INDEX

Explanations

phrases related to time and duration

oai_token-act-pair · gpt-3.5-turbo

references to ongoing discussions or topics of interest

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

 seiz

-0.54

 misunder

-0.50

 boarding

-0.48

 deadly

-0.48

 delicious

-0.47

 harrowing

-0.46

 destructive

-0.46

 thrilling

-0.46

 traged

-0.45

 resc

-0.45

POSITIVE LOGITS

along

0.61

none

0.61

emale

0.60

REDACTED

0.59

DIT

0.59

ITION

0.59

ilib

0.58

èĢħ

0.57

phasis

0.56

rompt

0.56

Activations Density 1.078%

phrases related to time and duration

references to ongoing discussions or topics of interest

No Comments

No Known Activations