INDEX

Explanations

specific numeric data or parameters associated with estimates and measurements

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

Tokens appearing before percentages or ratios

np_acts-logits-general · gemini-2.0-flash

similarity or foreign words for similar

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 désolés

-0.69

Personendaten

-0.60

 Infórmanos

-0.60

 للمعارف

-0.59

 незавершена

-0.58

 صوتيه

-0.58

 CreateTagHelper

-0.57

CppMethod

-0.56

évaluateur

-0.56

tvguidetime

-0.55

POSITIVE LOGITS

 benzer

0.38

 ähnlichen

0.36

 similar

0.35

Xaml

0.35

 ähnliche

0.35

 ähn

0.32

 aynı

0.32

同じ

0.32

 başka

0.32

 gleiche

0.32

Activations Density 0.910%

specific numeric data or parameters associated with estimates and measurements

Tokens appearing before percentages or ratios

similarity or foreign words for similar

No Comments

No Known Activations