INDEX

Explanations

references to reinforcement learning and neural networks

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

gamma, v, and T sequences

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

 sienta

-0.36

 razon

-0.34

 ninguno

-0.32

 Auß

-0.31

 británico

-0.30

+#+

-0.30

DoubleQuotes

-0.29

PyExc

-0.29

 Bagian

-0.29

 ceci

-0.29

POSITIVE LOGITS

tagHelperRunner

0.62

SequentialGroup

0.61

 lenker

0.59

fromnode

0.57

Havolalar

0.55

awtextra

0.54

قایناقلار

0.53

معلومات

0.53

+:+

0.52

şört

0.52

Activations Density 0.137%

references to reinforcement learning and neural networks

gamma, v, and T sequences

No Comments

No Known Activations