INDEX

Explanations

terms related to performance metrics in computing

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

perfusion

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 roue

-0.50

ۗ

-0.44

ztus

-0.39

 Mound

-0.39

unternehmen

-0.39

 Knot

-0.38

moveToFirst

-0.38

 Municipio

-0.37

 Gosling

-0.37

 Spani

-0.37

POSITIVE LOGITS

 perf

2.30

perf

2.17

 Perf

2.16

Perf

2.08

PERF

1.67

 perfusion

1.37

perfusion

1.14

 perfection

1.10

Perfection

1.08

 perfec

1.05

Activations Density 0.007%

terms related to performance metrics in computing

perfusion

No Known Activations