INDEX

Explanations

terms indicating precision or exactness in criteria or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_4/width_65k/average_l0_66

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.4.hook_mlp_out

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

 doInBackground

-0.68

Portale

-0.65

 kasarigan

-0.60

endpush

-0.57

BagLayout

-0.56

saraba

-0.56

principalTable

-0.55

MLLoader

-0.54

DockStyle

-0.53

LayoutStyle

-0.52

POSITIVE LOGITS

 للاسماء

0.61

 Riproduzione

0.59

 ciebie

0.56

Eſ

0.55

 anſ

0.54

 fubject

0.54

FieldNumber

0.53

 deſt

0.51

 Wayback

0.51

 urodz

0.51

Activations Density 0.000%

terms indicating precision or exactness in criteria or conditions

No Comments

No Known Activations