INDEX

Explanations

phrases related to incarceration or confinement

oai_token-act-pair · gpt-3.5-turbo

terms related to prisoners and their conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

drive

-0.78

UTH

-0.65

orp

-0.65

ilde

-0.64

lag

-0.63

dust

-0.63

 Boll

-0.63

 Pillar

-0.63

 Ples

-0.62

orem

-0.62

POSITIVE LOGITS

 incarcerated

1.03

 sentenced

1.01

 freed

0.90

 housed

0.89

 inmates

0.88

 jailed

0.85

 prisoners

0.84

 inmate

0.81

 prisoner

0.80

 confinement

0.79

Activations Density 0.053%

phrases related to incarceration or confinement

terms related to prisoners and their conditions

No Comments

No Known Activations