INDEX

Explanations

phrases related to positions or actions of authority

oai_token-act-pair · gpt-3.5-turbo

words related to physical actions or states of being

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

roma

-0.69

PORT

-0.64

 Means

-0.64

ONES

-0.63

ARB

-0.60

Rog

-0.56

 Sciences

-0.56

ACTED

-0.54

 Emerson

-0.53

offer

-0.53

POSITIVE LOGITS

ked

1.36

ky

1.32

king

1.26

kers

1.22

ks

1.11

kered

1.01

ksh

1.00

kish

0.94

ker

0.92

ki

0.90

Activations Density 0.127%

phrases related to positions or actions of authority

words related to physical actions or states of being

No Comments

No Known Activations