INDEX

Explanations

phrases related to providing complete information or full details

oai_token-act-pair · gpt-3.5-turbo

references to comprehensive or complete content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Interstitial

-0.72

abbit

-0.68

apters

-0.67

Wars

-0.67

Pac

-0.65

ucl

-0.63

DERR

-0.63

 Fever

-0.63

estern

-0.62

eers

-0.62

POSITIVE LOGITS

 extent

1.17

erton

1.04

 complement

0.99

 brunt

0.99

ness

0.95

hearted

0.91

 frontal

0.91

 spectrum

0.89

 blown

0.88

 fled

0.85

Activations Density 0.033%

phrases related to providing complete information or full details

references to comprehensive or complete content

No Comments

No Known Activations