INDEX

Explanations

terms related to computer programming and technical setups

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Palest

-0.67

Peb

-0.66

etsy

-0.56

wow

-0.55

 bandwagon

-0.54

entin

-0.54

 Adin

-0.54

 Celt

-0.54

 fortun

-0.53

ãģ£

-0.53

POSITIVE LOGITS

antam

0.67

vation

0.67

inctions

0.66

isine

0.65

yrics

0.65

Entry

0.64

ensable

0.64

oided

0.63

ertodd

0.63

Override

0.62

Activations Density 0.191%

terms related to computer programming and technical setups

No Comments

No Known Activations