INDEX

Explanations

instances of the word "granted."

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

isman

-1.78

uality

-1.63

ck

-1.60

 characterized

-1.54

ambda

-1.50

uria

-1.50

 reflected

-1.47

HECK

-1.45

Published

-1.40

 interrupted

-1.37

POSITIVE LOGITS

ĻĤ

2.36

¤

2.06

ļ

2.02

°

1.95

ģ

1.91

Ł

1.90

¦

1.88

Į

1.85

¶

1.82

ľ

1.81

Activations Density 1.587%

instances of the word "granted."

No Known Activations