INDEX

Explanations

positive adjectives indicating approval or admiration

oai_token-act-pair · gpt-3.5-turbo

words and phrases that convey positive sentiments or aesthetic appreciation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

casters

-0.77

eds

-0.76

bands

-0.76

agents

-0.76

puters

-0.76

eligible

-0.75

letters

-0.75

asters

-0.74

llers

-0.73

gans

-0.72

POSITIVE LOGITS

 blend

1.09

 distinction

1.08

 combination

1.08

 reminder

1.05

 twist

1.05

 assortment

1.03

 shade

1.02

 antidote

0.99

mix

0.99

 mixture

0.98

Activations Density 0.171%

positive adjectives indicating approval or admiration

words and phrases that convey positive sentiments or aesthetic appreciation

No Comments

No Known Activations