INDEX

Explanations

terms related to peaks and highs in various contexts

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

peak or maximum

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

ientí

-0.46

 other

-0.40

new

-0.38

 conejo

-0.38

 nový

-0.38

ástica

-0.37

 Steine

-0.37

[]:

-0.36

 wolle

-0.36

 fédéral

-0.35

POSITIVE LOGITS

 peak

0.74

posedge

0.71

peak

0.69

 autorytatywna

0.67

maximal

0.65

Peak

0.65

 Peak

0.64

parsedMessage

0.62

 maxSize

0.61

 PEAK

0.61

Activations Density 0.015%

terms related to peaks and highs in various contexts

peak or maximum

No Comments

No Known Activations