INDEX

Explanations

phrases related to control and management

oai_token-act-pair · gpt-3.5-turbo

instances of control or management over various aspects or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

nces

-0.89

lehem

-0.76

urai

-0.76

cellent

-0.73

mire

-0.71

å§

-0.69

hiro

-0.67

lli

-0.67

Champ

-0.66

olor

-0.66

POSITIVE LOGITS

 levers

1.01

 reins

0.92

 finances

0.79

 logistics

0.79

 interstate

0.72

 autonom

0.71

 faculties

0.69

 aspects

0.68

 environment

0.68

 airflow

0.67

Activations Density 0.272%

phrases related to control and management

instances of control or management over various aspects or entities

No Comments

No Known Activations