INDEX

Explanations

words related to mindsets and attitudes

oai_token-act-pair · gpt-3.5-turbo

references to different types of mentalities or mindsets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

tein

-0.76

cutting

-0.75

icles

-0.75

enegger

-0.73

weather

-0.70

lev

-0.70

gotten

-0.69

arers

-0.68

cuts

-0.68

selling

-0.67

POSITIVE LOGITS

 mentality

1.17

 mindset

1.02

yip

0.85

 attitude

0.81

ãħĭãħĭ

0.69

 attRot

0.68

 arrogance

0.67

istical

0.65

 inconsistency

0.65

achine

0.65

Activations Density 0.014%

words related to mindsets and attitudes

references to different types of mentalities or mindsets

No Comments

No Known Activations