INDEX

Explanations

terms related to nutritional information, specifically focusing on carbohydrate and protein content

oai_token-act-pair · gpt-3.5-turbo

terms related to rates and measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Wing

-0.73

 face

-0.70

 guard

-0.70

 Guard

-0.68

 miss

-0.65

 letter

-0.64

 wedding

-0.64

 locker

-0.63

JC

-0.63

 wearing

-0.63

POSITIVE LOGITS

rates

4.80

rate

3.29

rating

2.21

rations

2.08

rated

2.04

Rate

1.77

 Rates

1.71

ration

1.70

rat

1.68

rator

1.63

Activations Density 0.015%

terms related to nutritional information, specifically focusing on carbohydrate and protein content

terms related to rates and measurements

No Comments

No Known Activations