INDEX

Explanations

numerical values indicating distances and measurements

oai_token-act-pair · gpt-3.5-turbo

numerical values related to measurements or distances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

rics

-0.69

reply

-0.69

 Taxes

-0.68

cash

-0.68

 entitlement

-0.68

Reviewer

-0.67

Crime

-0.64

chart

-0.63

Cash

-0.62

 Accountability

-0.62

POSITIVE LOGITS

 kilometers

1.27

km

1.15

 cubic

1.13

 centimeters

1.11

 kilometres

1.10

 kilomet

1.07

 hectares

1.07

 degrees

1.03

 meters

1.02

 metres

1.01

Activations Density 0.237%

numerical values indicating distances and measurements

numerical values related to measurements or distances

No Comments

No Known Activations