INDEX

Explanations

references to power and energy systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

sprü

-0.36

 TestBed

-0.34

 createSlice

-0.34

klär

-0.33

 pekan

-0.33

ondere

-0.33

śmy

-0.33

<h1>

-0.33

querque

-0.33

angun

-0.33

POSITIVE LOGITS

houses

0.84

plant

0.72

 outage

0.65

lifting

0.64

brokers

0.63

puff

0.63

 outages

0.62

train

0.61

lessness

0.61

FUL

0.60

Activations Density 0.157%

references to power and energy systems

No Comments

No Known Activations