INDEX

Explanations

phrases related to conflicts, particularly in Eastern Europe

oai_token-act-pair · gpt-3.5-turbo

words related to pro- and anti- affiliations in a political context

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

 Byrd

-0.68

 fallacy

-0.65

 Handler

-0.64

 Hatch

-0.64

 Huck

-0.64

 toile

-0.63

 Lifetime

-0.62

 Wilde

-0.62

 Zucker

-0.62

 Lois

-0.62

POSITIVE LOGITS

government

1.08

democracy

1.04

Islam

1.01

 separatist

1.00

 Kurdish

0.99

Soviet

0.97

Moscow

0.96

Syrian

0.96

military

0.96

Assad

0.96

Activations Density 0.196%

phrases related to conflicts, particularly in Eastern Europe

words related to pro- and anti- affiliations in a political context

No Comments

No Known Activations