INDEX

Explanations

words related to time and sequencing

oai_token-act-pair · gpt-3.5-turbo

the end of text markers indicating the completion of thought or content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ãĤ§

-0.75

advertisement

-0.75

arse

-0.73

ascus

-0.68

ãĤ´ãĥ³

-0.67

inav

-0.65

////////////////

-0.63

iggurat

-0.62

Rated

-0.62

displayText

-0.62

POSITIVE LOGITS

 Britons

0.74

 Americans

0.72

 respondents

0.70

 archaeologists

0.68

 historians

0.68

 politicians

0.68

 anyone

0.67

 people

0.67

 voters

0.67

 ISPs

0.67

Activations Density 0.606%

words related to time and sequencing

the end of text markers indicating the completion of thought or content

No Comments

No Known Activations