INDEX

Explanations

sequences related to rankings or order of elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/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.12.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

 cara

-0.39

 late

-0.38

 long

-0.38

 early

-0.36

 kjent

-0.36

premi

-0.36

 carico

-0.35

 future

-0.35

 VIDA

-0.35

 impul

-0.34

POSITIVE LOGITS

posedge

0.86

#+#

0.79

minecraftforge

0.77

URLException

0.77

SharedDtor

0.76

 TestBed

0.75

 fieldNum

0.75

Étymologie

0.74

 jScrollPane

0.73

principalColumn

0.73

Activations Density 0.026%

sequences related to rankings or order of elements

No Comments

No Known Activations