INDEX

Explanations

proper nouns

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

mson

-0.76

ngth

-0.74

hower

-0.72

perse

-0.68

RNA

-0.63

åŃ

-0.63

æĸ¹

-0.61

DragonMagazine

-0.61

selves

-0.61

REDACTED

-0.60

POSITIVE LOGITS

jamin

1.56

nington

1.08

oit

1.03

cher

0.93

ghazi

0.93

chers

0.92

ches

0.87

itude

0.84

ning

0.84

imaru

0.83

Activations Density 0.537%

proper nouns

No Comments

No Known Activations