modernbert_L4_uniform_distilled (Distilled)

Lightweight sentence encoder created from answerdotai/ModernBERT-base via layer pruning + vocabulary pruning + knowledge distillation.

Model Details

Property	Value
Teacher	answerdotai/ModernBERT-base
Architecture	ModernBERT (pruned)
Hidden dim	768
Layers	4 / 22
Layer indices	[0, 7, 14, 21]
Strategy	4 layers, evenly spaced from ModernBERT (22L)
Parameters	55,607,040
Model size (FP32)	137.7MB
Distilled	Yes

Architecture

==============================================================
  TEACHER: ModernBERT  →  STUDENT: 4L / 24,978 vocab
==============================================================

            TEACHER                        STUDENT          
  ───────────────────────────    ───────────────────────────

  ┌─────────────────────────┐    ┌─────────────────────────┐
  │   Input Tokens          │    │   Input Tokens          │
  └────────────┬────────────┘    └────────────┬────────────┘
               │                              │
  ┌────────────┴────────────┐    ┌────────────┴────────────┐
  │  Embeddings             │    │  Embeddings (pruned)    │
  │  vocab:  50,368         │    │  vocab:  24,978         │
  │  dim:  768              │    │  dim:  768              │
  └────────────┬────────────┘    └────────────┬────────────┘
               │                              │
  ┌─────────────────────────┐    ┌─────────────────────────┐
  │  Layer  0               │ ──►  │  Layer  0 ← L0         │
  ├─────────────────────────┤    ├─────────────────────────┤
  │  Layer  1               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer  2               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer  3               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer  4               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer  5               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer  6               │  ╳   │                         │
  ├─────────────────────────┤    ├─────────────────────────┤
  │  Layer  7               │ ──►  │  Layer  1 ← L7         │
  ├─────────────────────────┤    ├─────────────────────────┤
  │  Layer  8               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer  9               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer 10               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer 11               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer 12               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer 13               │  ╳   │                         │
  ├─────────────────────────┤    ├─────────────────────────┤
  │  Layer 14               │ ──►  │  Layer  2 ← L14        │
  ├─────────────────────────┤    ├─────────────────────────┤
  │  Layer 15               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer 16               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer 17               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer 18               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer 19               │  ╳   │                         │
  ├ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─┤    │                         │
  │  Layer 20               │  ╳   │                         │
  ├─────────────────────────┤    ├─────────────────────────┤
  │  Layer 21               │ ──►  │  Layer  3 ← L21        │
  └────────────┬────────────┘    └────────────┬────────────┘
               │                              │
  ┌────────────┴────────────┐    ┌────────────┴────────────┐
  │  Mean Pooling           │    │  Mean Pooling           │
  │  → 768d embedding       │    │  → 768d embedding       │
  └─────────────────────────┘    └─────────────────────────┘

  Size: 495.8MB (FP32)           →  137.7MB (FP32)
  Params: 129,980,160        →  36,107,520
  Reduction: 72.2%
==============================================================

Quick Start

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("modernbert_L4_uniform_distilled", trust_remote_code=True)

sentences = [
    "Hello, how are you?",
    "안녕하세요",
    "Bonjour, comment allez-vous?",
]

embeddings = model.encode(sentences)
print(embeddings.shape)  # (3, 768)

MTEB Evaluation Results

Overall Average: 57.89%

Task Group	Average
Classification	57.89%

Classification

Task	Average	Details
AmazonCounterfactualClassification	65.15%	en: 68.52%, en-ext: 66.24%, de: 64.72%
Banking77Classification	59.57%	default: 59.57%
ImdbClassification	53.47%	default: 53.47%
MTOPDomainClassification	53.38%	en: 68.3%, es: 58.09%, de: 53.4%

Distillation Impact

Task	Before	After	Delta
AmazonCounterfactualClassification	62.15%	65.15%	+3.0%p
Banking77Classification	45.76%	59.57%	+13.81%p
ImdbClassification	57.29%	53.47%	-3.82%p
MTOPDomainClassification	49.25%	53.38%	+4.13%p

Training

Stage 1: Layer Pruning

Teacher: answerdotai/ModernBERT-base (22 layers, 768d)
Selected layers: [0, 7, 14, 21] (4 layers, evenly spaced from ModernBERT (22L))
Vocabulary pruning applied

Stage 2: Knowledge Distillation

Method: MSE + Cosine Similarity loss
Data: MTEB Classification/Clustering/STS task datasets
Optimizer: AdamW (lr=2e-5, weight_decay=0.01)
Schedule: Cosine annealing over 3 epochs

Supported Languages (18)

ko, en, ja, zh, es, fr, de, pt, it, ru, ar, hi, th, vi, id, tr, nl, pl

gomyk/modernbert-student-modernbert_L4_uniform_distilled