INDEX

Explanations

terms related to conflict, battles, and military engagements

oai_token-act-pair · gpt-3.5-turbo

terms related to combat or conflict situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Remastered

-0.71

Magazine

-0.69

CLA

-0.68

 Swim

-0.67

Morgan

-0.66

âĶĢâĶĢâĶĢâĶĢ

-0.64

 Upton

-0.64

Consumer

-0.64

 Richards

-0.64

Dick

-0.63

POSITIVE LOGITS

fighting

1.13

 Libyan

0.93

 brig

0.93

 insurgents

0.91

 Afghans

0.90

 milit

0.90

 waged

0.88

 ISIS

0.85

 Afghan

0.85

 ISIL

0.85

Activations Density 0.089%

terms related to conflict, battles, and military engagements

terms related to combat or conflict situations

No Comments

No Known Activations