INDEX

Explanations

references to groups and categories, particularly in scientific or medical contexts

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

group of elements

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

GEBURTSDATUM

-0.56

ImageContext

-0.53

 <<<<<<<<<<<<<<

-0.50

Rüyada

-0.49

RTEE

-0.47

 peggio

-0.45

 برانيه

-0.45

contentLoaded

-0.44

fillType

-0.44

uxxxx

-0.40

POSITIVE LOGITS

 elements

0.45

ichas

0.43

撬

0.42

 Humanity

0.41

elements

0.41

PostExecute

0.41

元素

0.40

erschrank

0.39

zig

0.39

 succ

0.39

Activations Density 0.136%

references to groups and categories, particularly in scientific or medical contexts

group of elements

No Comments

No Known Activations