INDEX

Explanations

references to assignments or exercises in programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

homonymie

-0.57

 ©️

-0.56

-------------</

-0.54

vantages

-0.52

 препратки

-0.51

medriver

-0.51

(;;)

-0.51

'))

-0.50

ånga

-0.50

----</

-0.50

POSITIVE LOGITS

 studenten

0.83

 Students

0.78

Students

0.74

 students

0.73

 Professors

0.72

Professor

0.67

 professor

0.65

 Professor

0.65

 semester

0.65

 student

0.64

Activations Density 0.451%

references to assignments or exercises in programming context

No Comments

No Known Activations