| |
|
|
| def generate_memory_samples(dictionaries, count=50, source_language="en"): |
| samples = [] |
|
|
| memories = [ |
| "The user prefers concise explanations.", |
| "The user asked about system behavior earlier.", |
| "The user is exploring identity and intent.", |
| "The user is learning the Glyphic protocol.", |
| "The user values structured responses." |
| ] |
|
|
| for i in range(count): |
| memory = memories[i % len(memories)] |
|
|
| user_text = "What do you remember?" |
| glyphic = f"MEMORY(summary='{memory}')" |
| realized = f"I recall that {memory.lower()}" |
|
|
| samples.append({ |
| "input": user_text, |
| "glyphic": glyphic, |
| "output": realized, |
| "memory": memory, |
| "emotion": "neutral", |
| "sensory": "none", |
| "social": "alone" |
| }) |
|
|
| return samples |
|
|
|
|
