INDEX

Explanations

locations or references to specific states and regions

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

inventoryQuantity

-0.84

potion

-0.74

xual

-0.71

apego

-0.70

azel

-0.70

rodu

-0.68

yrights

-0.68

DIR

-0.67

TAG

-0.66

POSITIVE LOGITS

 economies

0.90

wide

0.90

 nations

0.89

 borders

0.87

 ravaged

0.87

 bordering

0.86

 region

0.86

 governments

0.86

 annexed

0.86

 countries

0.84

Activations Density 0.372%

locations or references to specific states and regions

No Comments

No Known Activations