INDEX

Explanations

phrases related to militant groups and organizations

oai_token-act-pair · gpt-3.5-turbo

references to militant groups and their activities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.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

othy

-0.77

ministic

-0.75

 Titanic

-0.69

 Wheat

-0.68

ultz

-0.68

obyl

-0.67

Wax

-0.66

VERTISEMENT

-0.66

 Rockies

-0.66

 Diet

-0.65

POSITIVE LOGITS

 Boko

1.30

 Hezbollah

1.22

 militants

1.21

 jihadists

1.18

 insurgents

1.17

 allied

1.12

 militant

1.11

 Daesh

1.09

 jihadist

1.09

 Islamists

1.08

Activations Density 0.179%

phrases related to militant groups and organizations

references to militant groups and their activities

No Comments

No Known Activations