INDEX

Explanations

phrases related to close examination or scrutiny

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

ICAN

-0.80

AIN

-0.75

NZ

-0.74

urers

-0.67

ains

-0.64

 Pengu

-0.64

uria

-0.63

onso

-0.63

alian

-0.61

 Generator

-0.61

POSITIVE LOGITS

 proximity

1.20

sighted

0.94

 resemblance

0.92

 confinement

0.87

 vicinity

0.86

 relatives

0.86

 scrutiny

0.85

 shave

0.84

 confines

0.83

 ties

0.81

Activations Density 1.312%

phrases related to close examination or scrutiny

No Comments

No Known Activations