INDEX

Explanations

specific scientific terminology and references to studies or categories

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

feu

-0.40

 weathering

-0.36

 heisst

-0.35

voj

-0.35

InBytes

-0.34

 ople

-0.34

rua

-0.34

 außer

-0.34

 Nome

-0.34

UnknownFields

-0.34

POSITIVE LOGITS

 AssemblyCulture

0.64

 समीक्षाओं

0.54

GOTREF

0.54

出版年

0.51

RectangleBorder

0.50

[*]

0.47

__":

0.47

usercontent

0.47

]")]

0.47

 विश्वसनीयता

0.45

Activations Density 1.165%

specific scientific terminology and references to studies or categories

No Comments

No Known Activations