Training in progress, step 99000

README.md

@@ -44,7 +44,7 @@ More information needed
 
 # Resource Usage Comparison
 
-- VRAM Use: 7.4152 GB
+- VRAM Use: 7.4145 GB
 
 # Distillation (Teacher -> Student) Architecture Difference:
 
@@ -75,7 +75,7 @@ More information needed
 <br/>
 
 # Train Dataset
-Trained on 226,118,907 tokens from the [wikimedia/wikipedia](https://huggingface.co/datasets/wikimedia/wikipedia) dataset.
+Trained on 226,137,833 tokens from the [wikimedia/wikipedia](https://huggingface.co/datasets/wikimedia/wikipedia) dataset.
 
 - Num Samples: `396,000`
 - Subset: `20231101.en`
@@ -85,7 +85,7 @@ Trained on 226,118,907 tokens from the [wikimedia/wikipedia](https://huggingface
 # Training Objective
 
 ```
-DistillationObjective(logits_loss_component=LossComponent(label=logits, weight=1, loss_fn=kl), attn_loss_component=LossComponent(label=attn, weight=5, loss_fn=raw_mse, layer_mapper=layer-2, projector=orthogonal))
+DistillationObjective(logits_loss_component=LossComponent(label=logits, weight=1, loss_fn=kl), attn_loss_component=LossComponent(label=attn, weight=5, loss_fn=raw_mse, layer_mapper=layer-2, norm=layernorm, projector=mlp))
 ```
 
 # Hyperparameters
@@ -94,7 +94,7 @@ The following hyperparameters were used during training:
 <details>
 <summary>Expand</summary>
 
-- learning_rate: `0.0002`
+- learning_rate: `0.0001`
 - train_batch_size: `4`
 - eval_batch_size: `8`
 - seed: `42`
@@ -102,9 +102,9 @@ The following hyperparameters were used during training:
 - lr_scheduler_type: `polynomial`
 - lr_scheduler_warmup_ratio: `0.2`
 - num_epochs: `1.0`
-- distillation_objective: `DistillationObjective(logits_loss_component=LossComponent(label=logits, weight=1, loss_fn=kl), attn_loss_component=LossComponent(label=attn, weight=5, loss_fn=raw_mse, layer_mapper=layer-2, projector=orthogonal))`
+- distillation_objective: `DistillationObjective(logits_loss_component=LossComponent(label=logits, weight=1, loss_fn=kl), attn_loss_component=LossComponent(label=attn, weight=5, loss_fn=raw_mse, layer_mapper=layer-2, norm=layernorm, projector=mlp))`
 - train_embeddings: `True`
-- lr_scheduler: `<torch.optim.lr_scheduler.LambdaLR object at 0x7fe828138b50>`
+- lr_scheduler: `<torch.optim.lr_scheduler.LambdaLR object at 0x7fd06c06f5e0>`
 - student_model_name_or_path: `None`
 - student_config_name_or_path: `distilbert/distilgpt2`
 - student_model_config: `None`
@@ -136,4 +136,4 @@ The following hyperparameters were used during training:
 - Distily 0.4.1
 - Transformers 4.44.2
 - Pytorch 2.4.0+cu121
-- Datasets 2.21.0
+- Datasets 2.18.0

logs/attn_norm=layernorm, attn_projector=mlp, attn_weight=5, learning_rate=0.0002, per_device_train_batch_size=4, warmup_ratio=0.2/events.out.tfevents.1725095515.a7e428977e35

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e3ba72e867f1fbb7eac41369be0ead47d5bd76ccef34972669fd69f11399677
+size 47486046

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:860cfa913858a39655a28022a2681ccaad8886d1c5fd4b8e71de27d335dd6f60
+oid sha256:1f0bb459722ad4d52b5531c90b000d6f7650ac72e0ba7b7b416f77faf2404685
 size 163832792

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19290d884b3bb6e9c4a45223383220cf22649f63f02e2c9a3e5b7408b96e277d
+oid sha256:a836e6c45840cff0a4e772e9583729969b30f2fb83b7ffc09c8f641f85f3eb2f
 size 5560