INDEX

Explanations

instances involving the act of seeing or perception

oai_token-act-pair · gpt-3.5-turbo

instances of the word "seeing."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

cult

-0.79

prime

-0.78

 whistle

-0.76

trap

-0.72

cussion

-0.65

cia

-0.63

sett

-0.63

EStreamFrame

-0.62

 negotiator

-0.61

locked

-0.61

POSITIVE LOGITS

awed

0.81

seeing

0.77

enance

0.74

 visions

0.71

ometimes

0.71

 mete

0.71

agame

0.70

 seeing

0.68

 them

0.68

eye

0.67

Activations Density 0.013%

instances involving the act of seeing or perception

instances of the word "seeing."

No Comments

No Known Activations