INDEX

Explanations

references to academic or technical content, particularly related to economics and operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 فريبيس

-0.68

Referanser

-0.58

ClientSize

-0.57

badlogic

-0.57

SizeF

-0.56

 وتسجيلات

-0.55

 AssemblyProduct

-0.54

انيف

-0.54

CppMethod

-0.53

 Affected

-0.53

POSITIVE LOGITS

PreferredItem

0.68

 GenerationType

0.67

posób

0.54

MLLoader

0.54

 engraçadas

0.52

 preferencia

0.51

 sherds

0.51

 vermelhas

0.51

QMetaType

0.49

bellum

0.49

Activations Density 0.713%

references to academic or technical content, particularly related to economics and operations

No Comments

No Known Activations