INDEX

Explanations

descriptions of opinions or ratings, including different rating scales and feedback options

oai_token-act-pair · gpt-3.5-turbo

options, responses, and ratings related to decision-making scenarios

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĸļ

-0.83

atonin

-0.82

soDeliveryDate

-0.70

issance

-0.70

iership

-0.68

forth

-0.68

 supposedly

-0.68

conservancy

-0.68

illon

-0.67

 rejuven

-0.67

POSITIVE LOGITS

 Unsure

1.39

 Somew

1.14

Yes

1.07

 None

1.05

ecided

0.96

 Cancel

0.96

 Very

0.94

 Cannot

0.94

 Completed

0.93

(%)

0.92

Activations Density 0.208%

descriptions of opinions or ratings, including different rating scales and feedback options

options, responses, and ratings related to decision-making scenarios

No Comments

No Known Activations