Training in progress, step 61875

README.md

@@ -46,7 +46,7 @@ More information needed
 
 # Resource Usage Comparison
 
-- VRAM Use: 7.7872 GB
+- VRAM Use: 7.7871 GB
 
 # Distillation (Teacher -> Student) Architecture Difference:
 
@@ -76,7 +76,7 @@ Trained on 145,744,973 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=25.0, loss_fn=cos, layer_mapper=layer-2, projector=orthogonal_batchnorm))
+DistillationObjective(logits_loss_component=LossComponent(label=logits, weight=1, loss_fn=kl), attn_loss_component=LossComponent(label=attn, weight=25.0, loss_fn=raw_mse, layer_mapper=layer-2, norm=layernorm, projector=orthogonal))
 ```
 
 # Hyperparameters
@@ -93,9 +93,9 @@ The following hyperparameters were used during training:
 - lr_scheduler_type: `cosine_with_min_lr`
 - lr_scheduler_warmup_ratio: `0.5`
 - 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=25.0, loss_fn=cos, layer_mapper=layer-2, projector=orthogonal_batchnorm))`
+- distillation_objective: `DistillationObjective(logits_loss_component=LossComponent(label=logits, weight=1, loss_fn=kl), attn_loss_component=LossComponent(label=attn, weight=25.0, loss_fn=raw_mse, layer_mapper=layer-2, norm=layernorm, projector=orthogonal))`
 - train_embeddings: `True`
-- lr_scheduler: `<torch.optim.lr_scheduler.LambdaLR object at 0x74d6ca592a70>`
+- lr_scheduler: `<torch.optim.lr_scheduler.LambdaLR object at 0x7fe929482740>`
 - student_model_name_or_path: `None`
 - student_config_name_or_path: `None`
 - student_model_config: `None`
@@ -114,7 +114,7 @@ The following hyperparameters were used during training:
 - dataset_test_size: `0.01`
 - gradient_accumulation_steps: `1`
 - weight_decay: `0.0`
-- max_grad_norm: `100`
+- max_grad_norm: `1.0`
 - warmup_ratio: `0.5`
 - warmup_steps: `0`
 - gradient_checkpointing: `True`
@@ -124,7 +124,7 @@ The following hyperparameters were used during training:
 
 
 # Framework Versions
-- Distily 0.4.0
-- Transformers 4.44.2
-- Pytorch 2.3.0
+- Distily 0.4.1
+- Transformers 4.44.1
+- Pytorch 2.4.0+cu121
 - Datasets 2.21.0

config.json

@@ -33,7 +33,7 @@
     }
   },
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.44.2",
+  "transformers_version": "4.44.1",
   "use_cache": true,
   "vocab_size": 50257
 }

generation_config.json

@@ -2,5 +2,5 @@
   "_from_model_config": true,
   "bos_token_id": 50256,
   "eos_token_id": 50256,
-  "transformers_version": "4.44.2"
+  "transformers_version": "4.44.1"
 }

logs/attn_layer_mapper=layer-2, attn_loss_fn=cos, attn_norm=batchnorm, attn_projector=orthogonal, max_grad_norm=100/events.out.tfevents.1724957323.e3f806ea38c9

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d7a8c31cb86e95c80a9a549d9cf199135ea12ec2f4558160cc096201a75187f
+size 29625403

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5077f4511637f3ef362d7333bd324d0d3037cdcd0d1373884507acf75954834e
+oid sha256:36ded38759f4cae9e63259ab5ce175424a678eb861a18101c95ad8c582587f79
 size 248894656

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bc86e41c929d40fef9cfcf7ed0f9854b8455aa1cc23553c59926d2cbab39ae97
+oid sha256:09d1552da14c086e3adb6730e8cc1b2851ce2605b358b889ddc09ff6b6313361
 size 5560