INDEX

Explanations

mentions of a specific group of people

oai_token-act-pair · gpt-3.5-turbo

references to groups of individuals or communities

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

catentry

-0.87

inventoryQuantity

-0.87

ļéĨĴ

-0.78

Va

-0.75

 srfAttach

-0.75

Accessory

-0.68

Availability

-0.67

Customer

-0.66

 Lank

-0.66

etric

-0.65

POSITIVE LOGITS

folk

1.07

hood

1.03

 smugglers

0.97

 inhab

0.88

gnu

0.88

atics

0.79

ulkan

0.79

 born

0.76

 enslaved

0.76

 massac

0.76

Activations Density 0.094%

mentions of a specific group of people

references to groups of individuals or communities

No Comments

No Known Activations