INDEX

Explanations

references to specific types of vehicles involved in different scenarios

oai_token-act-pair · gpt-3.5-turbo

mentions of vehicles involved in incidents

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

utical

-0.81

 Pinball

-0.72

ãĤ´ãĥ³

-0.71

laus

-0.70

ettings

-0.68

algia

-0.68

èª

-0.68

oku

-0.66

ãĤ±

-0.65

 Flavoring

-0.65

POSITIVE LOGITS

 parked

1.11

 convoy

0.91

 towed

0.90

 dealership

0.87

load

0.86

ousel

0.84

 robbery

0.80

 belonging

0.79

 burglary

0.79

SUV

0.77

Activations Density 0.084%

references to specific types of vehicles involved in different scenarios

mentions of vehicles involved in incidents

No Comments

No Known Activations