INDEX

Explanations

variables and function definitions related to data handling or processing

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

ſelves

-0.82

DockStyle

-0.79

@"/

-0.75

 Monfieur

-0.73

 Roskov

-0.71

 esternos

-0.71

tvguidetime

-0.71

 kasarigan

-0.70

 Taktlose

-0.70

 ſtate

-0.68

POSITIVE LOGITS

0.33

Slf

0.33

0.31

ha

0.29

0.28

0.27

 that

0.27

Activations Density 0.011%

variables and function definitions related to data handling or processing

No Comments

No Known Activations