INDEX

Explanations

words related to files or documents

oai_token-act-pair · gpt-3.5-turbo

references to document files or reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ofi

-0.84

orical

-0.82

angular

-0.78

orically

-0.74

 Alban

-0.72

ophe

-0.71

apo

-0.71

 Palestin

-0.70

assi

-0.68

mal

-0.68

POSITIVE LOGITS

 cabinets

0.83

Upload

0.82

thumbnails

0.76

backs

0.74

 drawer

0.74

chens

0.73

 file

0.73

 FOIA

0.72

 Upload

0.72

 paperwork

0.72

Activations Density 0.019%

words related to files or documents

references to document files or reports

No Comments

No Known Activations