INDEX

Explanations

instances related to traffic violations and law enforcement interactions

oai_token-act-pair · gpt-3.5-turbo

references to speeding and traffic violations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

sonian

-0.82

 Lazarus

-0.75

hof

-0.73

theless

-0.69

romeda

-0.68

fact

-0.65

anyahu

-0.65

KEY

-0.65

sett

-0.64

utical

-0.64

POSITIVE LOGITS

 motorists

1.22

 speeding

1.13

 parked

1.10

 headlights

1.09

 traffic

1.08

 drivers

1.07

 windshield

1.04

 lanes

1.04

 accidents

1.02

 Drivers

1.02

Activations Density 0.200%

instances related to traffic violations and law enforcement interactions

references to speeding and traffic violations

No Comments

No Known Activations