INDEX

Explanations

phrases related to law enforcement actions and public reactions

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

quartered

-0.61

 varies

-0.58

 depends

-0.54

 inherently

-0.53

hovah

-0.51

 Historically

-0.51

ould

-0.50

yip

-0.49

icularly

-0.48

 vary

-0.47

POSITIVE LOGITS

 again

0.65

.[

0.64

 anew

0.60

Reloaded

0.57

 thereafter

0.56

quit

0.56

!!!!

0.53

 afterwards

0.53

!!!

0.53

livion

0.52

Activations Density 0.860%

phrases related to law enforcement actions and public reactions

No Comments

No Known Activations