INDEX

Explanations

time-related information, specifically phrases indicating the time of day or specific times

oai_token-act-pair · gpt-3.5-turbo

references to time, specifically terms indicating the timing of events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

acco

-0.67

Reviewer

-0.61

 Inher

-0.61

 heirs

-0.61

 Primordial

-0.58

asis

-0.58

CoC

-0.57

 Dating

-0.57

 Experience

-0.57

anty

-0.56

POSITIVE LOGITS

 afternoon

1.15

 morning

1.09

 evening

1.08

morning

1.07

week

1.03

 mornings

1.02

 Thursday

1.01

 Wednesday

0.99

aturdays

0.98

ursday

0.98

Activations Density 0.154%

time-related information, specifically phrases indicating the time of day or specific times

references to time, specifically terms indicating the timing of events

No Comments

No Known Activations