INDEX

Explanations

mentions of military forces, their operations, and deployments

oai_token-act-pair · gpt-3.5-turbo

references to military or armed groups

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

Negative Logits

 Gleaming

-0.69

gger

-0.64

dress

-0.64

ble

-0.63

holiday

-0.63

eson

-0.62

berry

-0.62

yer

-0.61

 Attend

-0.61

hn

-0.60

POSITIVE LOGITS

 forces

3.77

 Forces

2.67

forces

2.33

 force

2.03

 troops

1.67

 armies

1.55

force

1.50

 Force

1.42

 pressures

1.36

Force

1.33

Activations Density 0.020%

mentions of military forces, their operations, and deployments

references to military or armed groups

No Known Activations