Model save

README.md

@@ -1,203 +1,55 @@
 ---
-language: en
-license: apache-2.0
+library_name: transformers
 tags:
-- causal-lm
-- research
-- fp8
-- attention
-- normalization
-- neollm
-datasets:
-- HuggingFaceFW/fineweb-edu
+- generated_from_trainer
+model-index:
+- name: NeoLLM
+  results: []
 ---
 
-# NeoLLM
-
-NeoLLM is a **135 M parameter** decoder-only language model trained from scratch on
-[FineWeb-Edu](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu) in **FP8**
-precision, completing training in approximately **6 hours** on a single NVIDIA RTX 5090.
-It integrates a collection of recently published attention and normalization techniques
-into a single architecture, with the goal of studying how they interact during
-pretraining. The model is actively being developed and the current checkpoint represents
-an intermediate training state.
-
-> **Author / contact:** [@Kyokopom](https://x.com/Kyokopom) on X
-> **Repository:** [KitsuVp/NeoLLM](https://huggingface.co/KitsuVp/NeoLLM)
-
----
-
-## Architecture
-
-NeoLLM is a decoder-only transformer with the following configuration:
-
-| Parameter | Value |
-|---|---|
-| Hidden size | 512 |
-| Layers | 12 |
-| Attention heads | 8 |
-| KV heads (GQA) | 4 |
-| Head dim | 64 |
-| Intermediate size | 1536 |
-| Vocabulary | Qwen3 tokenizer (64,402 tokens) |
-| Context length | 512 tokens |
-
-### Parameter breakdown
-
-| Parameter bucket | Count |
-|---|---|
-| **Total parameters** | 113.07M (113,070,456) |
-| **Embedding parameters** (tied) | 32.97M (32,973,824) |
-| **Non-embedding parameters** | 80.10M (80,096,632) |
-| **Effective trainable parameters** | 113.07M (113,070,456) |
-
-> Weight tying is **enabled**: the input embedding matrix and the language-model head
-> share the same parameters, so the effective trainable budget is
-> `total − embed = 80.10M`.
-
-### Integrated techniques
-
-Each layer combines the following mechanisms simultaneously.
-
-**Normalization and residual stream**
-
-- **SeeDNorm** ([arXiv:2510.22777](https://arxiv.org/abs/2510.22777)) — Applied to Q and K
-  projections. Dynamically rescales the normalization based on the input's own statistics,
-  making the attention geometry more stable across varying input distributions.
-- **PolyNorm** ([arXiv:2602.04902](https://arxiv.org/abs/2602.04902)) — Replaces the standard
-  MLP activation with three branches: linear (x), quadratic (x²), and cubic (x³) — each
-  normalized and combined with learned weights. This allows the MLP to express both linear
-  and non-linear relationships simultaneously.
-- **GPAS** ([arXiv:2506.22049](https://arxiv.org/abs/2506.22049)) — Gradient-Preserving
-  Activation Scaling. Applied to residual connections between sublayers; helps gradients
-  flow more cleanly during training without distorting the residual stream.
-- **LayerNorm Scaling / LNS** ([arXiv:2502.05795](https://arxiv.org/abs/2502.05795)) — Each
-  layer's output is scaled by 1/√ℓ where ℓ is the layer index. Directly addresses the
-  "Curse of Depth" in Pre-LN transformers.
-
-**Attention mechanisms**
-
-- **FAN** ([arXiv:2502.21309](https://arxiv.org/abs/2502.21309)) — Fourier Analysis Networks.
-  A portion of the input projection channels are dedicated to representing periodic patterns
-  (cosine/sine pairs), while the remainder handle standard linear content.
-- **MEA** ([arXiv:2601.19611](https://arxiv.org/abs/2601.19611)) — Explicit Multi-head
-  Attention. Adds small learnable interaction matrices between attention heads for K and V.
-- **LUCID** ([arXiv:2602.10410](https://arxiv.org/abs/2602.10410)) — Applies a learned
-  lower-triangular preconditioner to V before attention, decorrelating value representations
-  across positions.
-- **Affine-Scaled Attention** ([arXiv:2602.23057](https://arxiv.org/abs/2602.23057)) — Adds
-  two learnable per-head scalars (α and β) to the softmax weights:
-  `[α·softmax(QKᵀ) + β]·V`.
-- **XSA** ([arXiv:2603.09078](https://arxiv.org/abs/2603.09078)) — Exclusive Self Attention.
-  After computing attention, removes the component of the output aligned with the token's
-  own value vector.
-- **Directional Routing** ([arXiv:2603.14923](https://arxiv.org/abs/2603.14923)) — Each head
-  learns K=4 directions in the output space; a learned router suppresses the attention output
-  along each direction per input.
-- **Gated Attention** ([arXiv:2505.06708](https://arxiv.org/abs/2505.06708)) — A sigmoid gate
-  is applied to the attention output before the output projection, introducing non-linearity
-  and preventing attention sinks.
-- **Momentum Attention** ([arXiv:2411.03884](https://arxiv.org/abs/2411.03884)) — Modifies Q
-  and K by subtracting a fraction of the previous position's Q and K values (causal
-  first-difference).
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
 
