INDEX

Explanations

occurrences of Java annotations related to configuration and bean management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.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

 local

-0.47

 trebui

-0.46

domada

-0.46

setUser

-0.45

 yakin

-0.45

run

-0.44

iyle

-0.43

 полу

-0.42

gyar

-0.42

vula

-0.42

POSITIVE LOGITS

 kasarigan

0.97

脚注の使い方

0.84

 GenerationType

0.77

UnusedPrivate

0.75

oa̍t

0.73

GraphicsUnit

0.71

XmlAccessType

0.71

ViewImports

0.71

 ModelExpression

0.69

 <=",

0.66

Activations Density 0.001%

occurrences of Java annotations related to configuration and bean management

No Comments

No Known Activations