INDEX

Explanations

terms related to the process of editing, especially editing software and equipment

oai_token-act-pair · gpt-3.5-turbo

instances of the word "editing."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hood

-0.83

ptoms

-0.78

bley

-0.74

ortium

-0.72

fall

-0.72

bands

-0.71

 Huck

-0.71

fal

-0.70

FE

-0.70

Ã£o

-0.69

POSITIVE LOGITS

 editing

1.06

 edits

0.97

 annex

0.86

 edit

0.85

 Editing

0.78

 suite

0.77

 cursor

0.76

ettings

0.72

 supervisor

0.71

 deleting

0.71

Activations Density 0.017%

terms related to the process of editing, especially editing software and equipment

instances of the word "editing."

No Comments

No Known Activations