INDEX

Explanations

places and locations

oai_token-act-pair · gpt-3.5-turbo

phrases related to locations or places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sqor

-0.77

 PsyNet

-0.71

ernaut

-0.69

ebus

-0.67

 sshd

-0.65

PsyNetMessage

-0.64

bps

-0.64

 Logged

-0.63

telling

-0.62

Enlarge

-0.61

POSITIVE LOGITS

olina

0.66

ova

0.65

igham

0.64

anas

0.63

atri

0.62

gio

0.62

Vas

0.60

 Eucl

0.60

onis

0.59

ocene

0.59

Activations Density 0.295%

places and locations

phrases related to locations or places

No Comments

No Known Activations