INDEX

Explanations

phrases indicating amounts or degrees

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

 Theſe

-0.57

 Chriftian

-0.54

 Schell

-0.53

ngOn

-0.53

 hohe

-0.53

mela

-0.53

 Perſ

-0.53

 whoſe

-0.53

 transfieras

-0.52

<&

-0.52

POSITIVE LOGITS

########.

0.80

FunctionFlags

0.63

mbggenerated

0.58

ValueStyle

0.57

ActionCreators

0.56

 ujednoznacz

0.55

DockStyle

0.55

 drier

0.53

Rüyada

0.53

Revenir

0.53

Activations Density 0.169%

phrases indicating amounts or degrees

No Comments

No Known Activations