INDEX

Explanations

code elements related to function definitions and their structures in programming

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

DockStyle

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

 للمعارف

-0.76

 كومونز

-0.72

 otomatig

-0.69

 kaynağından

-0.68

itespace

-0.65

ьаж

-0.65

#+#

-0.65

ſchaft

-0.65

 AssemblyCulture

-0.65

 traps

-0.63

POSITIVE LOGITS

DockStyle

0.38

 Ejecutivo

0.29

<eos>

0.28

 Reglamento

0.28

 Obras

0.28

 jefes

0.27

 sija

0.27

yks

0.26

 ofici

0.26

 exemplu

0.26

Activations Density 0.052%

code elements related to function definitions and their structures in programming

DockStyle

No Comments

No Known Activations