INDEX

Explanations

terms related to database systems, particularly focusing on mentions of "DB" followed by a numerical score

oai_token-act-pair · gpt-3.5-turbo

references to a specific database, particularly "DB" and "db"

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

olulu

-0.85

terday

-0.71

bourg

-0.69

ideshow

-0.68

 Centauri

-0.66

 willful

-0.66

 Yanuk

-0.64

 Hague

-0.63

ments

-0.63

 evangelical

-0.61

POSITIVE LOGITS

DB

0.97

ERG

0.89

pedia

0.87

ulous

0.82

MS

0.76

oard

0.75

dump

0.75

ranch

0.75

JECT

0.74

arser

0.74

Activations Density 0.009%

terms related to database systems, particularly focusing on mentions of "DB" followed by a numerical score

references to a specific database, particularly "DB" and "db"

No Comments

No Known Activations