INDEX

Explanations

phrases related to unidentified or unconfirmed entities

oai_token-act-pair · gpt-3.5-turbo

mentions of unidentified or vague entities

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

utic

-0.87

boa

-0.82

odcast

-0.81

acket

-0.81

ongyang

-0.78

©¶æ

-0.75

ushima

-0.73

owder

-0.73

nea

-0.71

ŃĶ

-0.71

POSITIVE LOGITS

 quantity

1.05

 unknown

0.76

 Mortal

0.76

 Unknown

0.71

comings

0.70

 quantities

0.69

ities

0.69

landish

0.68

sburg

0.68

Origin

0.68

Activations Density 0.037%

phrases related to unidentified or unconfirmed entities

mentions of unidentified or vague entities

No Comments

No Known Activations