INDEX

Explanations

ingredients and cooking instructions in a recipe

oai_token-act-pair · gpt-3.5-turbo

quantitative measurements and ingredient amounts in recipes

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

 milestones

-0.71

 regress

-0.71

 Intellectual

-0.70

 Visitors

-0.70

ORPG

-0.66

 Events

-0.66

Events

-0.64

 vulnerabilities

-0.63

 paranormal

-0.63

Operation

-0.63

POSITIVE LOGITS

 preferably

1.14

 preferable

0.94

 dried

0.93

 tablespoon

0.92

 softened

0.89

 pinch

0.89

optional

0.88

 powdered

0.88

 diced

0.87

 coconut

0.87

Activations Density 0.247%

ingredients and cooking instructions in a recipe

quantitative measurements and ingredient amounts in recipes

No Comments

No Known Activations