INDEX

Explanations

descriptions of ease and simplicity in processes or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

 kasarigan

-0.83

-0.74

SharedDtor

-0.70

Hochspringen

-0.68

 vera

-0.66

 protoimpl

-0.65

CreateIndex

-0.64

SequentialGroup

-0.64

sidemargin

-0.63

ValueGenerated

-0.62

POSITIVE LOGITS

 Majefty

0.93

 Houſe

0.87

 itſelf

0.84

 pleaſure

0.81

 houſe

0.80

 greateſt

0.77

nexpected

0.75

 myſelf

0.75

ſelf

0.74

 ſtate

0.73

Activations Density 0.015%

descriptions of ease and simplicity in processes or experiences

No Comments

No Known Activations