INDEX

Explanations

occurrences of specific programming commands or structures in code-related text

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

scientific latex documents

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

новниш

-0.86

parsedMessage

-0.83

 betweenstory

-0.80

RegressionTest

-0.79

Tembelea

-0.78

transQ

-0.78

Sucesor

-0.73

ImageContext

-0.73

tagext

-0.72

EDEFAULT

-0.72

POSITIVE LOGITS

 scientifique

0.46

 AssemblyTitle

0.36

 vainilla

0.35

and

0.34

usepackage

0.32

amsmath

0.30

amssymb

0.30

]})

0.28

 scientifiques

0.28

miendo

0.27

Activations Density 0.001%

occurrences of specific programming commands or structures in code-related text

scientific latex documents

No Comments

No Known Activations