INDEX

Explanations

references to files, devices, or physical entities in a digital or technical context

oai_token-act-pair · gpt-3.5-turbo

key elements or components related to structures and categories

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

 Canaver

-0.71

 Byrne

-0.68

Sol

-0.62

gypt

-0.62

 poles

-0.61

lys

-0.60

 EDITION

-0.60

utsche

-0.59

uly

-0.58

 Dion

-0.57

POSITIVE LOGITS

ä»£

0.76

 Secondary

0.67

viz

0.67

ceptor

0.65

 tert

0.65

oyer

0.64

 diversion

0.63

chest

0.62

 Reasons

0.61

lift

0.60

Activations Density 0.131%

references to files, devices, or physical entities in a digital or technical context

key elements or components related to structures and categories

No Comments

No Known Activations