INDEX

Explanations

scientific terminology and statistical concepts

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

showing results or facts

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

 saites

-0.63

AndEndTag

-0.58

 فريبيس

-0.55

ViewFeatures

-0.52

 AssemblyVersion

-0.52

thâu

-0.50

 estekak

-0.49

inaison

-0.49

板

-0.47

assuming

-0.47

POSITIVE LOGITS

 CreateTagHelper

0.45

 sorpresas

0.36

実は

0.36

 overras

0.34

 surprising

0.33

事實

0.33

interopRequire

0.32

 verborgen

0.32

 hidden

0.31

 rzeczyw

0.31

Activations Density 0.165%

scientific terminology and statistical concepts

showing results or facts

No Comments

No Known Activations