INDEX

Explanations

information related to explosives and bombs

oai_token-act-pair · gpt-3.5-turbo

terms related to explosives and detonations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Kitt

-0.71

Latin

-0.71

rait

-0.68

itt

-0.68

orest

-0.63

iw

-0.63

anse

-0.63

Â½

-0.62

ivia

-0.62

 View

-0.61

POSITIVE LOGITS

 explosives

3.72

 Explos

2.18

 deton

1.86

 detonated

1.83

 bombs

1.83

 grenades

1.77

 explosive

1.70

 Explosive

1.65

 munitions

1.61

TNT

1.53

Activations Density 0.025%

information related to explosives and bombs

terms related to explosives and detonations

No Comments

No Known Activations