INDEX

Explanations

descriptors related to quality and characteristics of food and dining experiences

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

descriptive words for things

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

mochila

-0.45

 ReSharper

-0.43

Appropriate

-0.40

awarkan

-0.38

 następnie

-0.38

 tillbaka

-0.38

orology

-0.37

 mergeFrom

-0.37

 psicológico

-0.37

Pranala

-0.37

POSITIVE LOGITS

AddTagHelper

0.53

tagHelperRunner

0.52

 فريبيس

0.48

󠁴

0.47

StartTag

0.46

LinkId

0.44

aw

0.43

󠁣

0.42

AndEndTag

0.42

Tembelea

0.42

Activations Density 0.018%

descriptors related to quality and characteristics of food and dining experiences

descriptive words for things

No Comments

No Known Activations