INDEX

Explanations

time durations specified in years

oai_token-act-pair · gpt-3.5-turbo

references to time durations, specifically involving the word "two" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Caption

-0.75

geist

-0.75

winner

-0.74

bg

-0.68

pse

-0.66

vl

-0.66

Role

-0.65

affer

-0.65

shield

-0.64

grave

-0.63

POSITIVE LOGITS

 weeks

1.57

 months

1.54

 decades

1.40

 years

1.37

 days

1.33

 hundred

1.21

 nights

1.21

 dozen

1.20

 hours

1.16

months

1.16

Activations Density 0.104%

time durations specified in years

references to time durations, specifically involving the word "two" and its variations

No Comments

No Known Activations