INDEX

Explanations

specific activating sequences of special characters

oai_token-act-pair · gpt-3.5-turbo

emotional expressions and personal reflections

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

 payroll

-0.83

 orchestr

-0.83

 frontline

-0.82

 displacement

-0.80

 phased

-0.80

 funnel

-0.80

 aggreg

-0.79

 dubbed

-0.77

 brid

-0.76

 negoti

-0.75

POSITIVE LOGITS

Anyway

1.48

Advertisements

1.45

Reply

1.41

Reviewer

1.37

âĿ

1.31

RAW

1.30

Loading

1.28

âĶĢâĶĢâĶĢâĶĢ

1.26

Trivia

1.21

--------------------------------------------------------

1.21

Activations Density 0.395%

specific activating sequences of special characters

emotional expressions and personal reflections

No Comments

No Known Activations