INDEX

Explanations

words related to paranoia, such as 'paranoid' and 'delusions'

oai_token-act-pair · gpt-3.5-turbo

terms related to paranoia and obsessive thoughts

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

hire

-0.97

ains

-0.93

ained

-0.80

adr

-0.77

inctions

-0.76

aver

-0.74

hops

-0.74

apers

-0.72

aining

-0.68

eric

-0.68

POSITIVE LOGITS

 schizophren

1.31

 delusions

1.06

 paranoid

1.04

 schizophrenia

0.99

 paranoia

0.98

 fantasies

0.80

 psychosis

0.79

ulent

0.77

 tendencies

0.75

 insanity

0.75

Activations Density 0.015%

words related to paranoia, such as 'paranoid' and 'delusions'

terms related to paranoia and obsessive thoughts

No Comments

No Known Activations