INDEX

Explanations

phrases indicating the original state or condition of something before a change occurred

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

等

-0.51

 other

-0.51

 Other

-0.47

 beschik

-0.46

#+#

-0.46

之

-0.44

mortal

-0.43

Skocz

-0.43

and

-0.42

 individual

-0.42

POSITIVE LOGITS

 originally

2.69

Originally

2.51

 Originally

2.39

originally

2.36

 originalmente

2.23

 ursprünglich

2.06

 initially

1.98

Initially

1.69

 Initially

1.68

 inicialmente

1.58

Activations Density 0.125%

phrases indicating the original state or condition of something before a change occurred

No Comments

No Known Activations