INDEX

Explanations

words related to misconceptions, false beliefs, and deception

oai_token-act-pair · gpt-3.5-turbo

concepts related to illusions and delusions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

atories

-0.72

RC

-0.69

uter

-0.65

ĵ

-0.65

iary

-0.64

utor

-0.64

ï¸ı

-0.63

RAFT

-0.62

rc

-0.62

received

-0.61

POSITIVE LOGITS

 illusion

3.28

 illusions

2.99

 Illusion

2.32

 delusion

2.02

illusion

1.85

 delusions

1.60

mir

1.31

 hallucinations

1.29

 impressions

1.27

 halluc

1.23

Activations Density 0.029%

words related to misconceptions, false beliefs, and deception

concepts related to illusions and delusions

No Comments

No Known Activations