INDEX

Explanations

instances of directional terms referring to physical positions

oai_token-act-pair · gpt-3.5-turbo

references to positional phrases in relation to objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

eers

-0.73

BIL

-0.71

fulness

-0.70

abilities

-0.69

ADA

-0.67

forts

-0.65

eday

-0.64

HCR

-0.64

NPR

-0.64

iors

-0.63

POSITIVE LOGITS

 globe

1.03

 spectrum

1.02

 equation

0.95

 circle

0.93

 compass

0.92

 continuum

0.91

 rectangle

0.88

 peninsula

0.88

 continent

0.88

 pyramid

0.87

Activations Density 0.143%

instances of directional terms referring to physical positions

references to positional phrases in relation to objects

No Comments

No Known Activations