INDEX

Explanations

terms associated with social commentary and critique, particularly related to identity and action

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

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

 Muses

-0.62

 chapels

-0.60

 Svar

-0.59

 Persians

-0.58

 Platon

-0.57

 ganglion

-0.56

 Pallas

-0.56

 Exacts

-0.56

 liturgical

-0.56

 Dede

-0.55

POSITIVE LOGITS

featureID

0.97

Хьажоргаш

0.88

:✨

0.86

Vidite

0.81

oredCriteria

0.79

 estimés

0.74

queryInterface

0.72

ьаж

0.71

<eos>

0.68

RegressionTest

0.68

Activations Density 0.013%

terms associated with social commentary and critique, particularly related to identity and action

No Comments

No Known Activations