INDEX

Explanations

numerical values

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

ãĤ¦ãĤ¹

-0.80

itta

-0.77

advertising

-0.76

advertisement

-0.76

ovember

-0.72

 sugg

-0.71

thinkable

-0.71

pattern

-0.70

urden

-0.70

 compr

-0.70

POSITIVE LOGITS

 Raven

1.06

Ben

1.03

Poe

1.02

 Monte

0.98

Nin

0.96

 Kirin

0.96

Uni

0.94

 Lion

0.94

Ken

0.94

 Chau

0.93

Activations Density 0.511%

numerical values

No Comments

No Known Activations