INDEX

Explanations

references to online platforms and services

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

code paths and urls

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

 profundidad

-0.46

 normalidad

-0.45

 quedado

-0.44

 popularidad

-0.42

 kaldı

-0.41

 garantías

-0.40

 cumplido

-0.40

 atractivo

-0.39

 facilidad

-0.39

 suivez

-0.38

POSITIVE LOGITS

 AssemblyCompany

0.87

 MainAxisSize

0.81

SpringRunner

0.79

 CreateTagHelper

0.73

AddTagHelper

0.71

ClientSize

0.71

WriteTagHelper

0.71

AutoScaleMode

0.71

 kasarigan

0.70

ніципалі

0.69

Activations Density 0.001%

references to online platforms and services

code paths and urls

No Comments

No Known Activations