INDEX

Explanations

concepts related to punishment

oai_token-act-pair · gpt-3.5-turbo

references to various types of punishment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

sonian

-0.90

gow

-0.83

ergy

-0.74

 Mich

-0.71

sie

-0.69

coe

-0.67

opic

-0.65

aug

-0.65

olated

-0.64

itect

-0.64

POSITIVE LOGITS

 punishment

1.04

 punishments

0.99

 inflicted

0.97

 punished

0.92

 sanction

0.85

 regimes

0.80

 punish

0.80

 harshly

0.79

 imposed

0.79

 penalty

0.79

Activations Density 0.022%

concepts related to punishment

references to various types of punishment

No Comments

No Known Activations