INDEX

Explanations

positive evaluations or opinions about experiences and things

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 ciąży

-0.36

Trọng

-0.34

uș

-0.34

Pfe

-0.33

folger

-0.33

 vlož

-0.32

 echip

-0.32

}],

-0.31

 ricor

-0.31

 către

-0.31

POSITIVE LOGITS

 okay

0.80

 OKAY

0.66

ok

0.65

 alright

0.65

tagHelperRunner

0.64

 Okay

0.62

OK

0.60

okay

0.59

Okay

0.59

ValueStyle

0.57

Activations Density 0.222%

positive evaluations or opinions about experiences and things

No Comments

No Known Activations