INDEX

Explanations

instances of criticism or negative assessment towards conventional practices or models

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

ValueStyle

-0.42

InitStruct

-0.31

 אד

-0.30

astify

-0.28

 inför

-0.28

karta

-0.28

牌

-0.28

XmlAccessType

-0.28

 barras

-0.27

ติด

-0.27

POSITIVE LOGITS

SharedDtor

0.72

verifyException

0.61

SharedCtor

0.56

httphttps

0.55

enderror

0.54

Completely

0.50

 initComponents

0.50

 Totally

0.49

oa̍t

0.49

 Completely

0.49

Activations Density 0.610%

instances of criticism or negative assessment towards conventional practices or models

No Comments

No Known Activations