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=
|
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=
|
145 |
-
- lr_scheduler: `<torch.optim.lr_scheduler.LambdaLR object at
|
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
|