INDEX

Explanations

queries related to technical support for software or systems

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

to show how

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

 subsequently

-0.52

 bowiem

-0.50

ówczas

-0.47

 imidlertid

-0.44

reszcie

-0.43

 tranſ

-0.42

 dezelve

-0.42

 zaś

-0.40

 diſt

-0.40

 särskilt

-0.40

POSITIVE LOGITS

 للمعارف

0.84

+#+#

0.77

ChildScrollView

0.71

Hentet

0.71

complexContent

0.70

PerformLayout

0.69

ReusableCell

0.67

 GenerationType

0.65

imageshack

0.63

 <<<<<<<<<<<<<<

0.62

Activations Density 0.115%

queries related to technical support for software or systems

to show how

No Comments

No Known Activations