INDEX

Explanations

verbs indicating states of being and actions related to existence or occurrence

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

RegressionTest

-0.91

__':

-0.81

 kaarangay

-0.78

 jsPsych

-0.77

={`/

-0.75

 Paglinawan

-0.75

 mourut

-0.74

 purpoſe

-0.72

__":

-0.72

CWE

-0.72

POSITIVE LOGITS

 also

0.76

the

0.52

0.50

 finally

0.49

also

0.48

 pula

0.48

 też

0.48

něž

0.46

 other

0.46

 ALSO

0.46

Activations Density 0.369%

verbs indicating states of being and actions related to existence or occurrence

No Comments

No Known Activations