INDEX

Explanations

words related to processed data or information

oai_token-act-pair · gpt-3.5-turbo

terms related to food preparation and packaging

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

-0.68

 âĢİ

-0.65

iP

-0.56

-0.55

Rap

-0.52

abs

-0.51

Roc

-0.51

Chicago

-0.48

audi

-0.48

 Scotch

-0.48

POSITIVE LOGITS

ãĤ´ãĥ³

0.70

KE

0.62

 guiActiveUn

0.60

ghazi

0.60

pload

0.59

pedia

0.58

 Annotations

0.58

ultimate

0.52

é¾įåĸļå£«

0.52

erker

0.52

Activations Density 2.181%

words related to processed data or information

terms related to food preparation and packaging

No Comments

No Known Activations