INDEX

Explanations

comparative language related to performance metrics or standards

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/4-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.4.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

³

-3.34

ģ

-3.10

Ŀ

-3.03

ĭ

-3.03

º

-3.02

ĵ

-3.00

Ĥ

-2.99

¡

-2.98

ľĵ

-2.95

¥

-2.95

POSITIVE LOGITS

 counterparts

2.80

 elsewhere

2.42

 others

2.19

 other

2.14

 counterpart

2.08

 ones

2.07

 placebo

2.04

other

2.02

 usual

1.96

 earlier

1.89

Activations Density 1.179%

comparative language related to performance metrics or standards

No Comments

No Known Activations