INDEX

Explanations

proper nouns related to specific events, people, or entities

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names associated with notable figures or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 sqor

-0.66

interstitial

-0.65

ãĥ³ãĤ¸

-0.63

detail

-0.62

 upside

-0.61

cffffcc

-0.60

ALSE

-0.57

bryce

-0.57

itability

-0.57

conservancy

-0.56

POSITIVE LOGITS

 began

1.28

 arrived

1.21

 became

1.20

 came

1.16

 debuted

1.15

 transitioned

1.08

 took

1.07

 disappeared

1.07

 commenced

1.06

 ceased

1.06

Activations Density 0.375%

proper nouns related to specific events, people, or entities

proper nouns and names associated with notable figures or entities

No Comments

No Known Activations