INDEX

Explanations

words related to barriers or obstacles

oai_token-act-pair · gpt-3.5-turbo

references to gatekeeping and barriers in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Hots

-0.77

ortium

-0.75

 Lomb

-0.70

issance

-0.68

 Norm

-0.65

arus

-0.65

ity

-0.63

yles

-0.62

ensional

-0.60

iguous

-0.60

POSITIVE LOGITS

keepers

1.56

keeper

1.52

ways

1.28

keeping

1.16

posts

1.12

house

1.05

way

1.04

fold

1.03

WAYS

0.95

cr

0.95

Activations Density 0.064%

words related to barriers or obstacles

references to gatekeeping and barriers in various contexts

No Comments

No Known Activations