INDEX

Explanations

names or terms related to legal and technical documents

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to specific actions or key items involved in processes or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

owship

-0.76

andro

-0.73

ukemia

-0.72

oti

-0.69

incoln

-0.69

arov

-0.68

iger

-0.68

ync

-0.68

 suspic

-0.68

evil

-0.67

POSITIVE LOGITS

cies

0.89

 Flavoring

0.75

itatively

0.73

hemy

0.73

itarian

0.73

oldown

0.72

RPM

0.69

CLR

0.67

culosis

0.67

isks

0.67

Activations Density 0.022%

names or terms related to legal and technical documents

words and phrases related to specific actions or key items involved in processes or conditions

No Comments

No Known Activations