INDEX

Explanations

terms related to stating a situation or condition

oai_token-act-pair · gpt-3.5-turbo

statements challenging conventional economic classifications

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

 Flavoring

-0.59

 watches

-0.58

 dumps

-0.56

 bedrooms

-0.55

velt

-0.55

 sleeps

-0.54

 mattress

-0.53

 fitt

-0.53

AMA

-0.52

 shoots

-0.52

POSITIVE LOGITS

 fraught

0.68

SourceFile

0.60

 daunting

0.60

maxwell

0.58

factor

0.58

requisite

0.57

stated

0.55

@#&

0.55

///

0.55

Activations Density 0.777%

terms related to stating a situation or condition

statements challenging conventional economic classifications

No Comments

No Known Activations