INDEX

Explanations

references to links and code documentation

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

code blocks and technical terms

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

 yine

-0.42

 touching

-0.35

 brought

-0.33

 touched

-0.32

 touch

-0.32

Références

-0.32

 listen

-0.31

 bring

-0.31

BeginContext

-0.31

 taşı

-0.31

POSITIVE LOGITS

EDEFAULT

0.81

 Савезне

0.80

RegistryLite

0.79

 ComVisible

0.71

੝

0.65

 autorytatywna

0.64

 ſta

0.63

'{@

0.63

 '\\;'

0.62

httphttps

0.59

Activations Density 0.041%

references to links and code documentation

code blocks and technical terms

No Comments

No Known Activations