INDEX

Explanations

time-related phrases mentioning specific hours for monitoring or activity

oai_token-act-pair · gpt-3.5-turbo

references to continuous time availability or coverage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

lda

-0.82

atown

-0.71

 Yankee

-0.63

 Gutenberg

-0.63

ovie

-0.61

enegger

-0.60

 Swanson

-0.60

 Sage

-0.59

 benign

-0.59

folk

-0.58

POSITIVE LOGITS

 hours

1.37

hrs

1.33

 hour

1.32

hours

1.32

 Hours

1.32

 Hour

1.30

hour

1.30

hr

1.18

Hour

1.17

hr

1.12

Activations Density 0.076%

time-related phrases mentioning specific hours for monitoring or activity

references to continuous time availability or coverage

No Comments

No Known Activations