INDEX

Explanations

mentions of rewards or incentives

oai_token-act-pair · gpt-3.5-turbo

references to various rewards and reward systems

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

abad

-0.73

lander

-0.70

ovie

-0.69

idents

-0.68

STD

-0.67

abases

-0.67

 Ange

-0.65

enium

-0.65

Osw

-0.64

head

-0.64

POSITIVE LOGITS

 reward

1.26

 rewards

1.17

 rewarded

1.03

Reward

0.95

 Reward

0.87

 rewarding

0.87

 payoff

0.83

 fulfillment

0.80

 recipients

0.80

 tiers

0.79

Activations Density 0.016%

mentions of rewards or incentives

references to various rewards and reward systems

No Comments

No Known Activations