INDEX

Explanations

terms related to legal reasoning and justification

oai_token-act-pair · gpt-3.5-turbo

phrases related to legal standards and justifications for the use of force

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 nerds

-0.67

 Billion

-0.65

luaj

-0.65

 iPads

-0.63

 Andromeda

-0.62

opa

-0.62

 Quartz

-0.62

 Brass

-0.62

 Marse

-0.61

soDeliveryDate

-0.61

POSITIVE LOGITS

reasonable

1.27

reason

1.22

 reasonably

1.21

 reasonable

1.14

 probable

1.14

 practicable

1.14

 impairment

1.12

 materially

1.09

sufficient

1.08

 impair

1.05

Activations Density 0.459%

terms related to legal reasoning and justification

phrases related to legal standards and justifications for the use of force

No Comments

No Known Activations