INDEX

Explanations

computer programming-related terms and syntax

oai_token-act-pair · gpt-3.5-turbo

references to numeric values and classification within documents

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

²¾

-0.65

 Proper

-0.62

inctions

-0.61

enko

-0.61

 experien

-0.58

regor

-0.58

 bumper

-0.58

DOC

-0.58

 Able

-0.58

 Portable

-0.56

POSITIVE LOGITS

actionDate

0.81

ItemThumbnailImage

0.76

LP

0.65

soDeliveryDate

0.64

dated

0.62

isSpecialOrderable

0.62

ãĥ¥

0.62

ItemImage

0.60

cerpt

0.58

 Balt

0.58

Activations Density 0.389%

computer programming-related terms and syntax

references to numeric values and classification within documents

No Comments

No Known Activations