INDEX

Explanations

references to violent incidents and military activity

oai_token-act-pair · gpt-3.5-turbo

content related to military operations and conflict

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ãĤ´ãĥ³

-0.87

 Oprah

-0.81

phia

-0.76

 immortality

-0.76

 envy

-0.74

 Proposition

-0.74

 Vernon

-0.71

livion

-0.71

 renegoti

-0.71

 Depression

-0.71

POSITIVE LOGITS

 Daesh

1.25

 gunmen

1.22

 attackers

1.20

IDF

1.18

 militants

1.16

 armoured

1.15

 ISIL

1.13

 gunfire

1.13

 grenade

1.13

 mortar

1.11

Activations Density 0.630%

references to violent incidents and military activity

content related to military operations and conflict

No Comments

No Known Activations