INDEX

Explanations

words related to the cooking process

oai_token-act-pair · gpt-3.5-turbo

references to cooking and food preparation

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

DAQ

-0.68

verse

-0.68

eph

-0.67

 Stra

-0.66

lex

-0.64

meter

-0.63

rius

-0.62

 Decl

-0.62

 Sentinel

-0.60

 Nass

-0.60

POSITIVE LOGITS

 cooked

3.44

cooked

2.52

 cooks

1.86

 hatched

1.79

 cooking

1.75

 roasted

1.73

 baked

1.73

 cook

1.72

 brewed

1.67

 boiled

1.59

Activations Density 0.017%

words related to the cooking process

references to cooking and food preparation

No Comments

No Known Activations