INDEX

Explanations

programming-related terminology and references to software functions or classes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Hozzáférés

-0.88

__":

-0.88

 pleaſure

-0.88

 Monfieur

-0.86

 ſever

-0.85

 houſe

-0.83

 purpoſe

-0.82

__':

-0.80

 ſtate

-0.80

 Houſe

-0.80

POSITIVE LOGITS

 BoxDecoration

0.45

 externi

0.39

prepareStatement

0.39

 acquistare

0.39

pak

0.38

 invokingState

0.37

 podrán

0.37

 XMLHttpRequest

0.37

nash

0.35

alt

0.35

Activations Density 0.595%

programming-related terminology and references to software functions or classes

No Comments

No Known Activations