INDEX

Explanations

references to archival content or activities

oai_token-act-pair · gpt-3.5-turbo

references to archives and archival materials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

pher

-0.77

hua

-0.77

yah

-0.75

lied

-0.74

gage

-0.73

nen

-0.72

pheus

-0.70

osuke

-0.70

ya

-0.69

acted

-0.68

POSITIVE LOGITS

 archive

1.05

 archives

1.03

 Archive

0.94

archive

0.86

 repository

0.84

 trove

0.83

conservancy

0.81

 documenting

0.81

 Archives

0.80

 gallery

0.77

Activations Density 0.013%

references to archival content or activities

references to archives and archival materials

No Comments

No Known Activations