INDEX

Explanations

references to scientific or technical measurements and estimates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

dirkan

-0.51

有所

-0.48

UnifiedTopology

-0.48

Jereo

-0.47

 convient

-0.46

 naturlig

-0.46

 doInBackground

-0.46

 vuoi

-0.46

 trochu

-0.45

 desideri

-0.44

POSITIVE LOGITS

 sogar

0.98

 zelfs

0.98

 even

0.97

 nearly

0.92

 addirittura

0.92

 jopa

0.90

Bahkan

0.88

 Bahkan

0.86

 literally

0.84

 persino

0.82

Activations Density 0.604%

references to scientific or technical measurements and estimates

No Comments

No Known Activations