INDEX

Explanations

references to programming concepts related to data structures and algorithms

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

Date or year with week information

np_acts-logits-general · gemini-2.0-flash

unique equivalence and distinctness

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

InputTagHelper

-0.56

Gön

-0.48

MLLoader

-0.47

cuela

-0.45

 FetchType

-0.44

ConverterFactory

-0.42

 betalen

-0.42

 '\\;'

-0.42

 Forecasting

-0.42

jsxFileName

-0.42

POSITIVE LOGITS

 patterns

0.48

 unique

0.47

 pattern

0.44

 Unique

0.42

 fingerprints

0.40

 OkHttpClient

0.40

 distinct

0.39

featureID

0.38

unique

0.38

Unique

0.37

Activations Density 0.477%

references to programming concepts related to data structures and algorithms

Date or year with week information

unique equivalence and distinctness

No Comments

No Known Activations