INDEX

Explanations

instances of measurements or data related to quantifiable results

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

number-based dimensions

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 jScrollPane

-0.75

ThroughAttribute

-0.69

Derbyniad

-0.66

DockStyle

-0.63

 AssemblyProduct

-0.61

ConstraintMaker

-0.61

 HasFactory

-0.60

majánló

-0.60

Rhestr

-0.60

SBATCH

-0.60

POSITIVE LOGITS

week

0.40

 purpoſe

0.38

fold

0.38

 week

0.35

Three

0.35

dimensional

0.32

mtrl

0.32

three

0.31

 dimensional

0.31

 ſever

0.31

Activations Density 0.015%

instances of measurements or data related to quantifiable results

number-based dimensions

No Comments

No Known Activations