INDEX

Explanations

updates or revisions made to a text or document

oai_token-act-pair · gpt-3.5-turbo

instances of the word "updated."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

aden

-0.93

avery

-0.86

vous

-0.82

odor

-0.76

ang

-0.76

ain

-0.75

 Fein

-0.74

zees

-0.73

verning

-0.73

mite

-0.73

POSITIVE LOGITS

 updates

0.91

 update

0.86

 notifications

0.83

 updating

0.80

 snapshots

0.78

 refresh

0.76

 versions

0.75

 visuals

0.74

 notification

0.73

 formulations

0.71

Activations Density 0.024%

updates or revisions made to a text or document

instances of the word "updated."

No Comments

No Known Activations