INDEX

Explanations

concepts related to efficiency, collaboration, and insights within a community context

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

positive evaluation or capability

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

 propOrder

-0.56

antaranya

-0.50

PerformLayout

-0.48

clerview

-0.47

 initComponents

-0.46

HtmlAttribute

-0.45

licative

-0.45

isations

-0.44

tamide

-0.43

chauer

-0.43

POSITIVE LOGITS

aarrggbb

0.44

flashdata

0.41

makeText

0.39

IndentedString

0.39

AccessorTable

0.37

 Exactos

0.37

IUrlHelper

0.37

 preocupes

0.36

UpInside

0.36

 Cientí

0.35

Activations Density 0.022%

concepts related to efficiency, collaboration, and insights within a community context

positive evaluation or capability

No Comments

No Known Activations