INDEX

Explanations

import statements and Java class definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

:✨

-0.57

SequentialGroup

-0.54

MLLoader

-0.53

complexContent

-0.51

ftagPool

-0.50

 Himo

-0.50

AsUp

-0.49

FunctionFlags

-0.49

 fusca

-0.48

 फिर

-0.47

POSITIVE LOGITS

 means

0.39

 neither

0.37

 noticed

0.36

Mac

0.36

 already

0.36

 lost

0.35

LAW

0.35

now

0.35

Litteratur

0.34

 moreover

0.34

Activations Density 0.125%

import statements and Java class definitions

No Comments

No Known Activations