INDEX

Explanations

technical terms related to data analysis and methodology

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apest

-0.83

eday

-0.69

ogram

-0.66

udeau

-0.66

bitious

-0.65

 honoured

-0.64

icipated

-0.64

ennial

-0.63

 Tradable

-0.62

arcity

-0.62

POSITIVE LOGITS

 balcon

0.77

ifice

0.74

hyde

0.74

 compartment

0.72

 workings

0.72

 boundaries

0.67

cers

0.67

 comfort

0.67

ĸļ

0.67

 cavern

0.67

Activations Density 0.166%

technical terms related to data analysis and methodology

No Comments

No Known Activations