INDEX

Explanations

comparative adverbs indicating a higher degree or intensity

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

interstitial

-0.71

DragonMagazine

-0.67

 Difference

-0.66

owe

-0.62

ļé

-0.60

advertising

-0.60

IP

-0.59

 DRAG

-0.59

 Editorial

-0.57

ilts

-0.57

POSITIVE LOGITS

 than

1.31

 closely

1.24

 efficiently

1.21

 broadly

1.16

 accurately

1.14

 aggressively

1.12

 prominently

1.07

 importantly

1.06

 tightly

1.05

 harshly

1.04

Activations Density 0.086%

comparative adverbs indicating a higher degree or intensity

No Comments

No Known Activations