Add files using upload-large-folder tool

backup_manifest.json

@@ -0,0 +1,72 @@
+[
+  {
+    "remote_path": "contextual/pct25/pretraining/pixel/alignment/final/pytorch_model.bin",
+    "source": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/pretraining/pixel/alignment/final/pytorch_model.bin",
+    "bytes": 1942239884
+  },
+  {
+    "remote_path": "contextual/pct25/pretraining/pixel/alignment/final/training_complete.json",
+    "source": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/pretraining/pixel/alignment/final/training_complete.json",
+    "bytes": 463
+  },
+  {
+    "remote_path": "contextual/pct25/pretraining/pixel/full_epoch1/final/pytorch_model.bin",
+    "source": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/pretraining/pixel/full_epoch1/final/pytorch_model.bin",
+    "bytes": 1942239884
+  },
+  {
+    "remote_path": "contextual/pct25/pretraining/pixel/full_epoch1/final/training_complete.json",
+    "source": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/pretraining/pixel/full_epoch1/final/training_complete.json",
+    "bytes": 577
+  },
+  {
+    "remote_path": "contextual/pct25/pretraining/pixel/full_epoch2/final/pytorch_model.bin",
+    "source": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/pretraining/pixel/full_epoch2/final/pytorch_model.bin",
+    "bytes": 1942239884
+  },
+  {
+    "remote_path": "contextual/pct25/pretraining/pixel/full_epoch2/final/training_complete.json",
+    "source": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/pretraining/pixel/full_epoch2/final/training_complete.json",
+    "bytes": 579
+  },
+  {
+    "remote_path": "contextual/pct25/finetuning/pixel/MSA_Dialect_Combined_1x/best_model.bin",
+    "source": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/finetuning/pixel/MSA_Dialect_Combined_1x/best_model.bin",
+    "bytes": 1942238590
+  },
+  {
+    "remote_path": "contextual/pct25/finetuning/pixel/MSA_Dialect_Combined_1x/training_complete.json",
+    "source": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/finetuning/pixel/MSA_Dialect_Combined_1x/training_complete.json",
+    "bytes": 631
+  },
+  {
+    "remote_path": "msa_curriculum/reset/pct50/pretraining/decoder/full_epoch1/final/pytorch_model.bin",
+    "source": "/root/baseline_cloud/work/checkpoints/msa_curriculum/reset/pct50/pretraining/decoder/full_epoch1/final/pytorch_model.bin",
+    "bytes": 1447437206
+  },
+  {
+    "remote_path": "msa_curriculum/reset/pct50/pretraining/decoder/full_epoch1/final/training_complete.json",
+    "source": "/root/baseline_cloud/work/checkpoints/msa_curriculum/reset/pct50/pretraining/decoder/full_epoch1/final/training_complete.json",
+    "bytes": 463
+  },
+  {
+    "remote_path": "msa_curriculum/reset/pct50/pretraining/decoder/full_epoch2/final/pytorch_model.bin",
+    "source": "/root/baseline_cloud/work/checkpoints/msa_curriculum/reset/pct50/pretraining/decoder/full_epoch2/final/pytorch_model.bin",
+    "bytes": 1447437206
+  },
+  {
+    "remote_path": "msa_curriculum/reset/pct50/pretraining/decoder/full_epoch2/final/training_complete.json",
+    "source": "/root/baseline_cloud/work/checkpoints/msa_curriculum/reset/pct50/pretraining/decoder/full_epoch2/final/training_complete.json",
+    "bytes": 581
+  },
+  {
+    "remote_path": "msa_curriculum/reset/pct50/finetuning/decoder/MSA_Dialect_Combined_1x/best_model.bin",
+    "source": "/root/baseline_cloud/work/checkpoints/msa_curriculum/reset/pct50/finetuning/decoder/MSA_Dialect_Combined_1x/best_model.bin",
+    "bytes": 1447436146
+  },
+  {
+    "remote_path": "msa_curriculum/reset/pct50/finetuning/decoder/MSA_Dialect_Combined_1x/training_complete.json",
+    "source": "/root/baseline_cloud/work/checkpoints/msa_curriculum/reset/pct50/finetuning/decoder/MSA_Dialect_Combined_1x/training_complete.json",
+    "bytes": 622
+  }
+]

