INDEX

Explanations

specific formatting elements or annotations within text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ValueStyle

-0.75

 للاسماء

-0.75

 nakalista

-0.60

MessageOf

-0.59

 ModelExpression

-0.57

 يتيمه

-0.56

 kasarigan

-0.55

HideFlags

-0.55

ConstraintMaker

-0.54

 հղումներ

-0.53

POSITIVE LOGITS

<bos>

0.45

 letzt

0.42

ło

0.38

tearDown

0.37

}]

0.37

yyl

0.35

setScene

0.34

❋

0.34

 ones

0.34

[]

0.33

Activations Density 0.569%

specific formatting elements or annotations within text

No Comments

No Known Activations