INDEX

Explanations

numerical and scientific terms related to measurements and identifiers

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

keywords associated with data management concepts, particularly in a technical context

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

dash-separated abbreviations and names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

".

-0.98

\{\\

-0.94

:");

-0.87

.")

-0.87

!")

-0.86

?";

-0.85

.");

-0.85

?")

-0.84

!");

-0.83

?");

-0.83

POSITIVE LOGITS

2.41

–

2.19

--

1.94

—

1.86

−

1.51

-,

1.40

---

1.38

$-$

1.34

—

1.30

--

1.27

Activations Density 0.121%

numerical and scientific terms related to measurements and identifiers

keywords associated with data management concepts, particularly in a technical context

dash-separated abbreviations and names

No Comments

No Known Activations