INDEX

Explanations

the word "Fenway"

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Nept

-0.65

 Newsletter

-0.63

EngineDebug

-0.61

 Passenger

-0.60

 Leilan

-0.60

utra

-0.59

eering

-0.59

bel

-0.59

 mater

-0.59

 Orient

-0.55

POSITIVE LOGITS

ruary

1.00

imore

0.85

cers

0.83

actory

0.81

selage

0.80

iesta

0.79

ãĥ¼ãĥĨãĤ£

0.78

Fen

0.74

ornia

0.74

shire

0.73

Activations Density 0.056%

the word "Fenway"

No Comments

No Known Activations