INDEX

Explanations

keywords related to programming concepts and code syntax

oai_token-act-pair · gpt-3.5-turbo

programming-related terms and variable declarations

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

 charism

-0.68

eches

-0.65

eers

-0.64

olls

-0.59

aucuses

-0.58

aws

-0.57

eatures

-0.57

zers

-0.56

 Marlins

-0.56

ashes

-0.55

POSITIVE LOGITS

âĪ

0.88

0.81

0.72

âī

0.72

 implies

0.70

0.69

 âī

0.67

vs

0.67

 denotes

0.66

Activations Density 0.430%

keywords related to programming concepts and code syntax

programming-related terms and variable declarations

No Comments

No Known Activations