Model save

README.md

@@ -0,0 +1,63 @@
+---
+library_name: transformers
+tags:
+- generated_from_trainer
+model-index:
+- name: lltransformer-fwe
+  results: []
+---
+
+<!-- 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. -->
+
+# lltransformer-fwe
+
+This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 11.9038
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 1234
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 32
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.03
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 95.3922       | 0.1860 | 3    | 11.9298         |
+| 95.3294       | 0.3721 | 6    | 11.9184         |
+| 95.1952       | 0.5581 | 9    | 11.9093         |
+| 95.1862       | 0.7442 | 12   | 11.9053         |
+| 95.0949       | 0.9302 | 15   | 11.9038         |
+
+
+### Framework versions
+
+- Transformers 4.51.3
+- Pytorch 2.7.0+cu128
+- Datasets 3.5.0
+- Tokenizers 0.21.1

config.json

@@ -2,11 +2,11 @@
   "architectures": [
     "LLTransformerForCausalLM"
   ],
-  "depth": 3,
-  "dim": 512,
+  "depth": 2,
+  "dim": 64,
   "hkv_processor_factory": "linear",
   "model_type": "LLTransformer",
-  "num_heads": 8,
+  "num_heads": 4,
   "torch_dtype": "float32",
   "transformers_version": "4.51.3",
   "vocab_size": 128256

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd3a2037a19081b5946774632dcd93f70c21af720bc5bbcf3e73d087bb2536f2
-size 581984520
+oid sha256:1e823fbba914aceb9b7e753be207e20c8b057db43cc22d71036ba2f8d960c136
+size 66261352

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1cc1fc6c47a3e52a5260457fc6cca0a6957547b851cac5c63eff861a577a196
+oid sha256:74725679d5ac1cb1920f53e39ae856a3ad7dce2a0fc256b1441e9bbadc487ec4
 size 5713