INDEX

Explanations

references to inner workings or internal aspects of people, systems, or organizations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orthy

-0.84

eday

-0.82

atoes

-0.81

dayName

-0.80

essors

-0.77

ILLE

-0.74

enegger

-0.74

nant

-0.73

enance

-0.72

HAHAHAHA

-0.72

POSITIVE LOGITS

most

1.28

 workings

1.22

 combustion

0.89

 sanct

0.84

 circle

0.81

 Mongolia

0.78

circle

0.78

wear

0.76

ranean

0.75

 turmoil

0.75

Activations Density 0.006%

references to inner workings or internal aspects of people, systems, or organizations

No Comments

No Known Activations