INDEX

Explanations

phrases describing transformation or conversion of something into a different form or state, often emphasizing a positive outcome

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "turned into"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inately

-0.77

yright

-0.77

ritz

-0.76

cies

-0.73

cation

-0.71

orno

-0.68

gang

-0.66

anwhile

-0.64

condition

-0.64

igent

-0.63

POSITIVE LOGITS

Ò

0.75

 adulthood

0.73

 usable

0.72

ãĤ§

0.71

 ashes

0.64

 something

0.64

pload

0.63

 insign

0.63

 unman

0.62

 Valhalla

0.62

Activations Density 0.055%

phrases describing transformation or conversion of something into a different form or state, often emphasizing a positive outcome

instances of the phrase "turned into"

No Comments

No Known Activations