INDEX

Explanations

titles or proper names

oai_token-act-pair · gpt-3.5-turbo

significant literary or cinematic references

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

 Titans

-0.93

 Jets

-0.82

tnc

-0.81

 Jordan

-0.77

 Robb

-0.76

 Kass

-0.74

 Debbie

-0.73

 flame

-0.73

 Brad

-0.72

 Tennessee

-0.72

POSITIVE LOGITS

Â

2.34

Â

2.24

Â²

1.94

Ãį

1.83

Ã

1.81

Ã¢

1.75

Â¨

1.74

Ãĥ

1.70

Ã

1.69

Ãī

1.62

Activations Density 0.227%

titles or proper names

significant literary or cinematic references

No Comments

No Known Activations