 statistique

-0.46

 goederen

-0.46

-0.45

 cadmium

-0.45

OrEmpty

-0.45

gemaakt

-0.45

 Bakter

-0.44

 gesteld

-0.44

POSITIVE LOGITS

 approach

1.40

approach

1.34

 Approach

1.33

 APPROACH

1.27

Approach

1.22

 approaches

1.16

 Approaches

1.01

 approached

0.98

 approche

0.96

approche

0.95

Activations Density 0.077%

references to methodologies or strategies

No Comments

No Known Activations