INDEX

Explanations

phrases and descriptors related to authenticity and truthfulness

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

rungsseite

-1.04

DockStyle

-1.00

principalTable

-0.90

majánló

-0.83

 vooz

-0.82

WithIOException

-0.82

migrationBuilder

-0.78

parsedMessage

-0.77

 miniaturka

-0.77

 laſſen

-0.76

POSITIVE LOGITS

ness

0.48

 adjective

0.45

 adjectives

0.34

 makeStyles

0.31

</strong>

0.31

hot

0.30

 qualities

0.30

adjective

0.30

 dull

0.30

 characteristics

0.30

Activations Density 0.036%

phrases and descriptors related to authenticity and truthfulness

No Comments

No Known Activations