INDEX

Explanations

terms related to terrorist organizations, specifically focusing on Al-Qaeda in the Arabian Peninsula

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Seah

-0.85

 Gron

-0.81

 Hubble

-0.78

 Vaugh

-0.75

 Constantin

-0.74

 Goodell

-0.73

ðĿ

-0.73

Wem

-0.72

kj

-0.71

 Kear

-0.71

POSITIVE LOGITS

 affiliates

1.11

 terrorists

1.08

 insurgents

1.04

 fighters

1.01

 operatives

1.01

 affiliate

1.00

 militants

0.99

 affiliated

0.99

 insurgency

0.99

 terrorist

0.96

Activations Density 0.019%

terms related to terrorist organizations, specifically focusing on Al-Qaeda in the Arabian Peninsula

No Comments

No Known Activations