INDEX

Explanations

phrases related to issues, statements, and controversies

oai_token-act-pair · gpt-3.5-turbo 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

 shine

-0.61

 reconstructed

-0.55

 awaited

-0.53

velt

-0.53

 fallout

-0.52

egal

-0.52

 Palestin

-0.51

 sung

-0.51

 benefit

-0.50

rans

-0.50

POSITIVE LOGITS

taboola

0.89

 guiActive

0.74

wheel

0.69

thumbnails

0.66

animous

0.65

 guiActiveUn

0.65

 isEnabled

0.64

river

0.64

 externalToEVAOnly

0.64

assis

0.63

Activations Density 0.017%

phrases related to issues, statements, and controversies

No Comments

No Known Activations