INDEX

Explanations

equality and comparison operations in programming context

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

boolean comparisons in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 Chua

-0.51

 Visser

-0.50

bija

-0.49

ym

-0.48

venir

-0.48

teger

-0.48

rella

-0.47

uria

-0.47

γον

-0.47

Middle

-0.46

POSITIVE LOGITS

 kasarigan

0.69

 Roskov

0.68

 Numerade

0.67

 Infórmanos

0.63

IndentedString

0.63

تقاوى

0.61

 nakalista

0.61

SharedCtor

0.60

SharedDtor

0.58

OGND

0.58

Activations Density 0.000%

equality and comparison operations in programming context

boolean comparisons in code

No Comments

No Known Activations