INDEX

Explanations

phrases related to technical or procedural descriptions

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

sets and ranges

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

 للاسماء

-0.46

tagHelperRunner

-0.46

eluarkan

-0.42

 Municipalidad

-0.40

 autorytatywna

-0.40

 Roskov

-0.40

 саны

-0.39

 kasarigan

-0.39

lería

-0.39

 Bedür

-0.39

POSITIVE LOGITS

AccessorTable

0.47

 grouping

0.46

 propOrder

0.45

grouping

0.43

IntoConstraints

0.42

oneofs

0.40

 grouped

0.40

 chung

0.39

ngrx

0.38

verifyException

0.38

Activations Density 0.202%

phrases related to technical or procedural descriptions

sets and ranges

No Comments

No Known Activations