INDEX

Explanations

events or appointments indicated by time

oai_token-act-pair · gpt-3.5-turbo

mentions of times, specifically in the format of "pm" followed by numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Hawkins

-0.68

Gro

-0.66

 Mold

-0.62

 Aber

-0.62

 Roots

-0.62

 Shel

-0.61

Cub

-0.61

 Tong

-0.60

Gri

-0.59

Dru

-0.59

POSITIVE LOGITS

pm

3.48

pm

2.19

PM

2.03

PM

1.80

nm

1.29

am

1.23

pc

1.14

gp

1.14

AM

1.08

fm

1.07

Activations Density 0.009%

events or appointments indicated by time

mentions of times, specifically in the format of "pm" followed by numerical values

No Comments

No Known Activations