INDEX

Explanations

words related to geopolitical events and government actions

oai_token-act-pair · gpt-3.5-turbo

references to violence and military actions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Bigfoot

-0.85

 Asheville

-0.71

 Guys

-0.71

 Finder

-0.71

 Fresno

-0.70

BYU

-0.70

 Glow

-0.70

 Milky

-0.69

 BART

-0.68

 PLEASE

-0.68

POSITIVE LOGITS

 parliamentary

0.98

 privat

0.97

 repud

0.95

 Nato

0.91

 renegoti

0.89

 parliament

0.88

 Parliamentary

0.85

 repatri

0.84

 mobil

0.84

 weakening

0.84

Activations Density 1.516%

words related to geopolitical events and government actions

references to violence and military actions

No Comments

No Known Activations