INDEX

Explanations

terms related to paralysis or being paralyzed

oai_token-act-pair · gpt-3.5-turbo

terms related to paralysis and its effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

rium

-0.78

endum

-0.73

iday

-0.73

ept

-0.72

gard

-0.71

eller

-0.69

gers

-0.69

ayn

-0.68

strom

-0.68

izens

-0.68

POSITIVE LOGITS

 paralyzed

1.41

 paraly

1.24

 paralysis

1.18

 crippled

0.95

 numb

0.93

alysed

0.92

 amput

0.92

 Paraly

0.90

locked

0.86

 polio

0.83

Activations Density 0.020%

terms related to paralysis or being paralyzed

terms related to paralysis and its effects

No Comments

No Known Activations