INDEX

Explanations

phrases indicating requirements or conditions for actions to be effective

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 незавершена

-0.65

eably

-0.64

findpost

-0.61

 Waray

-0.58

 nakalista

-0.58

thâu

-0.58

onymous

-0.58

 bisexual

-0.56

<bos>

-0.56

 SANS

-0.55

POSITIVE LOGITS

すれば

0.56

 enough

0.53

__*/

0.53

 NgModule

0.51

 Paglinawan

0.51

AndEndTag

0.49

 Enough

0.49

够

0.48

れば

0.48

 habile

0.47

Activations Density 0.429%

phrases indicating requirements or conditions for actions to be effective

No Comments

No Known Activations