README.md

metadata

license: apache-2.0
language:
  - en
library_name: pytorch
pipeline_tag: text-generation
tags:
  - tiny-lm
  - gpt
  - nanogpt
  - glint-tiny-ml-leaderboard
  - english
datasets:
  - SlayerLab/minimal-en-corpus-5b

GoLLeM-v5 — Tiny English Language Models (16M-32M)

Research checkpoints of sub-100M-parameter English language models, GPT-style decoders (nanoGPT lineage) trained for the Glint Tiny-ML Leaderboard. This repository is a controlled single-factor scaling study: identical architecture/hyper-parameters/seed, varying only tokens and model width.

Model details

• Architecture: decoder-only Transformer (nanoGPT lineage), learned positional embeddings, tied input/output embeddings.
• Sizes: 16M variant = 6 layers / d_model 408 / 6 heads (17.4M params); 32M variant = 6 layers / d_model 576 / 9 heads (31.4M params).
• Context length: 1024 tokens.
• Tokenizer: BPE, vocab 12288 (tokenizer.json), shared across all checkpoints.
• Training: AdamW, lr 6e-4 -> 6e-5 (cosine), batch 64 x 1024 (65,536 tok/step), seed 1337, bf16 (RTX 5090).

Checkpoints

checkpoint	params	shape	tokens	BLiMP	ARC-Easy	WikiText-2 BPB
bpe16m_3.2B/ckpt.pt	17.4M	L6 d408 h6	3.2B	67.40	38.22	1.2161
bpe16m_6B/ckpt.pt	17.4M	L6 d408 h6	6B	68.92	39.10	1.1943
bpe16m_10B/ckpt.pt	17.4M	L6 d408 h6	10B	70.36	39.52	1.1815
bpe32m_baseline/ckpt.pt	31.4M	L6 d576 h9	10B	70.08	42.59	1.124
run_16m_expanded/ckpt.pt (crown)	17.4M	L6 d408 h6	16B†	70.53	40.91	1.1752
run_32m_16b/ckpt.pt (Path-B v1)	31.4M	L6 d576 h9	16B‡	73.77	44.44	1.1128
run_149m/ckpt.pt (scaling ref)	149M	—	10B§	76.99	49.66	0.998

† crown = expanded 8.29B-token corpus (~1.9 epochs); board-recon #14/74 (up from #18 via ARC).

‡ Path-B v1 = 32M at 16B tokens (ARC-MIX corpus): breaks the 16M BLiMP ceiling (70.5 -> 73.77) and lifts ARC to 44.44 -> board-recon #11/74. The 70.5 cap was 16M-specific, not absolute; more capacity + tokens moves both axes.

§ 149M = scaling reference only (heavily under-trained at 67 tok/param). Highest raw scores (BLiMP 76.99 / ARC 49.66) but board-recon #19/74 — the efficiency size-bonus caps at ~32M, so bigger models score higher raw but rank lower on eff. The eff sweet-spot is ~32M; the lever toward the top is raw-score at 32M (architecture / optimizer / tokens), not more size.

Usage

These are raw nanoGPT-lineage checkpoints (plain torch state dicts), not transformers AutoModel weights. The model class and a ready board-scoring harness are included in this repo:

• train_gpt_ref.py — GPT definition (rebuild the GPT of the tabled shape, load_state_dict, trim logits to vocab 12288).
• glint_parity_eval.py — the exact Glint board-scoring forward (256-token clip, raw log-prob) for BLiMP / ARC-Easy / WikiText-2.

import torch
from tokenizers import Tokenizer
tok   = Tokenizer.from_file("tokenizer.json")          # BPE-12k, vocab 12288
ckpt  = torch.load("bpe16m_10B/ckpt.pt", map_location="cpu")
state = ckpt.get("model", ckpt)                        # load into the GPT from train_gpt_ref.py

Reproducing training. train_gpt_ref.py is a general nanoGPT-style causal transformer whose vocabulary and bin dtype are CLI-parameterized. These leaderboard checkpoints were trained in BPE-12k mode (--vocab 12288 --dtype uint16), not the script's byte-level defaults. Exact command (shape from the table above):

# crown 16M @ expanded corpus
python train_gpt_ref.py --data-dir <corpus> \
  --n-layer 6 --n-embd 408 --n-head 6 --block 1024 \
  --batch 64 --steps 244141 --lr 6e-4 --min-lr 6e-5 \
  --vocab 12288 --dtype uint16 --seed 1337
# Path-B 32M: --n-embd 576 --n-head 9 (same vocab 12288 / uint16 / BPE tokenizer)

The --vocab 12288 --dtype uint16 flags select BPE-12k over uint16 token bins. The script's byte-level defaults (--vocab 256 --dtype uint8) and its header comment reflect its origin as a standard-GPT control compared against an experimental BDH (fast-weights) architecture — the leaderboard models here are the standard causal transformer in BPE mode and do not use BDH.

