End of training
Browse files
README.md
CHANGED
@@ -85,7 +85,7 @@ Trained on 521,413,804 tokens from the [wikimedia/wikipedia](https://huggingface
|
|
85 |
# Training Objective
|
86 |
|
87 |
```
|
88 |
-
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=
|
89 |
```
|
90 |
|
91 |
# Hyperparameters
|
@@ -95,15 +95,15 @@ The following hyperparameters were used during training:
|
|
95 |
<summary>Expand</summary>
|
96 |
|
97 |
- learning_rate: `0.0002`
|
98 |
-
- train_batch_size: `
|
99 |
- eval_batch_size: `8`
|
100 |
- seed: `42`
|
101 |
- optimizer: `Adam with betas=(0.9,0.999) and epsilon=1e-08`
|
102 |
- lr_scheduler_type: `polynomial`
|
103 |
- num_epochs: `1.0`
|
104 |
-
- 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=
|
105 |
- train_embeddings: `True`
|
106 |
-
- lr_scheduler: `<torch.optim.lr_scheduler.LambdaLR object at
|
107 |
- student_model_name_or_path: `None`
|
108 |
- student_config_name_or_path: `distilbert/distilgpt2`
|
109 |
- student_model_config: `None`
|
|
|
85 |
# Training Objective
|
86 |
|
87 |
```
|
88 |
+
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, projector=orthogonal))
|
89 |
```
|
90 |
|
91 |
# Hyperparameters
|
|
|
95 |
<summary>Expand</summary>
|
96 |
|
97 |
- learning_rate: `0.0002`
|
98 |
+
- train_batch_size: `8`
|
99 |
- eval_batch_size: `8`
|
100 |
- seed: `42`
|
101 |
- optimizer: `Adam with betas=(0.9,0.999) and epsilon=1e-08`
|
102 |
- lr_scheduler_type: `polynomial`
|
103 |
- num_epochs: `1.0`
|
104 |
+
- 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, projector=orthogonal))`
|
105 |
- train_embeddings: `True`
|
106 |
+
- lr_scheduler: `<torch.optim.lr_scheduler.LambdaLR object at 0x7f27381a2740>`
|
107 |
- student_model_name_or_path: `None`
|
108 |
- student_config_name_or_path: `distilbert/distilgpt2`
|
109 |
- student_model_config: `None`
|
logs/attn_norm=layernorm_teacher_only, attn_projector=orthogonal, attn_weight=5, learning_rate=0.0002, per_device_train_batch_size=8, warmup_ratio=0/events.out.tfevents.1725280443.e3f806ea38c9
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b62253428b590f286e61308de8ff6ea42035aa5ef3ac9b373341d9eceacfd611
|
3 |
+
size 529
|