INDEX

Explanations

technical details regarding measurement parameters and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

Архівовано

-0.32

 comigo

-0.31

 samh

-0.30

 innehå

-0.28

 gustado

-0.28

äldrar

-0.27

conven

-0.26

 verksamhet

-0.26

必

-0.26

 conven

-0.26

POSITIVE LOGITS

StructEnd

0.61

 DBNull

0.60

IsMutable

0.58

tagHelperRunner

0.57

 JpaRepository

0.57

PCell

0.57

ftagPool

0.54

abstractmethod

0.54

ApiModelProperty

0.53

inSlope

0.53

Activations Density 0.488%

technical details regarding measurement parameters and conditions

No Comments

No Known Activations