INDEX

Explanations

phrases related to military actions and international conflict

oai_token-act-pair · gpt-3.5-turbo

references to military operations and geopolitical entities

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

agonist

-0.85

agame

-0.74

olkien

-0.73

 Canaver

-0.72

 Unsure

-0.72

thora

-0.72

 Companion

-0.66

escription

-0.65

ologue

-0.64

ptroller

-0.62

POSITIVE LOGITS

 facilities

1.34

 targets

1.29

 buildings

1.24

 properties

1.20

 installations

1.18

 assets

1.17

 sites

1.14

 residences

1.13

 property

1.12

 landmarks

1.12

Activations Density 0.327%

phrases related to military actions and international conflict

references to military operations and geopolitical entities

No Comments

No Known Activations