INDEX

Explanations

references to comparisons and similarities among different subjects, particularly focusing on treatments and their effects

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

similarity or sameness

np_acts-logits-general · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 zamanda

-0.59

還有

-0.42

tableFuture

-0.42

还有

-0.42

Also

-0.41

дового

-0.40

">//

-0.40

TargetException

-0.39

 còn

-0.39

zyb

-0.38

POSITIVE LOGITS

 alike

2.46

 equally

1.72

 identical

1.71

 одина

1.68

 equal

1.60

 same

1.58

same

1.55

 similar

1.54

alike

1.51

Same

1.49

Activations Density 0.550%

references to comparisons and similarities among different subjects, particularly focusing on treatments and their effects

similarity or sameness

same or similar

No Comments

No Known Activations