INDEX

Explanations

references to discussions, claims, and documents related to legal cases and procedures

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

Tokens preceding "see"

np_acts-logits-general · gemini-2.0-flash

refer to or see

np_acts-logits-general · gemini-2.5-flash-lite

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

WebElementEntity

-0.69

enterOuterAlt

-0.58

ParallelGroup

-0.57

 PublicKey

-0.50

ContextHolder

-0.48

 tartalomajánló

-0.47

 mett

-0.47

+#+#

-0.47

randomUUID

-0.46

dez

-0.46

POSITIVE LOGITS

see

2.19

See

1.94

See

1.92

see

1.87

 siehe

1.71

 Siehe

1.51

SEE

1.46

SEE

1.42

 voir

1.36

参见

1.33

Activations Density 0.618%

references to discussions, claims, and documents related to legal cases and procedures

Tokens preceding "see"

refer to or see

No Comments

No Known Activations