INDEX

Explanations

specific locations or addresses containing the word "Drive"

oai_token-act-pair · gpt-3.5-turbo

references to street names or locations that include the word "Drive."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

umbn

-0.76

icipated

-0.74

itious

-0.73

imated

-0.71

uba

-0.71

guyen

-0.69

ANN

-0.68

ark

-0.67

astics

-0.67

phabet

-0.67

POSITIVE LOGITS

 Drive

1.05

 Yards

0.81

Dri

0.80

 Parkway

0.80

 Drift

0.77

away

0.77

Drive

0.74

 Driving

0.73

 Highway

0.72

bys

0.71

Activations Density 0.010%

specific locations or addresses containing the word "Drive"

references to street names or locations that include the word "Drive."

No Comments

No Known Activations