INDEX

Explanations

instances where a range or maximum limit is mentioned

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a range or limit

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Released

-0.78

 consulted

-0.75

 headquartered

-0.72

acons

-0.70

fecture

-0.70

tymology

-0.70

 governs

-0.68

achy

-0.66

PDATE

-0.66

 Cosponsors

-0.66

POSITIVE LOGITS

 pasture

0.72

 shelves

0.68

 Siberian

0.66

 seams

0.65

 electrodes

0.65

wards

0.65

 Abbey

0.64

othal

0.64

 stairs

0.63

 Siberia

0.63

Activations Density 0.564%

instances where a range or maximum limit is mentioned

phrases indicating a range or limit

No Comments

No Known Activations