INDEX

Explanations

instances of the word "it."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÃŃs

-2.25

ĵ

-1.90

ķ

-1.81

resents

-1.65

edes

-1.61

ĺ

-1.60

¥

-1.59

Ħ

-1.57

Ģ

-1.57

ĭ

-1.56

POSITIVE LOGITS

place

1.71

ncbi

1.64

imize

1.47

icken

1.34

ubottu

1.31

strand

1.30

amate

1.27

amo

1.27

 lane

1.26

blogger

1.26

Activations Density 0.398%

instances of the word "it."

No Comments

No Known Activations