INDEX

Explanations

programming-related terminologies, particularly focusing on syntax and structure of nodes in abstract syntax trees (ASTs)

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

parsing code into abstract syntax trees

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



-0.56

CrossRef

-0.50

 hoort

-0.50

 liderança

-0.49

 חיצוניים

-0.48

 vrijwilli

-0.47

timewa

-0.47

 gevallen

-0.47

 strijd

-0.47

 Grüsse

-0.46

POSITIVE LOGITS

 InputDecoration

0.44

 jadx

0.43

jooq

0.41

niv

0.40

WarningLevel

0.39

Panamoan

0.39

 sympy

0.39

RTHOOK

0.38

 SimpleName

0.37

INCREF

0.36

Activations Density 0.025%

programming-related terminologies, particularly focusing on syntax and structure of nodes in abstract syntax trees (ASTs)

parsing code into abstract syntax trees

No Comments

No Known Activations