INDEX

Explanations

quantities and comparisons related to sizes and classifications

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

NameInMap

-0.46

exitRule

-0.45

ambién

-0.43

antaranya

-0.41

curities

-0.40

 argint

-0.40

 suprême

-0.40

entlig

-0.40

 împre

-0.40

mybatisplus

-0.39

POSITIVE LOGITS

 المعيارى

0.53

 mini

0.48

 ProtoMessage

0.45

 Brief

0.45

 brief

0.45

minecraftforge

0.44

 članak

0.43

 sort

0.42

يكب

0.42

 Vesti

0.41

Activations Density 0.153%

quantities and comparisons related to sizes and classifications

No Comments

No Known Activations