INDEX

Explanations

verbs related to doing work or tasks

oai_token-act-pair · gpt-3.5-turbo

references to working or effort-related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 constitu

-0.76

 Bubble

-0.75

ylon

-0.69

anamo

-0.67

 Bour

-0.63

Wars

-0.63

 Unknown

-0.63

 Offic

-0.62

Sov

-0.62

Sov

-0.61

POSITIVE LOGITS

 ethic

1.09

hops

1.01

bench

0.96

horse

0.94

heet

0.88

flows

0.88

manship

0.88

hirt

0.83

work

0.80

mates

0.80

Activations Density 0.065%

verbs related to doing work or tasks

references to working or effort-related activities

No Comments

No Known Activations