INDEX

Explanations

the start of a new section or topic within a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

og

-0.54

rsiniz

-0.50

 tartalomajánló

-0.49

ara

-0.49

dem

-0.48

 mano

-0.48

dan

-0.47

te

-0.47

res

-0.47

av

-0.47

POSITIVE LOGITS

styleType

1.00

 houſe

0.92

 Majefty

0.89

 pleaſure

0.84

UnusedPrivate

0.80

 ſche

0.79

 ſever

0.78

 ſtate

0.78

 acorns

0.78

 auffi

0.77

Activations Density 0.000%

the start of a new section or topic within a text

No Comments

No Known Activations