INDEX

Explanations

specific terms related to written prescriptions and verbal transcriptions

oai_token-act-pair · gpt-3.5-turbo

terminology related to prescriptions or official documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Vand

-0.66

Ib

-0.63

 beds

-0.62

 Love

-0.62

 Messi

-0.61

 Chancellor

-0.60

 plates

-0.60

Hor

-0.59

mer

-0.58

 roller

-0.58

POSITIVE LOGITS

cript

4.72

cription

2.62

cribed

2.34

criptions

2.11

CRIP

1.58

Script

1.38

ascript

1.36

chool

1.16

cend

1.15

 Script

1.14

Activations Density 0.020%

specific terms related to written prescriptions and verbal transcriptions

terminology related to prescriptions or official documents

No Comments

No Known Activations