INDEX

Explanations

mentions of the Trump administration and related entities

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

iencies

-0.68

NAD

-0.68

SAM

-0.67

çĶ°

-0.63

ASA

-0.62

 Mortal

-0.61

FTP

-0.60

 sqor

-0.60

 mater

-0.60

Laf

-0.59

POSITIVE LOGITS

care

1.01

TRUMP

0.96

Trump

0.95

 Trump

0.92

Donald

0.88

realDonaldTrump

0.88

 Donald

0.87

eting

0.86

Care

0.81

donald

0.79

Activations Density 3.266%

mentions of the Trump administration and related entities

No Comments

No Known Activations