INDEX

Explanations

references to being imprisoned or held captive

oai_token-act-pair · gpt-3.5-turbo

references to prisoners and their experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

orp

-0.79

 Boll

-0.75

OPA

-0.70

amera

-0.68

wig

-0.66

orie

-0.66

drive

-0.64

alore

-0.64

ulously

-0.64

ories

-0.63

POSITIVE LOGITS

 prisoners

1.01

 prisoner

0.90

 captives

0.88

 inmates

0.87

 sentenced

0.83

 detainees

0.79

 incarcerated

0.78

 confinement

0.77

icts

0.73

 captive

0.73

Activations Density 0.025%

references to being imprisoned or held captive

references to prisoners and their experiences

No Comments

No Known Activations