INDEX

Explanations

instances of punctuation and format indicators in text

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

 Explorer

-1.61

amethasone

-1.42

________________

-1.42

MSC

-1.42

aceae

-1.37

onial

-1.37

otech

-1.35

dale

-1.33

hores

-1.33

ayer

-1.33

POSITIVE LOGITS

ķ

4.39

ĸ´

4.09

ĸ

4.04

¿

3.92

£

3.89

ĥ½

3.86

Ī

3.82

¦

3.81

§

3.77

©

3.71

Activations Density 3.129%

instances of punctuation and format indicators in text

No Comments

No Known Activations