INDEX

Explanations

references to online resources, particularly from Stack Overflow and documentation sites

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

technical q&a links

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

 الرياضيه

-0.50

 ProtoMessage

-0.45

 NSCoder

-0.45

 EconPapers

-0.43

 mxArray

-0.41

PageRoute

-0.41

 referrerpolicy

-0.40

NUMX

-0.39

 تضيفلها

-0.39

يكب

-0.39

POSITIVE LOGITS

stackoverflow

0.60

PYX

0.53

sproz

0.46

Pyx

0.46

addPreferredGap

0.45

 Lesson

0.43

 lesson

0.43

TRAILING

0.43

IsContent

0.42

CppMethod

0.41

Activations Density 0.030%

references to online resources, particularly from Stack Overflow and documentation sites

technical q&a links

No Comments

No Known Activations