INDEX

Explanations

phrases related to casualties or injuries in various incidents or conflicts

oai_token-act-pair · gpt-3.5-turbo

phrases related to injuries and casualties in conflicts or incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Programs

-0.76

sonian

-0.72

inventoryQuantity

-0.68

Nothing

-0.67

 Instruction

-0.67

 Evolution

-0.67

cknow

-0.67

 Gleaming

-0.65

 srfAttach

-0.65

 Gutenberg

-0.65

POSITIVE LOGITS

 injured

1.10

 hospitalized

1.07

athi

1.04

 killed

1.02

 detained

1.01

involved

0.99

 abducted

0.95

 murdered

0.94

 arrested

0.91

 evacuated

0.90

Activations Density 0.190%

phrases related to casualties or injuries in various incidents or conflicts

phrases related to injuries and casualties in conflicts or incidents

No Comments

No Known Activations