INDEX

Explanations

words related to human perception

oai_token-act-pair · gpt-3.5-turbo

instances of the word "perception" and its variations

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

SU

-0.70

Zip

-0.66

must

-0.65

HOU

-0.64

wic

-0.64

ump

-0.64

 Optional

-0.63

stock

-0.62

 replaced

-0.62

wana

-0.61

POSITIVE LOGITS

 perception

3.58

 perceptions

2.62

 Perception

2.07

 misconception

1.71

 perceived

1.69

 perce

1.68

 perceive

1.62

 perceptual

1.57

 impression

1.55

 illusion

1.43

Activations Density 0.018%

words related to human perception

instances of the word "perception" and its variations

No Comments

No Known Activations