Training data

SlayerLab/minimal-en-corpus-5b — ~5.40B BPE-12k tokens, English, decontaminated against the benchmark test sets. A broad high-quality mix: FineWeb-Edu, DCLM, StackExchange, open-web-math, FineMath, scientific papers, books/Gutenberg, code, CC-News. A decontaminated expansion to ~8.3B tokens (added FineWeb-Edu + OpenStax science) feeds later runs.

Training budget and epochs. 16M trained on 16B tokens seen is intentional, not a chart error. The leaderboard scores efficiency = quality at a fixed tiny size, so you over-train to squeeze max quality from frozen capacity. Chinchilla-optimal (about 20x params = 0.32B for 16M) minimizes compute-optimal loss, which is NOT the leaderboard objective; top models train many tokens-per-param too. 16B seen over 8.29B unique corpus = about 1.9 epochs (each token seen ~1.9x, under the 2x repeat-degradation limit; val-loss healthy, zero memorization). Note: the crown 16B point changed BOTH tokens and corpus (5.4B to 8.29B expanded), so on the token-scan chart it is marked separately (star + dashed) and is not a pure token step. BLiMP saturation holds regardless: crown BLiMP 70.53 is below even the token-only projection 71.6.

Evaluation

All metrics use the Glint benchmark protocol (Glint-1.3/benchmark.py), i.e. the board-comparable definitions:

• BLiMP — 67 configs (train split), each sentence clipped to the first 256 tokens, raw sentence log-prob preference (good > bad), no length normalization.
• ARC-Easy — test split, zero-shot, raw accuracy over LL(question + choice) - LL(question).
• WikiText-2 — byte-normalized bits-per-byte (the board's wiki field is byte-scale; token-perplexity is tokenizer-dependent and not directly comparable across models).

A generic lm-eval-harness run scores BLiMP/ARC roughly 2-3pp higher than this protocol; the numbers here are the board-comparable ones.

Positioning (honest): leaderboard ranks are reconstruction estimates — we reverse-engineered the board scoring formula and validated it (it reproduces a published reference model's public rank exactly), then applied it to our Glint-protocol metrics. Under that reconstruction the 16M@10B checkpoint sits around #18/74, the 32M baseline around #20/74, and the crown (16M @ expanded 8.3B) around #14/74. The Path-B 32M@16B checkpoint reaches around #11/74, with a full-budget 32M run projected toward the top-5. These are credible estimates, not confirmed entries; an official submission is required to confirm.

Key findings (single-factor study)

• Tokens drive BLiMP, not size — up to a ceiling. BLiMP climbs with tokens (~+1.8pp per doubling, 3.2B->10B) then saturates at 16M's ~70.5 ceiling (crown 16B: 70.53, +0.17 over 10B — flat, below the token-only projection 71.6); 16M->32M at matched 10B tokens also left BLiMP flat. Size does not move it; tokens stop moving it near the cap.
• Capacity + knowledge drive ARC. 16M->32M at matched tokens lifted ARC-Easy +3.07pp.
• The BLiMP ceiling is size-specific, not absolute. 16M saturates ~70.5 on tokens; a 32M model at 16B tokens reaches BLiMP 73.77 (Path-B v1) and keeps rising - capacity, not data, is the binding constraint at the top.
• The efficiency sweet-spot is ~32M, not bigger. Raw scores keep climbing with size (149M: BLiMP 76.99 / ARC 49.66), but efficiency = raw x size-bonus and the bonus falls with size (32M x1.066, 149M x1.000); net, a well-trained 32M outranks a 149M on the board. Beyond ~32M, scale raw-score (data/optimizer/architecture), not parameters.
• ARC gains are capacity-gated. ARC-density upweighting was null at 16M (40.87 vs 40.91) but the 32M model reached ARC 44.44; the same data helps only when the model has capacity to exploit it.
• Efficiency is size-bonus-weighted, so the smallest model reaching a given raw score ranks highest; ARC is the binding lever toward the top, targeted next via knowledge distillation.

Roadmap

• Crown run (done): 16M @ expanded 8.29B corpus, 16B tokens -> BLiMP 70.53 / ARC 40.91, board-recon #14/74. Finding: BLiMP data-lever exhausted at 16M (~70.5 cap); ARC is the sole lever toward #1.
• ARC boost: knowledge distillation from a strong in-house teacher (decontaminated), plus science-dense data.
• Larger raw-score variant under evaluation.

Limitations

Base (not instruction-tuned) research models at 16-32M parameters, English-only. Expect limited factual knowledge and coherence; not intended for production use.

Provenance

Full dialectical record, evaluation artifacts and eval-protocol details in labvault 21_09_GoLLeM-v5-Skalowanie-Glint/ (see 90-Ewaluacja/EvalHarnessParity.md). Trained on RunPod RTX 5090.