INDEX

Explanations

phrases related to political figures or actions

oai_token-act-pair · gpt-3.5-turbo

terms associated with political or financial ideologies, specifically those related to hawkish stances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Incarnation

-0.77

 Investigator

-0.74

 Mysteries

-0.72

 Sacrament

-0.70

URE

-0.70

 Piercing

-0.68

 Framework

-0.68

 Result

-0.68

 Scotia

-0.67

 Mankind

-0.67

POSITIVE LOGITS

haw

1.41

keye

1.32

kish

1.07

 hawk

1.05

keyes

1.01

aii

0.94

indal

0.88

alus

0.81

pers

0.80

wat

0.80

Activations Density 0.008%

phrases related to political figures or actions

terms associated with political or financial ideologies, specifically those related to hawkish stances

No Comments

No Known Activations