INDEX

Explanations

mentions of time periods, specifically 'weeks'

oai_token-act-pair · gpt-3.5-turbo

mentions of time duration specifically related to weeks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

inav

-0.69

jri

-0.65

ifice

-0.64

 Ital

-0.64

ioch

-0.64

cellence

-0.61

 Crunch

-0.60

inis

-0.59

 VIDEOS

-0.59

Reviewer

-0.58

POSITIVE LOGITS

ago

1.12

Ago

1.10

days

1.02

 weeks

0.98

 gestation

0.84

DAY

0.83

pring

0.83

orrow

0.83

 Weeks

0.80

night

0.79

Activations Density 0.017%

mentions of time periods, specifically 'weeks'

mentions of time duration specifically related to weeks

No Comments

No Known Activations