INDEX

Explanations

references to archives or stored information

oai_token-act-pair · gpt-3.5-turbo

mentions of archives and related archival terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

full

-0.74

liv

-0.73

lying

-0.65

 fish

-0.61

LL

-0.61

soy

-0.61

 medically

-0.60

 mercury

-0.60

 wellness

-0.60

ly

-0.59

POSITIVE LOGITS

 Archive

3.70

 Upload

1.01

archive

1.01

 Archives

0.98

Wire

0.89

 Artifact

0.83

Topic

0.83

 Collections

0.82

 Observatory

0.82

 Agency

0.82

Activations Density 0.011%

references to archives or stored information

mentions of archives and related archival terms

No Comments

No Known Activations