INDEX

Explanations

negative health outcomes and related scientific terms

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-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.2.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

sole

-1.54

agment

-1.50

 polls

-1.45

vier

-1.40

ts

-1.38

 offline

-1.37

uses

-1.37

illa

-1.35

itons

-1.34

imet

-1.34

POSITIVE LOGITS

·

1.51

Verlag

1.50

¾

1.50

Ĥ

1.41

---|---|---

1.40

 COURT

1.35

---|---

1.34

]{}\^

1.34

compl

1.33

½

1.33

Activations Density 0.048%

negative health outcomes and related scientific terms

No Comments

No Known Activations