INDEX

Explanations

political statements or commentary

oai_token-act-pair · gpt-3.5-turbo

linguistic markers of emphasis or significance

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

 sighting

-0.85

 bounced

-0.82

 mushroom

-0.80

 recovered

-0.79

 jelly

-0.79

 salv

-0.77

 spotted

-0.77

 intrigued

-0.76

 retri

-0.75

orb

-0.75

POSITIVE LOGITS

ï¸ı

1.24

Therefore

1.15

âĢ

1.08

Democrats

1.06

Republicans

1.05

Unless

1.03

Polit

1.02

Govern

1.02

Americans

1.01

 Governments

0.98

Activations Density 0.307%

political statements or commentary

linguistic markers of emphasis or significance

No Comments

No Known Activations