INDEX

Explanations

elements related to structured data and functions in programming, particularly in the context of arrays and their manipulation

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

list, followed by, calculated

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-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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sloppy

-0.45

 leaky

-0.44

JTable

-0.41

pushFollow

-0.40

 cAMP

-0.40

ConfigureAwait

-0.39

addView

-0.38

 suicide

-0.38

AMC

-0.38

 skate

-0.37

POSITIVE LOGITS

 defaultstate

0.61

 HttpNotFound

0.52

 بيها

0.52

domada

0.50

RenderAtEndOf

0.49

сылкі

0.47

 autorytatywna

0.46

enschappelijke

0.46

hyrchwyd

0.45

encodeWith

0.45

Activations Density 0.007%

elements related to structured data and functions in programming, particularly in the context of arrays and their manipulation

list, followed by, calculated

No Comments

No Known Activations