INDEX

Explanations

phrases related to detailed information or documentation

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the" in relation to various subjects, especially political and administrative contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

antes

-0.71

elaide

-0.67

landers

-0.67

tackle

-0.67

quit

-0.66

uba

-0.66

ava

-0.65

aval

-0.65

ingly

-0.65

avier

-0.64

POSITIVE LOGITS

 workings

1.34

 situation

1.02

 upcoming

0.99

 contents

0.98

 latest

0.95

 process

0.89

 relationship

0.89

 phenomenon

0.89

 impending

0.88

 inner

0.88

Activations Density 0.377%

phrases related to detailed information or documentation

occurrences of the word "the" in relation to various subjects, especially political and administrative contexts

No Comments

No Known Activations