INDEX

Explanations

mentions of cognitive states or cognitive abilities

oai_token-act-pair · gpt-3.5-turbo

terminology related to cognitive functions or mental faculties

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

ItemTracker

-0.77

nesses

-0.74

found

-0.70

emale

-0.68

ngth

-0.68

upid

-0.66

Soc

-0.66

cur

-0.65

 satisf

-0.65

pleasant

-0.65

POSITIVE LOGITS

 administration

1.36

 Administration

1.05

Ĥİ

0.87

 admin

0.77

ä½ľ

0.75

 Cheong

0.72

aders

0.69

 Admin

0.68

axis

0.66

ains

0.64

Activations Density 0.000%

mentions of cognitive states or cognitive abilities

terminology related to cognitive functions or mental faculties

No Comments

No Known Activations