INDEX

Explanations

phrases related to international diplomacy and negotiations, particularly regarding Iran's nuclear program

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

 LAPD

-0.77

 Trafford

-0.74

 Chest

-0.72

iatrics

-0.70

 Everton

-0.69

ãĤī

-0.68

 Dover

-0.68

 Southampton

-0.68

 Nottingham

-0.67

PASS

-0.67

POSITIVE LOGITS

 Hussein

0.92

enei

0.90

pour

0.89

 regime

0.88

 retali

0.85

 dissidents

0.84

 regimes

0.79

 dictator

0.79

 missile

0.76

ollah

0.75

Activations Density 0.614%

phrases related to international diplomacy and negotiations, particularly regarding Iran's nuclear program

No Comments

No Known Activations