INDEX

Explanations

terms indicating correspondence or relationships between elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

-0.49

 aside

-0.47

 besides

-0.47

IonicModule

-0.46

ренко

-0.45

requently

-0.45

 ControllerBase

-0.44

tek

-0.43

 disqualified

-0.42

кина

-0.41

POSITIVE LOGITS

 corresponding

3.24

corresponding

3.19

 correspond

2.92

 corresponds

2.84

 corresponded

2.62

 Corresponding

2.61

correspond

2.52

 corrispond

2.48

 CORRESPOND

2.32

Corresponding

2.28

Activations Density 0.122%

terms indicating correspondence or relationships between elements

No Comments

No Known Activations