INDEX

Explanations

references to HTML and related coding terminology

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

 jsPsych

-0.51

 AppComponent

-0.50

NameInMap

-0.48

Erl

-0.47

routeProvider

-0.46

lau

-0.45

yta

-0.45

NewUrlParser

-0.44

 blik

-0.44

 Hinton

-0.44

POSITIVE LOGITS

ⓧ

0.69

IContainer

0.65

DoubleQuotes

0.59

 Monks

0.59

webElement

0.58

AddTagHelper

0.56

SharedCtor

0.55

ules

0.54

]--;

0.54

 BorderRadius

0.52

Activations Density 0.014%

references to HTML and related coding terminology

No Comments

No Known Activations