INDEX

Explanations

locations or proper nouns

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

 Haku

-0.81

 Museum

-0.78

 Auckland

-0.73

stadt

-0.73

 microw

-0.72

 Hungry

-0.71

 Nose

-0.71

oids

-0.71

 Waves

-0.71

POSITIVE LOGITS

per

1.02

pet

0.91

mer

0.91

ãĥ¯

0.90

person

0.87

pe

0.86

PC

0.85

par

0.85

iber

0.84

COM

0.84

Activations Density 0.146%

locations or proper nouns

No Comments

No Known Activations