INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

mentions of specific geographic locations

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

 Vincent

-0.62

 Spartan

-0.60

 Gast

-0.60

 exhausted

-0.58

var

-0.58

 slim

-0.58

pl

-0.58

 Holly

-0.58

 thin

-0.57

VO

-0.57

POSITIVE LOGITS

abad

5.32

ashtra

1.18

alore

1.16

alam

1.14

Pakistan

1.10

agar

1.08

rael

1.07

ullah

1.05

 Bombay

0.99

 Kabul

0.99

Activations Density 0.012%

locations or places

mentions of specific geographic locations

No Comments

No Known Activations