INDEX

Explanations

mentions of bounties or rewards offered for various purposes

oai_token-act-pair · gpt-3.5-turbo

references to bounties and bounty hunting

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

case

-0.76

ppard

-0.72

VW

-0.71

icken

-0.70

INK

-0.70

slow

-0.69

BP

-0.69

CON

-0.67

mus

-0.67

orsche

-0.67

POSITIVE LOGITS

 bounty

1.68

 bount

1.44

 Bounty

1.16

 hunter

1.13

 hunters

1.02

iless

0.96

vertisement

0.93

 challeng

0.79

 reward

0.79

ounty

0.78

Activations Density 0.004%

mentions of bounties or rewards offered for various purposes

references to bounties and bounty hunting

No Comments

No Known Activations