INDEX

Explanations

specific locations or places

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to cities and locations

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

leigh

-0.58

bread

-0.58

 Lomb

-0.57

 Burgess

-0.57

 Kash

-0.56

rano

-0.56

 Staten

-0.55

bottom

-0.55

 Greenberg

-0.54

ogether

-0.54

POSITIVE LOGITS

ION

1.18

HI

1.06

IONS

0.99

ONDON

0.95

ANCE

0.94

ING

0.94

ENN

0.93

LIN

0.93

TON

0.92

IST

0.91

Activations Density 0.034%

specific locations or places

proper nouns related to cities and locations

No Comments

No Known Activations