INDEX

Explanations

phrases related to specific actions or achievements

oai_token-act-pair · gpt-3.5-turbo

references to the actions of various groups or individuals

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

fml

-0.81

meat

-0.74

/"

-0.72

forge

-0.72

>(

-0.72

iHUD

-0.68

=#

-0.68

bender

-0.67

wcsstore

-0.65

................................................................

-0.65

POSITIVE LOGITS

 others

1.02

 those

0.96

 individuals

0.95

 whoever

0.87

 politicians

0.87

 minorities

0.82

 fellow

0.82

 scholars

0.82

 countless

0.81

 passers

0.80

Activations Density 0.262%

phrases related to specific actions or achievements

references to the actions of various groups or individuals

No Comments

No Known Activations