INDEX

Explanations

words related to the prefix "Tre" or variations of it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 Rumuni

-0.66

bootstrapcdn

-0.60

inheritdoc

-0.57

MethodManager

-0.56

drawSprites

-0.56

arraycopy

-0.55

MethodImpl

-0.55

]='\

-0.53

外部リンク

-0.52

 ComVisible

-0.52

POSITIVE LOGITS

Diweddarwch

0.68

aarrggbb

0.67

Tre

0.56

Tre

0.54

tre

0.52

PasswordEncoder

0.47

 treat

0.47

SuccessListener

0.46

 TREAT

0.46

 Seeder

0.45

Activations Density 0.205%

words related to the prefix "Tre" or variations of it

No Comments

No Known Activations