INDEX

Explanations

measurements and units related to various quantities

oai_token-act-pair · gpt-3.5-turbo

measurements and units of quantification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

itialized

-0.82

Ü

-0.71

 handwriting

-0.68

 acron

-0.67

 grips

-0.65

 regrets

-0.63

Joy

-0.62

 logos

-0.62

æĦ

-0.60

 Journals

-0.60

POSITIVE LOGITS

 squared

1.05

imeter

0.95

 basis

0.94

Â²

0.89

 average

0.85

abyte

0.83

 multiplied

0.80

 averaged

0.79

per

0.78

 capita

0.77

Activations Density 0.091%

measurements and units related to various quantities

measurements and units of quantification

No Comments

No Known Activations