INDEX

Explanations

mentions of the term "Big"

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific entity or organization referred to as "Big" followed by a number

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

istry

-0.83

 confir

-0.76

idency

-0.76

cia

-0.71

theless

-0.70

yrim

-0.68

 autop

-0.66

 muster

-0.64

ILA

-0.63

 polarized

-0.63

POSITIVE LOGITS

gest

1.49

ger

1.26

gie

1.06

gins

1.02

 Brother

0.95

gers

0.93

wig

0.91

glers

0.90

ging

0.88

 Bang

0.85

Activations Density 0.028%

mentions of the term "Big"

mentions of a specific entity or organization referred to as "Big" followed by a number

No Comments

No Known Activations