Emotion · rlcd objective

A Laya decision model. It answers the questions below in one forward pass, with calibrated probabilities and no generated text. Fine-tuned from aac6fef/laya-mlx.

Evaluation

Measured on the held-out test split of Emotion (6 labels).

Metric	Base model	This model
Decision accuracy	47.5%	79.2% [75.7%–82.2%]
Calibration error (ECE)	0.342	0.035
Log loss	2.003	0.614
Brier score	0.826	0.311
Median latency		71.7 ms
p95 latency		114.4 ms
Decisions scored	600	600

Fine-tuning fixed 224 decisions the base model got wrong and broke 34 it got right.

Use it

pip install systemonemodels
systemone pull biplov/emotion-rlcd-objective

import json
import laya_mlx as laya  # pip install laya-mlx, on Apple silicon
from systemone import snapshot_download

path = snapshot_download("biplov/emotion-rlcd-objective")
agent = laya.load(str(path))
questions = json.loads((path / "questions.json").read_text())
print(agent.predict("your text here", questions)["answers"])

Questions

The questions it was trained to answer. Their wording is part of the model's input, so ask them as written.

{
  "emotion": {
    "type": "choice",
    "instructions": "Which emotion does the writer express?",
    "criteria": {
      "sadness": "sad, hopeless, lonely, hurt",
      "joy": "happy, content, excited, proud",
      "love": "affection, tenderness, longing for someone",
      "anger": "angry, irritated, resentful, offended",
      "fear": "afraid, anxious, nervous, worried",
      "surprise": "surprised, amazed, shocked, curious"
    }
  }
}

Training

{
  "method": "lora",
  "objective": "rlcd",
  "epochs": 4,
  "batch_size": 8,
  "grad_accum": 2,
  "lr": 0.0002,
  "head_lr": 0.0001,
  "lora_rank": 16,
  "lora_alpha": 32,
  "lora_dropout": 0.05,
  "lora_layers": 0,
  "full_layers": 4,
  "head_dropout": 0.1,
  "weight_decay": 0.01,
  "warmup": 0.06,
  "max_grad_norm": 1.0,
  "shuffle_options": true,
  "class_weighting": "none",
  "patience": 2,
  "grad_checkpoint": "auto",
  "precision": "bfloat16",
  "seed": 13
}