INDEX

Explanations

time-related terms indicating the duration of events

oai_token-act-pair · gpt-3.5-turbo

references to durations of time, particularly involving months and years

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

itably

-0.75

tainment

-0.68

 Cosponsors

-0.62

 behalf

-0.59

EMBER

-0.58

 Important

-0.57

BuyableInstoreAndOnline

-0.56

versely

-0.56

York

-0.56

rone

-0.55

POSITIVE LOGITS

of

1.02

 spent

0.97

 hiatus

0.92

 slog

0.89

 deliber

0.83

ynthesis

0.80

 langu

0.79

ourcing

0.76

 without

0.76

 hiber

0.74

Activations Density 0.084%

time-related terms indicating the duration of events

references to durations of time, particularly involving months and years

No Comments

No Known Activations