INDEX

Explanations

dates written in a specific format: month, day, and year, followed by a comma

oai_token-act-pair · gpt-3.5-turbo

occurrences of commas in the text

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

 corridors

-0.53

 tailor

-0.53

etheless

-0.53

 ageing

-0.52

 exha

-0.51

 marqu

-0.50

 planes

-0.50

 conscientious

-0.50

 glac

-0.49

 secretaries

-0.48

POSITIVE LOGITS

1.38

1.32

1.31

1.30

1.24

1.17

1.16

1.07

1.06

1.05

Activations Density 0.031%

dates written in a specific format: month, day, and year, followed by a comma

occurrences of commas in the text

No Comments

No Known Activations