INDEX

Explanations

terms related to political proposals and government actions

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

 partName

-0.89

spot

-0.70

Sym

-0.69

 Correspond

-0.69

OGR

-0.68

 moments

-0.67

Eva

-0.67

mist

-0.65

 Passion

-0.65

 Frie

-0.65

POSITIVE LOGITS

 decriminal

1.31

 legalized

1.19

 amnesty

1.17

 repeal

1.17

 abol

1.16

 privat

1.16

 privatization

1.14

 legislation

1.11

 legalize

1.09

 reforms

1.09

Activations Density 3.394%

terms related to political proposals and government actions

No Comments

No Known Activations