INDEX

Explanations

phrases related to theoretical research and investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

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

AddTagHelper

-1.31

ValueStyle

-0.97

MemoryWarning

-0.92

 ostavi

-0.92

parsedMessage

-0.89

 cherchés

-0.89

abestanden

-0.89

enumii

-0.89

 NgModule

-0.85

InputTagHelper

-0.82

POSITIVE LOGITS

 until

0.50

というと

0.42

di

0.42

 either

0.41

 traditional

0.41

 traditionally

0.40

 prior

0.39

die

0.38

Until

0.38

«

0.38

Activations Density 0.558%

phrases related to theoretical research and investigations

No Comments

No Known Activations