INDEX

Explanations

names of individuals and their occupations

oai_token-act-pair · gpt-3.5-turbo

subjects described by the word "who."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Bound

-0.74

ÂłÂłÂłÂł

-0.72

 Anything

-0.70

 Basics

-0.69

 VIDEOS

-0.66

 Processing

-0.65

 Delicious

-0.65

BACK

-0.64

 Checking

-0.63

 Around

-0.63

POSITIVE LOGITS

 accompanies

1.02

soever

0.99

oping

0.99

 attended

0.99

 attends

0.98

 oversaw

0.98

 preceded

0.97

 frequ

0.95

 resided

0.92

 presided

0.90

Activations Density 0.140%

names of individuals and their occupations

subjects described by the word "who."

No Comments

No Known Activations