INDEX

Explanations

phrases related to breaking something into parts or pieces

oai_token-act-pair · gpt-3.5-turbo

phrases indicating elevation or an increase

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

 Clicker

-0.69

 commencement

-0.65

FK

-0.63

redo

-0.61

vice

-0.60

riz

-0.60

natureconservancy

-0.59

 availability

-0.58

Trade

-0.57

bt

-0.57

POSITIVE LOGITS

by

1.02

BY

0.83

sole

0.81

estone

0.74

by

0.73

 aback

0.73

monton

0.69

 territ

0.68

 onto

0.67

upon

0.67

Activations Density 0.146%

phrases related to breaking something into parts or pieces

phrases indicating elevation or an increase

No Comments

No Known Activations