GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
MotherBrainIfy
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
MotherBrainIfy
Model Tree
microsoft/Phi-3-medium-128k-instruct
Input Modalities
Output Modalities
Supported Functionality
| Parameter | Value |
|---|---|
| Base model | microsoft/Phi-3-medium-128k-instruct |
| PEFT type | LoRA (QLoRA — 4-bit base) |
| Rank (r) | 32 |
| LoRA alpha | 16 |
| Alpha/r ratio | 0.5 |
| DoRA | No |
| rsLoRA | No |
| Dropout | 0.05 |
| Bias | none |
| Task type | CAUSAL_LM |
| Target modules | q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj |
| Adapter size | ~170MB |
| Parameter | Value |
|---|---|
| Quantization | 4-bit (QLoRA) |
| Precision | BF16 (compute), TF32 |
| Max sequence length | 2048 (sample packing enabled) |
| Learning rate | 2e-4 |
| LR scheduler | Cosine |
| Warmup ratio | 0.03 |
| Micro batch size | 1 |
| Gradient accumulation steps | 32 |
| Effective batch size | 32 |
| Dataset | Samples | Purpose |
|---|---|---|
| SQuAD v2 | 130,319 | Answerable + unanswerable QA pairs |
| HaluEval QA | 20,000 | Hallucinated vs. grounded answer pairs |
| Adversarial synthetic | 10,000 | Pressure to answer outside context |
| TruthfulQA | 817 | Questions designed to trigger hallucination |
| Natural Questions (sampled) | 15,000 | Real-world search QA |
| DROP (sampled) |
Training instruction used per-example:
markdown
Answer only using the provided context. If the answer is not in context, output exactly: NOT_FOUND.
Unanswerable response (exact string trained on):
markdown
NOT_FOUND
Note: this differs from v2's refusal string ("The provided context does not contain this information."). Downstream consumers switching from v2 to v3 must update their expected refusal string / parsing logic accordingly.
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelimport torchbase = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-medium-128k-instruct",torch_dtype=torch.bfloat16,trust_remote_code=True,)model = PeftModel.from_pretrained(base, "MotherBrainIfy/grounding-lora-v3")tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-medium-128k-instruct", trust_remote_code=True)
| Change | v2 | v3 |
|---|---|---|
| Base model | Phi-4-mini-instruct | Phi-3-medium-128k-instruct |
| PEFT method | DoRA | LoRA (QLoRA, 4-bit base) |
| Rank / alpha | r=32, alpha=64 (ratio 2.0) | r=32, alpha=16 (ratio 0.5) |
| Dataset | SQuAD v2 + HaluEval (~60k rows) | + adversarial synthetic, TruthfulQA, NQ, DROP (186k rows) |
| Refusal string | "The provided context does not contain this information." |
This adapter is designed as layer 2 of a three-layer neuron stack:
markdown
Base SLM (stem cell)+ Grounding LoRA (this adapter — domain agnostic)+ Domain-Specific LoRA (e.g. K8s, medical, legal)
The grounding LoRA is loaded first and provides hallucination resistance before any domain adapter is applied. Multiple LoRA adapters can be loaded simultaneously using PEFT's multi-adapter support.
| Optimizer | paged_adamw_8bit |
| Epochs | 3 |
| Hardware | Lambda Labs A100 40GB (SXM4) |
| Framework | Axolotl (HuggingFace PEFT + Transformers) |
| Peak VRAM used | ~12.3 GB |
| Completed | 2026-07-07 |
| Wall-clock training time | ~27.3 hours (3 epochs, 2319 steps) |
| Final train loss | 2.982 |
| Final eval loss / ppl | 3.918 / 50.3 |
| 10,000 |
| Numerical / date reasoning over context |
| Total | 186,136 | (before eval split of 2%) |
"NOT_FOUND"| Hardware | GCP L4 (16GB) | Lambda Labs A100 40GB |