INDEX

Explanations

phrases indicating structured guidance or recommendations

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

enumerating types or methods

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

 wcześniej

-0.37

 centaines

-0.34

 whereas

-0.33

 <<<<<<<<<<<<<<

-0.33

 מכל

-0.32

 Others

-0.32

 yıll

-0.32

 Materiales

-0.32

 vorher

-0.32

ResumeLayout

-0.32

POSITIVE LOGITS

 Roskov

0.72

XmlAccessType

0.63

 CreateTagHelper

0.56

tagHelperRunner

0.55

########.

0.53

 يتيمه

0.51

 mergeFrom

0.49

 MyBatis

0.49

localctx

0.49

Παραπομπές

0.47

Activations Density 0.021%

phrases indicating structured guidance or recommendations

enumerating types or methods

No Comments

No Known Activations