INDEX

Explanations

phrases related to praise or honoring something or someone

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "glory."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

BOOK

-0.91

strong

-0.75

artment

-0.72

ANY

-0.68

perature

-0.65

leneck

-0.65

bern

-0.64

chens

-0.63

SEA

-0.62

heter

-0.62

POSITIVE LOGITS

 glory

1.03

fully

0.84

ous

0.82

ful

0.81

eous

0.79

ifully

0.78

 pageant

0.76

 deeds

0.75

 fide

0.75

antly

0.75

Activations Density 0.022%

phrases related to praise or honoring something or someone

references to the concept of "glory."

No Comments

No Known Activations