INDEX

Explanations

phrases indicating attention or emphasis, usually at the end of a sentence

oai_token-act-pair · gpt-3.5-turbo

sentences that convey definite assertions or conclusions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

gress

-0.79

 constituted

-0.77

 consisted

-0.76

respective

-0.75

éŃĶ

-0.73

withstanding

-0.73

rued

-0.71

lyak

-0.70

ombo

-0.69

velop

-0.68

POSITIVE LOGITS

 Says

1.29

 Plus

1.10

Adds

1.08

 Adds

1.08

 PLUS

1.02

 Meanwhile

1.01

 Photographer

0.90

Enlarge

0.87

 Speaking

0.87

 Anyway

0.87

Activations Density 0.597%

phrases indicating attention or emphasis, usually at the end of a sentence

sentences that convey definite assertions or conclusions

No Comments

No Known Activations