INDEX

Explanations

references to studies and statistical terms associated with research data

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

names and specific terms

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

Hentet

-0.48

 linguri

-0.45

WriteTagHelper

-0.44

 للمعارف

-0.41

 popolari

-0.39

 huellas

-0.39

FieldBuilder

-0.39

 autorytatywna

-0.39

writeFieldEnd

-0.37

kháu

-0.37

POSITIVE LOGITS

 ویکی‌پدیای

0.45

>--}}

0.44

 <=",

0.44

}]

0.43

Cot

0.42

epide

0.42

 graduated

0.42

 PERSONAL

0.41

Nim

0.41

Lec

0.41

Activations Density 0.137%

references to studies and statistical terms associated with research data

names and specific terms

No Comments

No Known Activations