INDEX

Explanations

phrases indicating complexity and contradiction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 6-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_6/width_16k/average_l0_133

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.6.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

far

-0.54

</blockquote>

-0.47

ConstraintMaker

-0.46

 experiment

-0.46

 active

-0.45

 study

-0.45

 raise

-0.45

 build

-0.43

 plus

-0.43

POSITIVE LOGITS

PhysRev

0.82

 poveznice

0.81

دانشنامهٔ

0.81

ollectionView

0.70

 الحره

0.70

imanapun

0.69

 saites

0.69

awtextra

0.66

SequentialGroup

0.66

WriteAttribute

0.65

Activations Density 0.301%

phrases indicating complexity and contradiction

No Comments

No Known Activations