INDEX

Explanations

proper nouns related to football players

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

uate

-0.92

uated

-0.87

uates

-0.86

ting

-0.80

WHERE

-0.76

izations

-0.75

uating

-0.70

ted

-0.68

aneously

-0.67

eenth

-0.66

POSITIVE LOGITS

 Rodgers

0.93

mith

0.87

manship

0.79

sey

0.78

 Yards

0.74

0.71

gers

0.71

haw

0.70

ystem

0.69

SON

0.68

Activations Density 0.017%

proper nouns related to football players

No Comments

No Known Activations