INDEX

Explanations

references to the formatting or structuring functions in code

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

Īĺ

-2.26

·¸

-2.21

Ĥ

-2.18

Ĥ¬

-2.10

ı

-2.09

ĭ

-2.00

¦

-1.92

±

-1.85

µ

-1.84

ĥ

-1.83

POSITIVE LOGITS

 festivals

1.57

 profession

1.57

ios

1.52

erals

1.48

 lantern

1.43

nell

1.36

 festival

1.36

nad

1.35

ings

1.35

chat

1.35

Activations Density 0.021%

references to the formatting or structuring functions in code

No Comments

No Known Activations