INDEX

Explanations

terms related to accreditation or approval

oai_token-act-pair · gpt-3.5-turbo

references to accreditation and associated statuses

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

OPLE

-0.82

PT

-0.79

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.79

çīĪ

-0.75

zzo

-0.70

hs

-0.70

åĭ

-0.70

è»

-0.70

DEN

-0.69

geist

-0.69

POSITIVE LOGITS

redited

1.17

ommod

1.15

reditation

1.07

enture

1.04

laim

1.02

ompl

1.00

imate

0.96

uracy

0.94

identally

0.92

urate

0.92

Activations Density 0.008%

terms related to accreditation or approval

references to accreditation and associated statuses

No Comments

No Known Activations