INDEX

Explanations

programming-related keywords and elements, particularly in the context of function definitions and documentation

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

 يتيمه

-0.81



-0.75

########.

-0.72

afficheront

-0.64

TypedDataSet

-0.61

 MainAxisSize

-0.61

 ligiloj

-0.61

 चीज़ों

-0.58

出版年

-0.57

 intptr

-0.56

POSITIVE LOGITS

0.61

the

0.58

complexType

0.53

0.46

hline

0.46

memberof

0.46

__(/*!

0.44

0.43

The

0.42

SuppressLint

0.42

Activations Density 0.008%

programming-related keywords and elements, particularly in the context of function definitions and documentation

No Comments

No Known Activations