INDEX

Explanations

references to descriptions in a data schema or metadata context

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_32k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.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

">'.$

-0.89

 Theſe

-0.82

__*/

-0.82

 étoient

-0.82

 uſed

-0.81

Efq

-0.79

;">

-0.78

'>

-0.77

$_"

-0.77

='';

-0.75

POSITIVE LOGITS

Description

1.00

description

0.97

 Description

0.88

DESCRIPTION

0.88

 description

0.88

getDescription

0.84

 getDescription

0.83

setDescription

0.78

 DESCRIPTION

0.67

 Mendez

0.64

Activations Density 0.029%

references to descriptions in a data schema or metadata context

No Comments

No Known Activations