INDEX

Explanations

car-related terms and activities

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to performance evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

soDeliveryDate

-0.74

NetMessage

-0.70

onym

-0.69

 Authorization

-0.69

channelAvailability

-0.68

soType

-0.66

 Lent

-0.66

optional

-0.65

 Prohibition

-0.64

OPLE

-0.64

POSITIVE LOGITS

 outper

1.51

 excel

1.27

 perform

1.25

 performs

1.18

 fared

1.17

 exce

1.17

 fare

1.09

 performed

1.07

 performance

1.05

 deliver

1.04

Activations Density 0.477%

car-related terms and activities

words and phrases related to performance evaluation

No Comments

No Known Activations