INDEX

Explanations

mentions of adjustments or alignments

oai_token-act-pair · gpt-3.5-turbo

phrases related to mechanical adjustments or modifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Rum

-0.72

Online

-0.70

 Markets

-0.70

Synopsis

-0.70

Money

-0.68

 Movies

-0.68

DEM

-0.67

 Towns

-0.67

 Beyon

-0.66

 festivals

-0.65

POSITIVE LOGITS

 protr

1.46

 perpendicular

1.29

 snug

1.25

 indent

1.23

 thickness

1.23

 diameter

1.21

 width

1.20

 angled

1.19

 hinge

1.15

 seam

1.13

Activations Density 0.533%

mentions of adjustments or alignments

phrases related to mechanical adjustments or modifications

No Comments

No Known Activations