INDEX

Explanations

mentions related to military and conflicts

oai_token-act-pair · gpt-3.5-turbo

references to military and defense-related topics

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Cary

-0.81

FTC

-0.72

Champ

-0.71

 Briggs

-0.70

 Madison

-0.69

 Milwaukee

-0.69

 Byrd

-0.69

 Consumer

-0.67

 femin

-0.67

 Fram

-0.66

POSITIVE LOGITS

zbollah

0.91

 Lavrov

0.85

 Hezbollah

0.84

 Jinping

0.81

 Uzbek

0.76

 Kazakh

0.76

Ukraine

0.76

Ä

0.72

ÅĤ

0.71

 Iranians

0.70

Activations Density 1.105%

mentions related to military and conflicts

references to military and defense-related topics

No Comments

No Known Activations