INDEX

Explanations

references to the concept of ego

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "ego."

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

Recomm

-0.78

-0.72

EVA

-0.71

Effective

-0.67

Ô

-0.65

NING

-0.65

Interstitial

-0.65

 Flavoring

-0.64

urgy

-0.64

scenes

-0.63

POSITIVE LOGITS

istical

1.18

ego

1.11

 gratification

0.85

entric

0.83

istically

0.81

istic

0.81

nect

0.81

rolet

0.77

regate

0.76

 infl

0.75

Activations Density 0.005%

references to the concept of ego

references to the concept of "ego."

No Comments

No Known Activations