INDEX

Explanations

phrases related to spatial locations or directions

oai_token-act-pair · gpt-3.5-turbo

references to geographical locations and regional distinctions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

oko

-0.71

SPONSORED

-0.71

fulness

-0.71

enough

-0.70

nance

-0.70

 anew

-0.70

âĢİ

-0.68

bal

-0.67

EVA

-0.66

 humanity

-0.66

POSITIVE LOGITS

 latter

1.25

 periphery

1.17

 upper

1.03

 poorest

1.02

 wealthiest

1.01

 smallest

1.00

 lowest

1.00

 wealthier

1.00

 weakest

0.99

 extremes

0.97

Activations Density 0.294%

phrases related to spatial locations or directions

references to geographical locations and regional distinctions

No Comments

No Known Activations