INDEX

Explanations

structural elements or tags in data representations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

}{@

-0.65

principalColumn

-0.64

 onCancelled

-0.63

isti

-0.60

⪜

-0.59

verifyException

-0.59

astify

-0.57

picasso

-0.57

$_(

-0.57

 InputDecoration

-0.55

POSITIVE LOGITS

0.55

0.50

<h2>

0.47

↵↵↵↵↵

0.35

0.34

 cerrada

0.33

↵↵↵↵

0.33

showAll

0.33

QMetaType

0.32

Activations Density 0.536%

structural elements or tags in data representations

No Comments

No Known Activations