INDEX

Explanations

numeric expressions indicating an increase

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

venge

-0.77

lon

-0.75

rog

-0.72

porter

-0.71

lar

-0.69

notation

-0.68

recy

-0.67

rius

-0.66

matic

-0.66

cape

-0.66

POSITIVE LOGITS

 hours

1.01

 IMAGES

0.99

 acres

0.96

mph

0.94

 years

0.94

 gallons

0.90

 pounds

0.90

 pages

0.90

MPH

0.88

 miles

0.88

Activations Density 0.031%

numeric expressions indicating an increase

No Comments

No Known Activations