INDEX

Explanations

terms related to detention centers

oai_token-act-pair · gpt-3.5-turbo

references to detention facilities and associated concepts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Toast

-0.72

ãĥ£

-0.70

 toast

-0.64

chrom

-0.64

 deductible

-0.64

 Decay

-0.62

bold

-0.61

UTH

-0.60

Cum

-0.58

 Russo

-0.57

POSITIVE LOGITS

 centres

1.09

 facilities

1.07

 facility

1.06

 detention

1.05

 centers

1.02

 Facility

1.00

 camps

0.95

 detain

0.94

 confinement

0.93

 detainees

0.90

Activations Density 0.073%

terms related to detention centers

references to detention facilities and associated concepts

No Comments

No Known Activations