INDEX

Explanations

"pivot" or related words denoting a change in direction or strategy

oai_token-act-pair · gpt-3.5-turbo

terms related to pivots and pivoting in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

enic

-0.72

WAY

-0.71

aver

-0.70

Interstitial

-0.70

à¦

-0.70

Copyright

-0.67

ENA

-0.65

avering

-0.65

chn

-0.64

 Birthday

-0.63

POSITIVE LOGITS

 pivot

1.50

piv

1.20

ivot

1.13

ilus

0.91

olicy

0.84

inson

0.74

ipolar

0.74

 hinges

0.74

olate

0.72

 hinge

0.72

Activations Density 0.006%

"pivot" or related words denoting a change in direction or strategy

terms related to pivots and pivoting in various contexts

No Comments

No Known Activations