INDEX

Explanations

complex concepts or items

oai_token-act-pair · gpt-3.5-turbo

references to complexity in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

chance

-0.75

agues

-0.74

 Loot

-0.68

 Attend

-0.66

 Bers

-0.64

 Printed

-0.64

 Backup

-0.64

ures

-0.64

ileaks

-0.64

 Editors

-0.63

POSITIVE LOGITS

 complex

3.72

complex

2.78

 Complex

2.56

 complexes

2.34

 complicated

2.24

 complexity

1.68

 complexities

1.67

 sophisticated

1.58

 intricate

1.54

 nuanced

1.47

Activations Density 0.025%

complex concepts or items

references to complexity in various contexts

No Comments

No Known Activations