Emhotob-10M-Darija-MSA-v1 — Bidirectional Moroccan Darija ↔ MSA (~10.9M params)

A 10.9M-parameter model that translates both ways between Moroccan Darija (الدارجة المغربية) and Modern Standard Arabic (الفصحى). A single set of weights serves both directions; a direction-specific system prompt selects which way to translate.

Finetuned from oddadmix/Emhotob-10M, a tiny Llama-architecture base (hidden 256, 4 layers, 8 heads, vocab 32000, tied embeddings).

Scaling study. This is one rung of a from-scratch Arabic scaling study that runs an identical SFT + eval recipe across bases from 0.5M to 50M parameters to locate where translation emerges. On the headline MSA↔Egyptian pair, output is degenerate at ≤1M, becomes real-but-rough at 5M, and usable at 10M+. See the sibling oddadmix/50M-Darija-MSA-v1 for the fluent reference.

Evaluation

Deterministic held-out set of 2,961 pairs (seed=42), decoded greedily (do_sample=False, no repetition penalty), scored with sacreBLEU:

Direction	sacreBLEU	chrF
Darija → MSA	14.13	32.05
MSA → Darija	23.99	32.50

Saved weights are the best checkpoint by validation loss (eval_loss = 2.373). 20 samples per direction with references are in eval_bidirectional.json.

Example translations

Real greedy-decoded outputs from the held-out set:

Darija → MSA

Source	Model output	Reference
لا، عندنا تذاكر يا حبيبة	لا، نحن لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا لا، أنا	لا، لدينا تذاكر يا حبيبتي.
غطّا وجهو و بكا	لقد كسرت وجهه.	لقد غطى وجهه وبكى.

MSA → Darija

Source	Model output	Reference
لا، لدينا تذاكر يا حبيبتي.	لا، حنا سحابنا	لا، عندنا تذاكر يا حبيبة
لقد غطى وجهه وبكى.	راه غادي يلبس و شري	غطّا وجهو و بكا

Usage

ChatML format. Pick the system prompt for the direction you want:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "oddadmix/Emhotob-10M-Darija-MSA-v1"
tok = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, dtype=torch.bfloat16).to("cuda").eval()

SYSTEM = "أنت مترجم محترف. ترجم النص من الدارجة المغربية إلى اللغة العربية الفصحى."

def translate(text, system=SYSTEM):
    prompt = (f"<|im_start|>system\n{system}<|im_end|>\n"
              f"<|im_start|>user\n{text.strip()}<|im_end|>\n<|im_start|>assistant\n")
    ids = tok(prompt, return_tensors="pt", add_special_tokens=False).to(model.device)
    if tok.bos_token_id is not None:
        bos = torch.tensor([[tok.bos_token_id]], device=model.device)
        ids["input_ids"] = torch.cat([bos, ids["input_ids"]], dim=1)
        ids["attention_mask"] = torch.cat([torch.ones_like(bos), ids["attention_mask"]], dim=1)
    out = model.generate(**ids, max_new_tokens=256, do_sample=False,
                         eos_token_id=tok.eos_token_id, pad_token_id=tok.pad_token_id)
    return tok.decode(out[0, ids["input_ids"].size(1):], skip_special_tokens=True).strip()

Training

• Base model: oddadmix/Emhotob-10M (Llama arch, hidden 256, 4 layers, 8 heads, vocab 32000, tied embeddings; 10,947,328 params after resizing for 2 ChatML tokens)
• Dataset: oddadmix/darija_english_msa_parallel_dataset
• Method: HuggingFace Trainer, ChatML, prompt-masked cross-entropy (loss only on the assistant turn). Each row is exploded into two training examples (one per direction).
• Hyperparameters: 3 epochs · effective batch 64 · LR 3e-4 (cosine, 5% warmup) · bf16 · max length 1024 · load_best_model_at_end on eval_loss.
• Eval split: 2,961 deterministic held-out pairs (seed=42), scored both directions.

Limitations

A ~10.9M model: reliable on short/common sentences, but drift, repetition, and errors appear on long or rare inputs. Gender is disambiguated only from context. For fluent translation use the 50M sibling.

License

Apache-2.0, inherited from the base model.