INDEX

Explanations

keywords related to editing or revising content

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Edit" indicating modification or editing actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

milo

-0.99

ciating

-0.97

 pigeon

-0.67

Interstitial

-0.66

gart

-0.65

wagen

-0.65

SIG

-0.61

 behavi

-0.61

 pige

-0.60

gom

-0.59

POSITIVE LOGITS

edit

0.76

 edit

0.73

ril

0.72

 Edit

0.72

 Machina

0.70

ģ«

0.69

edia

0.69

iton

0.68

ulum

0.65

Spoiler

0.63

Activations Density 0.012%

keywords related to editing or revising content

instances of the word "Edit" indicating modification or editing actions

No Comments

No Known Activations