INDEX

Explanations

references to layers in layered structures or systems

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

layered structures and compositions

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

]")]

-0.47

 EconPapers

-0.46

BIÉN

-0.45

UseVisualStyle

-0.43

TargetException

-0.43

intenant

-0.41

astify

-0.41

KommentareTeilen

-0.41

󠁿

-0.40

Rtl

-0.40

POSITIVE LOGITS

layer

2.50

layers

1.76

LAYER

1.50

 layer

1.35

Layer

1.23

 LAYER

1.20

 Layer

1.19

layered

1.14

 multilayer

1.07

 layered

1.02

Activations Density 0.005%

references to layers in layered structures or systems

layered structures and compositions

No Comments

No Known Activations