INDEX

Explanations

proper nouns related to specific geographic locations

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

vana

-1.07

istration

-0.72

thood

-0.70

letter

-0.69

rahim

-0.69

wana

-0.68

ogun

-0.67

 recip

-0.67

utations

-0.67

acca

-0.66

POSITIVE LOGITS

 City

1.15

 skyline

1.12

 County

1.08

 Mayor

1.08

 suburb

1.08

 Airport

1.07

 suburbs

1.07

 Heights

1.05

 Metropolitan

1.05

 Municipal

1.04

Activations Density 1.425%

proper nouns related to specific geographic locations

No Comments

No Known Activations