INDEX

Explanations

words related to giving up something important for a greater cause or purpose

oai_token-act-pair · gpt-3.5-turbo

concepts related to self-sacrifice

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Nap

-0.73

 Cheong

-0.68

BuyableInstoreAndOnline

-0.68

ergy

-0.67

realDonaldTrump

-0.66

Correct

-0.66

updated

-0.66

sure

-0.65

col

-0.64

WHO

-0.63

POSITIVE LOGITS

 sacrifice

1.19

rifice

1.17

 sacrificed

1.08

 sacrifices

1.05

 sacrificing

0.98

 sacrific

0.97

ificial

0.97

rament

0.86

 altar

0.84

 Sacrifice

0.80

Activations Density 0.016%

words related to giving up something important for a greater cause or purpose

concepts related to self-sacrifice

No Comments

No Known Activations