INDEX

Explanations

quantitative data about increases or changes, particularly in the context of research and statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

########.

-0.75

conio

-0.67

 kasarigan

-0.53

 estekak

-0.53

chila

-0.52

IntoConstraints

-0.51

ContentLoaded

-0.50

Manbalar

-0.49

Bek

-0.48

qrstuvwxyz

-0.48

POSITIVE LOGITS

 double

2.01

 twice

1.96

double

1.84

 doubled

1.77

 quadruple

1.77

 DOUBLE

1.74

 doubling

1.73

 Twice

1.73

twice

1.71

 triple

1.70

Activations Density 0.655%

quantitative data about increases or changes, particularly in the context of research and statistics

No Comments

No Known Activations