INDEX

Explanations

nutritional information in the format of milligrams (mg) or grams (g)

oai_token-act-pair · gpt-3.5-turbo

references to nutritional information 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Crus

-0.75

 revolution

-0.70

 uncovered

-0.67

 Catalan

-0.66

 Catal

-0.66

 basket

-0.66

 astronomy

-0.62

 carriage

-0.61

 Colonial

-0.61

 uncover

-0.60

POSITIVE LOGITS

mg

3.99

mg

2.67

MG

1.72

 Î¼g

1.41

ml

1.38

mc

1.34

mL

1.28

mL

1.19

mc

1.16

Î¼

1.15

Activations Density 0.013%

nutritional information in the format of milligrams (mg) or grams (g)

references to nutritional information and measurements

No Comments

No Known Activations