INDEX

Explanations

phrases starting with "These are" or "Those are" followed by descriptions or lists

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the presence or importance of certain subjects or topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

etheless

-0.85

enium

-0.76

cation

-0.76

leneck

-0.74

mentation

-0.73

ledge

-0.70

oyer

-0.69

rification

-0.69

osate

-0.69

culosis

-0.68

POSITIVE LOGITS

 constants

0.80

 examples

0.79

 truths

0.74

 acron

0.73

wolf

0.72

types

0.71

hots

0.70

 indications

0.69

 representations

0.69

 reminders

0.69

Activations Density 0.080%

phrases starting with "These are" or "Those are" followed by descriptions or lists

phrases indicating the presence or importance of certain subjects or topics

No Comments

No Known Activations