INDEX

Explanations

phrases indicating something has not yet happened or been confirmed

oai_token-act-pair · gpt-3.5-turbo

phrases indicating something that has not yet occurred or been completed

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

ï

-0.64

rities

-0.62

warts

-0.58

packs

-0.57

ocker

-0.57

Reporting

-0.56

ufact

-0.56

printf

-0.55

GD

-0.55

ospace

-0.55

POSITIVE LOGITS

TBD

0.66

 succumbed

0.63

orthy

0.62

 unde

0.62

icz

0.61

 surpassed

0.60

 another

0.59

 awaited

0.58

 succumb

0.58

jee

0.58

Activations Density 0.017%

phrases indicating something has not yet happened or been confirmed

phrases indicating something that has not yet occurred or been completed

No Comments

No Known Activations