lapp0 commited on
Commit
a18b7a0
·
verified ·
1 Parent(s): 5186c08

End of training

Browse files
README.md CHANGED
@@ -125,7 +125,7 @@ Trained on 521,413,804 tokens from the [wikimedia/wikipedia](https://huggingface
125
  # Training Objective
126
 
127
  ```
128
- 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_teacher_only_affine, projector=mlp_64_l4))
129
  ```
130
 
131
  # Hyperparameters
@@ -141,8 +141,8 @@ The following hyperparameters were used during training:
141
  - optimizer: `Adam with betas=(0.9,0.999) and epsilon=1e-08`
142
  - lr_scheduler_type: `polynomial`
143
  - num_epochs: `1.0`
144
- - 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_teacher_only_affine, projector=mlp_64_l4))`
145
- - lr_scheduler: `<torch.optim.lr_scheduler.LambdaLR object at 0x7f5b3ddf36d0>`
146
  - student_model_name_or_path: `None`
147
  - student_config_name_or_path: `distilbert/distilgpt2`
148
  - student_model_config: `None`
 
125
  # Training Objective
126
 
127
  ```
128
+ 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_teacher_only_affine, projector=orthogonal))
129
  ```
130
 
131
  # Hyperparameters
 
141
  - optimizer: `Adam with betas=(0.9,0.999) and epsilon=1e-08`
142
  - lr_scheduler_type: `polynomial`
143
  - num_epochs: `1.0`
144
+ - 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_teacher_only_affine, projector=orthogonal))`
145
+ - lr_scheduler: `<torch.optim.lr_scheduler.LambdaLR object at 0x7f5b3de02770>`
146
  - student_model_name_or_path: `None`
147
  - student_config_name_or_path: `distilbert/distilgpt2`
148
  - student_model_config: `None`
logs/attn_norm=layernorm_teacher_only_affine, attn_projector=orthogonal, attn_weight=5, learning_rate=0.0002, per_device_train_batch_size=16, warmup_ratio=0/events.out.tfevents.1725524224.a7e428977e35 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef342266758b756a8098d2c910499a75301ffd3168f9bb686a3fe6cba8ee65d6
3
+ size 529