INDEX

Explanations

phrases related to completing tasks efficiently and effectively

oai_token-act-pair · gpt-3.5-turbo

references to jobs and tasks being completed or managed

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

Chance

-0.63

 incorpor

-0.63

 cameo

-0.61

ewitness

-0.61

 Dwell

-0.61

 commenter

-0.59

arij

-0.59

BN

-0.58

lime

-0.57

 plea

-0.56

POSITIVE LOGITS

 efficiently

1.62

 smoothly

1.32

 safely

1.30

 properly

1.28

 correctly

1.22

 effectively

1.13

 cheaply

1.08

 reliably

1.08

 satisf

1.05

 without

1.04

Activations Density 0.574%

phrases related to completing tasks efficiently and effectively

references to jobs and tasks being completed or managed

No Comments

No Known Activations