INDEX

Explanations

the word "aim" as well as instances of the concept of intent or purpose

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.0.hook_mlp_out

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

uminous

-1.57

 Principal

-1.56

midt

-1.54

TRODUCTION

-1.48

say

-1.40

$$

-1.40

}$$

-1.39

separator

-1.38

oyal

-1.35

$$

-1.34

POSITIVE LOGITS

 arsenal

1.70

edo

1.54

iol

1.51

enstein

1.51

olta

1.49

ulator

1.49

iop

1.48

etics

1.45

aje

1.43

ory

1.42

Activations Density 0.010%

the word "aim" as well as instances of the concept of intent or purpose

No Known Activations