INDEX

Explanations

numbers related to statistics or quantities

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

 Despair

-0.69

dit

-0.67

imar

-0.66

lace

-0.64

axter

-0.63

ETHOD

-0.61

Materials

-0.60

sbm

-0.59

à©

-0.59

INTON

-0.59

POSITIVE LOGITS

plates

0.84

number

0.80

enance

0.80

 crunch

0.78

 numbers

0.77

 number

0.76

plate

0.75

of

0.73

metry

0.73

ginx

0.71

Activations Density 0.031%

numbers related to statistics or quantities

No Comments

No Known Activations