-**MLP**
-
-- **Learnable Multipliers** ([arXiv:2601.04890](https://arxiv.org/abs/2601.04890)) — Adds
-  per-row and per-column learnable scalar parameters to each linear layer.
-- **SimpleGPT** ([arXiv:2602.01212](https://arxiv.org/abs/2602.01212)) — A normalization
-  strategy derived from second-order geometry analysis, applied inside MLP projections to
-  improve optimization stability.
-
----
-
-## Training
-
-| Setting | Value |
-|---|---|
-| Dataset | FineWeb-Edu (sample-10BT) |
-| Tokens seen | ~1.54B (46,875 steps × batch 64 × length 512) |
-| Precision | FP8 native (E4M3 weights/activations, E5M2 gradients) + BF16 fallback |
-| Optimizer | Conda (Column-Normalized Adam) + GPA |
-| Learning rate | 6e-04 with linear warmup (10 % of steps) |
-| Weight decay | 0.1 |
-| Training time | ~3h 31m |
-| Hardware | NVIDIA RTX 5090 (single GPU) |
-
-### Training curve
-
-| Step | Train Loss | Val Loss |
-|---|---|---|
-| 5,000 | 4.005 | 3.905 |
-| 10,000 | 3.727 | 3.637 |
-| 15,000 | 3.616 | 3.534 |
-| 20,000 | 3.559 | 3.475 |
-| 25,000 | 3.521 | 3.437 |
-| 30,000 | 3.494 | 3.410 |
-| 35,000 | 3.479 | 3.390 |
-| 40,000 | 3.395 | 3.321 |
-| 45,000 | 3.355 | 3.273 |
-| 46,875 | — | 3.266 |
-
----
-
-## Limitations
-
-- **Token budget** — ~1.5 B tokens seen; below estimated optimum. Knowledge-intensive tasks
-  will improve with more training.
-- **Gradient spike at step 40k** — Reorganized the attention pattern in layer 9 that
-  previously captured long-range token correlations. A checkpoint from ~step 38k is expected
-  to have better aggregate benchmark scores.
-- **PolyNorm exclusivity** — The quadratic branch has become partially redundant with the
-  linear branch. Will be corrected in the next training run.
-- **Base model only** — Not instruction-tuned or aligned; purely a next-token-prediction
-  base model.
-
----
+# NeoLLM
 
-## References
+This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- eval_loss: 3.3300
+- eval_runtime: 88.5321
+- eval_samples_per_second: 170.277
+- eval_steps_per_second: 2.666
+- epoch: 0.96
+- step: 45000
 
-All papers whose techniques are integrated into NeoLLM's architecture:
+## Model description
 
