INDEX

Explanations

the start of a document or section, indicated by specific token markers

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

-0.79

-0.61

 […]

-0.56

-0.55

…

-0.54

ज़

-0.52

-0.51

occ

-0.51

↵

-0.49

POSITIVE LOGITS

 ModelExpression

0.83

__':

0.79

 Audiodateien

0.79

 autorytatywna

0.79

tvguidetime

0.76

 maxn

0.75

archiviato

0.75

rungsseite

0.73

URLException

0.73

].)

0.72

Activations Density 0.022%

the start of a document or section, indicated by specific token markers

No Comments

No Known Activations