INDEX

Explanations

mentions of a specific term or acronym represented by "yp."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "type" and related variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Catalyst

-0.72

DERR

-0.71

FUL

-0.70

VERT

-0.66

Ou

-0.65

fully

-0.65

renheit

-0.65

 UNITED

-0.63

 congr

-0.62

gaard

-0.62

POSITIVE LOGITS

onent

0.98

aired

0.95

posium

0.95

utic

0.94

olicy

0.93

yp

0.90

ocalypse

0.90

olit

0.90

amph

0.87

ublic

0.87

Activations Density 0.008%

mentions of a specific term or acronym represented by "yp."

occurrences of the word "type" and related variations

No Comments

No Known Activations