INDEX

Explanations

phrases indicating large quantities or populations

oai_token-act-pair · gpt-4o-mini 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

ãĥ¤

-0.71

agate

-0.67

 Redemption

-0.66

Delivery

-0.64

iban

-0.63

Interstitial

-0.63

mentioned

-0.62

 Armory

-0.61

Gay

-0.58

 menace

-0.58

POSITIVE LOGITS

aires

0.99

aire

0.92

 dollars

0.86

 ILCS

0.84

dollar

0.82

 upon

0.78

of

0.78

undo

0.77

ourcing

0.77

ourced

0.74

Activations Density 0.042%

phrases indicating large quantities or populations

No Comments

No Known Activations