INDEX

Explanations

locations or places where violent incidents occurred

oai_token-act-pair · gpt-3.5-turbo

locations and names associated with incidents or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

////////////////////////////////

-0.67

Disclaimer

-0.66

lishes

-0.65

wcsstore

-0.63

DVD

-0.62

================================================================

-0.62

TRUMP

-0.61

VALUE

-0.61

åŃ

-0.61

Management

-0.61

POSITIVE LOGITS

 suburb

0.92

 countryside

0.84

 airport

0.83

 gunfire

0.80

 prisons

0.79

 ambush

0.79

 streets

0.77

 Harbor

0.76

 suburbs

0.76

 jails

0.76

Activations Density 0.342%

locations or places where violent incidents occurred

locations and names associated with incidents or events

No Comments

No Known Activations