INDEX

Explanations

phrases indicating the continuation or interruption of a story or article

oai_token-act-pair · gpt-3.5-turbo

repeated phrases indicating continuation in text

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

bats

-0.79

wit

-0.70

onomic

-0.69

onom

-0.66

liest

-0.65

Bey

-0.65

ogy

-0.64

haps

-0.64

aclysm

-0.62

lication

-0.61

POSITIVE LOGITS

 BELOW

0.74

 sclerosis

0.70

CLASSIFIED

0.64

 trending

0.63

...]

0.63

 Continued

0.63

 enlarge

0.62

 scrolling

0.61

 PARK

0.60

 CONTIN

0.59

Activations Density 0.011%

phrases indicating the continuation or interruption of a story or article

repeated phrases indicating continuation in text

No Comments

No Known Activations