INDEX

Explanations

numerical sizes and measurements

oai_token-act-pair · gpt-3.5-turbo

size classifications and ratings of quality

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

phies

-0.67

 Statue

-0.65

 Bureau

-0.63

 Spit

-0.62

CLAIM

-0.61

tips

-0.60

 Brus

-0.60

 Donation

-0.59

 sorely

-0.58

 Clintons

-0.58

POSITIVE LOGITS

-,

1.00

 versus

0.92

(>

0.90

-.

0.88

or

0.88

missive

0.87

ecided

0.87

+,

0.84

-)

0.83

/-

0.82

Activations Density 0.455%

numerical sizes and measurements

size classifications and ratings of quality

No Comments

No Known Activations