contextual/pct25/finetuning/pixel/MSA_Dialect_Combined_1x/best_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d9bae9e8d58feb6497718567ee21c02b76270cf8c8c1343f4855fe07a0f419a
+size 1942238590

contextual/pct25/finetuning/pixel/MSA_Dialect_Combined_1x/training_complete.json

@@ -0,0 +1,19 @@
+{
+  "dataset_path": "/root/baseline_cloud/contextual_25_work/data/contextual/downstream/MSA_Dialect_Combined_1x_max64",
+  "dataset_rows": 584406,
+  "epochs": 1,
+  "global_step": 4566,
+  "input_checkpoint": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/pretraining/pixel/full_epoch2/final/pytorch_model.bin",
+  "learning_rate": 3e-05,
+  "metrics": {
+    "epoch": 1.0,
+    "train_loss": 1.3086010231764964,
+    "train_runtime": 1181.8981,
+    "train_samples_per_second": 494.464,
+    "train_steps_per_second": 3.863
+  },
+  "model_type": "pixel",
+  "scheduler": "cosine",
+  "seed": 42,
+  "warmup_ratio": 0.1
+}

contextual/pct25/pretraining/pixel/alignment/final/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7f7701cee72ab1ba580a61ce46a2e31c3270521aaf3d0af89c46f3d8c9a601a
+size 1942239884

contextual/pct25/pretraining/pixel/alignment/final/training_complete.json

@@ -0,0 +1,21 @@
+{
+  "epochs": 1,
+  "global_step": 11739,
+  "initial_weights": null,
+  "learning_rate": 0.0003,
+  "metrics": {
+    "epoch": 1.0,
+    "train_loss": 3.069912617916813,
+    "train_runtime": 6426.3644,
+    "train_samples_per_second": 935.289,
+    "train_steps_per_second": 1.827
+  },
+  "model_type": "pixel",
+  "percentage": 25,
+  "scheduler": "cosine",
+  "seed": 42,
+  "stage": "alignment",
+  "subset_rows": 6010510,
+  "total_rows": 24042041,
+  "warmup_ratio": 0.1
+}

contextual/pct25/pretraining/pixel/full_epoch1/final/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7639c7d2ec286772b2f305c21876942b21f5e6f73acc745ed8357f445fdb404
+size 1942239884

contextual/pct25/pretraining/pixel/full_epoch1/final/training_complete.json

@@ -0,0 +1,21 @@
+{
+  "epochs": 1,
+  "global_step": 11739,
+  "initial_weights": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/pretraining/pixel/alignment/final/pytorch_model.bin",
+  "learning_rate": 0.0003,
+  "metrics": {
+    "epoch": 1.0,
+    "train_loss": 1.1720564518320935,
+    "train_runtime": 7655.0557,
+    "train_samples_per_second": 785.169,
+    "train_steps_per_second": 1.533
+  },
+  "model_type": "pixel",
+  "percentage": 25,
+  "scheduler": "cosine",
+  "seed": 42,
+  "stage": "full",
+  "subset_rows": 6010510,
+  "total_rows": 24042041,
+  "warmup_ratio": 0.1
+}

contextual/pct25/pretraining/pixel/full_epoch2/final/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84152f3b4fd138a09d2ca67ebcac7176f9986921899ebcb8fac8f2ee53165f1e
+size 1942239884

