INDEX

Explanations

references to political events and laws

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

ragon

-0.70

etime

-0.65

 estab

-0.63

cko

-0.62

Otherwise

-0.62

eways

-0.62

 sidx

-0.61

hesda

-0.61

é¾į

-0.61

TeX

-0.60

POSITIVE LOGITS

 these

0.74

 this

0.73

these

0.66

it

0.63

 respectively

0.63

 Garland

0.61

features

0.61

this

0.61

 however

0.59

angular

0.58

Activations Density 0.306%

references to political events and laws

No Comments

No Known Activations