INDEX

Explanations

technical terminology related to feature selection in computational science

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 myſelf

-1.00

 pleaſure

-0.97

 itſelf

-0.95

 poffible

-0.94

 NSCoder

-0.93

 reaſon

-0.88

 leaſt

-0.86

CodedInputStream

-0.86

ſelf

-0.84

 themſelves

-0.84

POSITIVE LOGITS

0.62

me

0.41

RadioGroup

0.40

0.39

0.38

叫

0.38

sal

0.38

me

0.38

Sal

0.37

so

0.37

Activations Density 1.184%

technical terminology related to feature selection in computational science

No Comments

No Known Activations