-| Technique | Paper title | arXiv |
-|---|---|---|
-| SeeDNorm | Self-Rescaled Dynamic Normalization | [2510.22777](https://arxiv.org/abs/2510.22777) |
-| MEA | Explicit Multi-head Attention | [2601.19611](https://arxiv.org/abs/2601.19611) |
-| Learnable Multipliers | Freeing the Scale of Language Model Matrix Layers | [2601.04890](https://arxiv.org/abs/2601.04890) |
-| Directional Routing | Directional Routing in Transformers | [2603.14923](https://arxiv.org/abs/2603.14923) |
-| XSA | Exclusive Self Attention | [2603.09078](https://arxiv.org/abs/2603.09078) |
-| Gated Attention | Gated Attention for LLMs | [2505.06708](https://arxiv.org/abs/2505.06708) |
-| Affine-Scaled Attention | Affine-Scaled Attention | [2602.23057](https://arxiv.org/abs/2602.23057) |
-| LNS | The Curse of Depth in LLMs | [2502.05795](https://arxiv.org/abs/2502.05795) |
-| LUCID | Attention with Preconditioned Representations | [2602.10410](https://arxiv.org/abs/2602.10410) |
-| FAN | Fourier Analysis Networks | [2502.21309](https://arxiv.org/abs/2502.21309) |
-| SimpleGPT | SimpleGPT | [2602.01212](https://arxiv.org/abs/2602.01212) |
-| GPAS | Gradient-Preserving Activation Scaling | [2506.22049](https://arxiv.org/abs/2506.22049) |
-| PolyNorm | PolyNorm / PolyCom | [2602.04902](https://arxiv.org/abs/2602.04902) |
-| Momentum Attention | Momentum Attention | [2411.03884](https://arxiv.org/abs/2411.03884) |
-| TWEO (analysis ref.) | Transformers Without Extreme Outliers | [2511.23225](https://arxiv.org/abs/2511.23225) |
+More information needed
 
----
+## Intended uses & limitations
 
-## Citation
+More information needed
 
-```bibtex
-@misc{neollm2026,
-  title  = {NeoLLM: A Research Language Model Integrating Recent Attention and Normalization Techniques},
-  author = {KitsuVp},
-  year   = {2026},
-  url    = {https://huggingface.co/KitsuVp/NeoLLM}
-}
-```
+## Training and evaluation data
 
----
+More information needed
 
-## Author
+## Training procedure
 
-[@Kyokopom](https://x.com/Kyokopom) on X
+### Training hyperparameters
 
----
+The following hyperparameters were used during training:
+- learning_rate: 0.0006
+- train_batch_size: 64
+- eval_batch_size: 64
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 0.1
+- num_epochs: 1
 
-## License
+### Framework versions
 
-Apache 2.0
+- Transformers 5.8.1
+- Pytorch 2.12.0+cu130
+- Datasets 4.8.5
+- Tokenizers 0.22.2

config.json

@@ -22,7 +22,6 @@
   "fan_ratio_ffn": 0.0625,
   "generator_d_seed": 128,
   "generator_k": 3,
-  "generator_khronos_chunk_size": 16,
   "generator_krank": 64,
   "generator_num_knots": 32,
   "generator_num_modes": 8,
@@ -73,7 +72,7 @@
   "stack_memory_cache_size": 2048,
   "stack_slots": 16,
   "tie_word_embeddings": false,
-  "transformers_version": "5.8.0",
+  "transformers_version": "5.8.1",
   "use_affine_scaled_attention": false,
   "use_attn_res": false,
   "use_cache": false,
@@ -101,7 +100,7 @@
   "use_siamesenorm": true,
   "use_spelling_bee_embeddings": true,
   "use_stack_memory": false,
-  "use_token_generator": false,
+  "use_token_generator": true,
   "use_versatile_ffn": false,
   "use_xsa": true,
   "versatile_active_experts": 2,

generation_config.json

@@ -7,5 +7,5 @@
   "output_attentions": false,
   "output_hidden_states": false,
   "pad_token_id": 0,
-  "transformers_version": "5.8.0"
+  "transformers_version": "5.8.1"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:adf4cf118e8db2b8f126d7d1ecf9c6795ab7df172095edae7788e31d3c46cae2
-size 234563356
+oid sha256:8dd96ef8d8c02d6ff194e1f5d4fa8c27936abd94c0f60240f29f16f6f9b571a5
+size 173668524

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f3cb95d622aaf5240b2ab92566316ecd151c8b597d3d86bcfc154933227d34f5
+oid sha256:b1c3c25d3f289220adf32328d1a78c81a4fe69b32cfac36940998fdfcda2254a
 size 5457