INDEX

Explanations

references to or mentions of the United States

oai_token-act-pair · gpt-3.5-turbo

references to the United States in the context of military actions or policies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

venue

-0.74

bred

-0.72

smanship

-0.69

emetery

-0.69

idated

-0.67

 Reincarn

-0.64

equality

-0.62

 Chatt

-0.62

 Sabbath

-0.61

 srfAttach

-0.60

POSITIVE LOGITS

 embassy

1.10

 Embassy

1.08

 airstrikes

1.05

 counterterrorism

0.93

 drone

0.93

 ally

0.93

AAF

0.93

 diplomats

0.91

 envoy

0.91

 troop

0.89

Activations Density 0.210%

references to or mentions of the United States

references to the United States in the context of military actions or policies

No Comments

No Known Activations