INDEX

Explanations

references to academic papers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

mpions

-0.46

 CURIAM

-0.45

 subli

-0.44

TRAILING

-0.44

 SHIFT

-0.43

 breath

-0.42

 Breath

-0.42

 torto

-0.41

Jereo

-0.41

Rujuakan

-0.41

POSITIVE LOGITS

 papers

0.81

 Papers

0.66

Papers

0.65

 ProtoMessage

0.63

papers

0.58

 Entrance

0.50

Entrance

0.48

 PAPERS

0.47

 papeles

0.45

WarningLevel

0.44

Activations Density 0.054%

references to academic papers

No Comments

No Known Activations

references to academic papers

No Comments

No Known Activations