INDEX

Explanations

dates represented in a two-digit format

oai_token-act-pair · gpt-3.5-turbo

sequences of numbers and numerical identifiers

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

olicy

-0.89

 tremend

-0.82

isode

-0.78

atre

-0.78

pherd

-0.77

uppet

-0.74

etsk

-0.74

onse

-0.74

achu

-0.72

acle

-0.72

POSITIVE LOGITS

1.31

1.21

th

0.91

teen

0.88

0.85

0.84

0.83

0.82

66666666

0.81

Activations Density 0.041%

dates represented in a two-digit format

sequences of numbers and numerical identifiers

No Comments

No Known Activations