INDEX

Explanations

references to academic or technical papers

oai_token-act-pair · gpt-3.5-turbo

various forms of the word "paper" and other related terms

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

STD

-0.75

Qual

-0.72

 exha

-0.71

Brend

-0.71

 referen

-0.67

shall

-0.66

 unden

-0.64

¿½

-0.63

LESS

-0.63

VID

-0.62

POSITIVE LOGITS

Â®

0.71

('

0.70

ives

0.70

utics

0.68

 extraord

0.68

ilon

0.64

igans

0.64

dq

0.63

ffiti

0.63

("

0.63

Activations Density 0.338%

references to academic or technical papers

various forms of the word "paper" and other related terms

No Comments

No Known Activations