INDEX

Explanations

mentions of cleaning or removing something

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

yip

-0.99

abul

-0.65

XT

-0.65

akening

-0.64

trl

-0.64

ommod

-0.64

PsyNetMessage

-0.63

arsity

-0.63

Mand

-0.63

MP

-0.62

POSITIVE LOGITS

liness

1.05

 cleaned

0.92

 ashore

0.89

 cleaner

0.88

 linen

0.85

clean

0.85

 towels

0.80

 cleaners

0.80

 toilets

0.79

 stains

0.79

Activations Density 2.821%

mentions of cleaning or removing something

No Comments

No Known Activations