INDEX

Explanations

references to military commanders

oai_token-act-pair · gpt-3.5-turbo

references to military ranks or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

eat

-0.76

employment

-0.67

icles

-0.66

rin

-0.65

-0.64

dust

-0.63

Democrats

-0.62

osponsors

-0.62

esville

-0.61

encers

-0.61

POSITIVE LOGITS

chief

0.86

 commander

0.78

utenant

0.75

 Kham

0.74

 commanding

0.73

 commands

0.73

 commanders

0.73

stration

0.73

selage

0.72

 stationed

0.71

Activations Density 0.017%

references to military commanders

references to military ranks or titles

No Comments

No Known Activations