INDEX

Explanations

phrases or words related to porridge

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "por."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 GOODMAN

-0.76

ITNESS

-0.73

enegger

-0.73

INESS

-0.69

 Dickinson

-0.67

 sorts

-0.66

schild

-0.65

 Finder

-0.65

ITED

-0.64

 Wanted

-0.64

POSITIVE LOGITS

cel

1.04

folios

1.02

cup

1.01

oshenko

0.98

agraph

0.95

folio

0.95

por

0.94

quet

0.94

cus

0.93

ters

0.91

Activations Density 0.012%

phrases or words related to porridge

mentions of the word "por."

No Comments

No Known Activations