contextual/pct25/pretraining/pixel/full_epoch2/final/training_complete.json

@@ -0,0 +1,21 @@
+{
+  "epochs": 1,
+  "global_step": 11739,
+  "initial_weights": "/root/baseline_cloud/contextual_25_work/checkpoints/contextual/pct25/pretraining/pixel/full_epoch1/final/pytorch_model.bin",
+  "learning_rate": 0.0003,
+  "metrics": {
+    "epoch": 1.0,
+    "train_loss": 0.9379562959119402,
+    "train_runtime": 7651.2785,
+    "train_samples_per_second": 785.556,
+    "train_steps_per_second": 1.534
+  },
+  "model_type": "pixel",
+  "percentage": 25,
+  "scheduler": "cosine",
+  "seed": 42,
+  "stage": "full",
+  "subset_rows": 6010510,
+  "total_rows": 24042041,
+  "warmup_ratio": 0.1
+}

msa_curriculum/reset/pct50/finetuning/decoder/MSA_Dialect_Combined_1x/best_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0aeea707d355d1f1f46f42b13ef423762332ee0ceaee7f8a44159a5cd7abf0ed
+size 1447436146

msa_curriculum/reset/pct50/finetuning/decoder/MSA_Dialect_Combined_1x/training_complete.json

@@ -0,0 +1,19 @@
+{
+  "dataset_path": "/root/baseline_cloud/work/prepared/msa_curriculum/train/decoder/MSA_Dialect_Combined_1x",
+  "dataset_rows": 584406,
+  "epochs": 1,
+  "global_step": 4565,
+  "input_checkpoint": "/root/baseline_cloud/work/checkpoints/msa_curriculum/reset/pct50/pretraining/decoder/full_epoch2/final/pytorch_model.bin",
+  "learning_rate": 3e-05,
+  "metrics": {
+    "epoch": 1.0,
+    "train_loss": 1.2391240839629158,
+    "train_runtime": 1861.9311,
+    "train_samples_per_second": 313.871,
+    "train_steps_per_second": 2.452
+  },
+  "model_type": "decoder",
+  "scheduler": "cosine",
+  "seed": 42,
+  "warmup_ratio": 0.1
+}

msa_curriculum/reset/pct50/pretraining/decoder/full_epoch1/final/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:634c806467097fab94653d9277d67d7d2833f0219546cb606d2e8796ca3fbc23
+size 1447437206

msa_curriculum/reset/pct50/pretraining/decoder/full_epoch1/final/training_complete.json

@@ -0,0 +1,21 @@
+{
+  "epochs": 1,
+  "global_step": 23478,
+  "initial_weights": null,
+  "learning_rate": 0.0003,
+  "metrics": {
+    "epoch": 1.0,
+    "train_loss": 0.6118796945928218,
+    "train_runtime": 26768.1406,
+    "train_samples_per_second": 449.079,
+    "train_steps_per_second": 0.877
+  },
+  "model_type": "decoder",
+  "percentage": 50,
+  "scheduler": "cosine",
+  "seed": 42,
+  "stage": "full",
+  "subset_rows": 12021020,
+  "total_rows": 24042041,
+  "warmup_ratio": 0.1
+}

msa_curriculum/reset/pct50/pretraining/decoder/full_epoch2/final/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef77f63b874315f0058e88c99eef81d5f3a7ead7975805c83424a8e1cebf113d
+size 1447437206

msa_curriculum/reset/pct50/pretraining/decoder/full_epoch2/final/training_complete.json

@@ -0,0 +1,21 @@
+{
+  "epochs": 1,
+  "global_step": 23478,
+  "initial_weights": "/root/baseline_cloud/work/checkpoints/msa_curriculum/reset/pct50/pretraining/decoder/full_epoch1/final/pytorch_model.bin",
+  "learning_rate": 0.0003,
+  "metrics": {
+    "epoch": 1.0,
+    "train_loss": 0.8671167392383007,
+    "train_runtime": 40630.0806,
+    "train_samples_per_second": 295.865,
+    "train_steps_per_second": 0.578
+  },
+  "model_type": "decoder",
+  "percentage": 50,
+  "scheduler": "cosine",
+  "seed": 42,
+  "stage": "full",
+  "subset_rows": 12021020,
+  "total_rows": 24042041,
+  "warmup_ratio": 0.1
+}