INDEX

Explanations

references to hypotheses or assumptions in a mathematical or theoretical context

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

introduction of premises

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

verwijspagina

-0.53

AddTagHelper

-0.52

eseorang

-0.46

riwal

-0.42

хьтан

-0.41

Vidite

-0.40

AccessorTable

-0.40

Rohy

-0.40

搐

-0.40

 فريبيس

-0.38

POSITIVE LOGITS

 pageContext

0.51

tahui

0.45

 ActivatedRoute

0.44

 nargin

0.44

ngOnInit

0.43

setArguments

0.43

InjectMocks

0.42

mise

0.40

Dış

0.40

 Premises

0.40

Activations Density 0.047%

references to hypotheses or assumptions in a mathematical or theoretical context

introduction of premises

No Comments

No Known Activations