INDEX

Explanations

references to archives or archiving activities

oai_token-act-pair · gpt-3.5-turbo

references to archives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

-0.80

lied

-0.73

ndra

-0.73

hua

-0.72

ptive

-0.71

nen

-0.70

gage

-0.69

yah

-0.68

laws

-0.68

pheus

-0.67

POSITIVE LOGITS

 archive

1.12

 archives

1.11

 Archive

1.02

archive

0.91

 repository

0.84

 trove

0.84

 gallery

0.80

 repositories

0.77

 Archives

0.75

conservancy

0.74

Activations Density 0.006%

references to archives or archiving activities

references to archives

No Comments

No Known Activations