INDEX

Explanations

technical terms associated with products or product descriptions

oai_token-act-pair · gpt-3.5-turbo

references to products and their features

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Coy

-0.75

 labels

-0.69

obo

-0.66

Fat

-0.65

 Mist

-0.62

 Castro

-0.61

orient

-0.61

ie

-0.60

Neg

-0.60

 camps

-0.59

POSITIVE LOGITS

Product

2.02

Premium

2.01

urated

1.55

âĦ¢:

1.52

 patented

1.51

Offline

1.50

Micro

1.22

Legendary

1.20

Ep

1.08

Untitled

1.07

Activations Density 0.062%

technical terms associated with products or product descriptions

references to products and their features

No Comments

No Known Activations