INDEX

Explanations

terms that indicate comparisons or distinctions between multiple entities or categories

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ment

-1.61

pent

-1.51

ments

-1.40

rives

-1.39

 finale

-1.39

 completes

-1.37

 organizer

-1.37

 diversion

-1.36

 sucks

-1.36

 access

-1.35

POSITIVE LOGITS

Ļª

4.26

Īĺ

4.23

¸

4.23

³

4.17

¼

4.16

¿½

4.15

ľ

4.14

¶

4.09

į

4.09

¦

4.08

Activations Density 0.418%

terms that indicate comparisons or distinctions between multiple entities or categories

No Comments

No Known Activations