INDEX

Explanations

instructions or guides on how to do specific tasks or activities

oai_token-act-pair · gpt-3.5-turbo

actions and instructions related to the process of making or doing something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

eatures

-0.77

blogspot

-0.75

vic

-0.73

ipped

-0.65

hement

-0.65

uploads

-0.64

court

-0.64

eks

-0.64

axter

-0.63

allery

-0.62

POSITIVE LOGITS

uate

0.89

 efficiently

0.70

ulate

0.68

 oneself

0.67

 yourself

0.67

 attribution

0.67

 properly

0.65

TEXTURE

0.64

 Harlem

0.61

 mult

0.61

Activations Density 0.229%

instructions or guides on how to do specific tasks or activities

actions and instructions related to the process of making or doing something

No Comments

No Known Activations