INDEX

Explanations

phrases or descriptions related to someone's character or impact

oai_token-act-pair · gpt-3.5-turbo

articles followed by descriptors or attributes related to individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

anism

-0.80

Edit

-0.77

otte

-0.72

AIDS

-0.67

events

-0.67

 Include

-0.67

querade

-0.67

Access

-0.66

mares

-0.65

views

-0.65

POSITIVE LOGITS

 terrific

1.30

 fantastic

1.16

tad

1.15

 wonderful

1.14

 fascinating

1.11

 formidable

1.06

 delightful

1.06

 great

1.05

bit

1.04

 decent

1.04

Activations Density 0.214%

phrases or descriptions related to someone's character or impact

articles followed by descriptors or attributes related to individuals

No Comments

No Known Activations