INDEX

Explanations

prepositions followed by locations or directions

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate movement or features associated with geographical boundaries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

adem

-0.71

TY

-0.71

ann

-0.70

uum

-0.69

ulous

-0.69

oric

-0.69

apolis

-0.68

zel

-0.67

ags

-0.65

inguishable

-0.65

POSITIVE LOGITS

side

1.01

 routes

0.92

 Route

0.87

 Interstate

0.87

 highways

0.86

 corridors

0.86

 paths

0.82

 route

0.80

 Aven

0.78

 roads

0.77

Activations Density 0.024%

prepositions followed by locations or directions

phrases that indicate movement or features associated with geographical boundaries

No Comments

No Known Activations