INDEX

Explanations

numeric values or data points within a scientific context

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

citations ending in ]]]. or ))

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

፩

-0.64

Демографія

-0.61

 Fland

-0.58

IBLIO

-0.54

 queſta

-0.53

 myſelf

-0.53

ðsíða

-0.49

Historio

-0.49

ReusableCell

-0.49

arangay

-0.49

POSITIVE LOGITS

 كومونز

0.52

 <<<<<<<<<<<<<<

0.46

kháu

0.42

DockStyle

0.37

 arşivlendi

0.37

gnore

0.35

 einst

0.35

enderror

0.34

erweise

0.34

tapan

0.34

Activations Density 0.007%

numeric values or data points within a scientific context

citations ending in ]]]. or ))

No Comments

No Known Activations