INDEX

Explanations

mentions or descriptions of pans and cooking actions related to pans

oai_token-act-pair · gpt-3.5-turbo

references to "pan"

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

éĽ

-0.69

éĩ

-0.65

design

-0.64

æ©Ł

-0.64

 Highlander

-0.59

adv

-0.58

IELD

-0.58

 Gilmore

-0.57

 Reserve

-0.55

EAR

-0.55

POSITIVE LOGITS

pan

1.12

orama

1.05

 pans

0.94

handler

0.87

Pan

0.85

 frying

0.82

eful

0.82

efully

0.81

ocular

0.81

omaly

0.80

Activations Density 0.005%

mentions or descriptions of pans and cooking actions related to pans

references to "pan"

No Comments

No Known Activations