INDEX

Explanations

words related to technology products and brands

oai_token-act-pair · gpt-3.5-turbo

brand names and product categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

population

-0.74

immigration

-0.73

immigrant

-0.73

paying

-0.71

public

-0.70

 unwelcome

-0.68

continue

-0.68

 vanquished

-0.66

soDeliveryDate

-0.65

shall

-0.65

POSITIVE LOGITS

Â®

1.27

âĦ¢

1.11

XT

1.07

SX

1.06

 Precision

1.04

Â®,

1.03

 Mini

1.01

 Nano

1.01

 Vapor

1.00

 Electronics

0.99

Activations Density 0.297%

words related to technology products and brands

brand names and product categories

No Comments

No Known Activations