INDEX

Explanations

phrases related to statistics, data analysis, and findings

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

eers

-0.83

BIL

-0.73

 Akin

-0.71

arians

-0.70

ARA

-0.66

SPONSORED

-0.64

maid

-0.63

ALLY

-0.62

aris

-0.62

 those

-0.61

POSITIVE LOGITS

 equation

1.32

 spectrum

1.15

 universe

0.98

 continuum

0.95

 puzzle

0.95

 workforce

0.91

 populace

0.91

 population

0.90

 electorate

0.90

 world

0.88

Activations Density 1.231%

phrases related to statistics, data analysis, and findings

No Comments

No Known Activations