INDEX

Explanations

specific phrases containing the word "the."

oai_token-act-pair · gpt-3.5-turbo

repeated instances of the word "the."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Joined

-0.81

lier

-0.76

eday

-0.76

eus

-0.71

fulness

-0.70

places

-0.70

packs

-0.70

autions

-0.69

acing

-0.69

fights

-0.66

POSITIVE LOGITS

 dreaded

0.99

 aforementioned

0.92

 latter

0.91

 requisite

0.88

 sexes

0.85

 elusive

0.81

 Racial

0.71

 beloved

0.70

 fetus

0.70

 desired

0.69

Activations Density 0.421%

specific phrases containing the word "the."

repeated instances of the word "the."

No Comments

No Known Activations