INDEX

Explanations

ingredients or cooking instructions in a recipe

oai_token-act-pair · gpt-3.5-turbo

references to various types of cooking oils

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

RON

-0.78

EMBER

-0.77

BY

-0.74

 Eucl

-0.73

 Visitors

-0.73

ARGET

-0.71

rons

-0.71

Earth

-0.71

 Prob

-0.71

INAL

-0.69

POSITIVE LOGITS

oil

1.15

Oil

1.08

oil

1.07

Oil

1.04

seed

1.03

 slic

0.99

zyme

0.94

 glands

0.91

anol

0.90

 oils

0.90

Activations Density 0.014%

ingredients or cooking instructions in a recipe

references to various types of cooking oils

No Comments

No Known Activations