INDEX

Explanations

terms related to events, objects, or actions within a structured system

oai_token-act-pair · gpt-3.5-turbo

terms related to structured organizations, systems, and legal frameworks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤ½

-0.67

Chance

-0.65

cale

-0.64

DragonMagazine

-0.62

ync

-0.62

Redd

-0.62

Ô

-0.62

azo

-0.60

peed

-0.60

lap

-0.59

POSITIVE LOGITS

 itself

1.36

's

0.94

eers

0.84

 herself

0.79

wright

0.78

 homepage

0.76

iverse

0.71

keeper

0.70

 opener

0.70

runners

0.69

Activations Density 0.419%

terms related to events, objects, or actions within a structured system

terms related to structured organizations, systems, and legal frameworks

No Comments

No Known Activations