INDEX

Explanations

references to Canadian identity and associated terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 sterling

-0.70

 quarters

-0.65

 bleeding

-0.63

assetsadobe

-0.60

 whit

-0.60

 nightly

-0.58

 contention

-0.57

bush

-0.56

 Hallow

-0.56

 Cheong

-0.56

POSITIVE LOGITS

iox

0.87

ue

0.81

cially

0.80

illo

0.79

icip

0.79

abis

0.78

ortun

0.76

ensis

0.76

uel

0.76

ibly

0.75

Activations Density 0.005%

references to Canadian identity and associated terms

No Comments

No Known Activations