INDEX

Explanations

instances related to the word 'Plat'

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names or brands

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

CVE

-0.70

tainment

-0.66

lde

-0.64

corn

-0.64

 peel

-0.62

 scissors

-0.60

 corrosion

-0.59

 sink

-0.59

————

-0.58

 Scientology

-0.58

POSITIVE LOGITS

onic

1.03

atform

1.03

itudes

1.02

imes

0.93

yp

0.93

astical

0.91

ilateral

0.88

oken

0.87

agonist

0.87

oral

0.87

Activations Density 0.039%

instances related to the word 'Plat'

mentions of specific names or brands

No Comments

No Known Activations