INDEX

Explanations

references to global or worldwide contexts and events

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

gerald

-0.85

ippi

-0.85

romancer

-0.81

veyard

-0.80

maid

-0.79

roma

-0.77

alid

-0.75

enhagen

-0.72

neapolis

-0.71

yp

-0.70

POSITIVE LOGITS

 renown

0.86

wide

0.85

 worldwide

0.81

 acclaim

0.81

 warming

0.80

 wide

0.79

 dissemination

0.78

 distribution

0.77

 bestselling

0.73

 premiere

0.73

Activations Density 0.024%

references to global or worldwide contexts and events

No Comments

No Known Activations