INDEX

Explanations

phrases related to recipes, ingredients, and cooking instructions

oai_token-act-pair · gpt-3.5-turbo

recipes and food-related terms

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bably

-0.91

 parity

-0.86

 regress

-0.84

 marginalized

-0.84

 tremend

-0.79

yond

-0.78

 tumult

-0.78

 broadcasters

-0.77

 weary

-0.76

 demoral

-0.76

POSITIVE LOGITS

 Recipe

1.32

 Combo

1.32

 Collection

1.30

 Recipes

1.28

 Adapter

1.22

 Solution

1.22

 Starter

1.22

 Jacket

1.22

 Replacement

1.21

 Coup

1.18

Activations Density 0.344%

phrases related to recipes, ingredients, and cooking instructions

recipes and food-related terms

No Comments

No Known Activations