Upload folder using huggingface_hub

.gitattributes

@@ -42,3 +42,4 @@ checkpoints-v5.5/checkpoint-10240/eval_state.json filter=lfs diff=lfs merge=lfs
 checkpoints-v5.5/checkpoint-16384/eval_state.json filter=lfs diff=lfs merge=lfs -text
 checkpoints-v4.4+/checkpoint-7168/eval_state.json filter=lfs diff=lfs merge=lfs -text
 checkpoints-v5.5/checkpoint-24576/eval_state.json filter=lfs diff=lfs merge=lfs -text
+checkpoints-v5.6/checkpoint-4096/eval_state.json filter=lfs diff=lfs merge=lfs -text

checkpoints-v5.6/checkpoint-4096/ema.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b347ddb30e8f07210cd2e13f22f02b92b453ff0eeb5da7ed4fdd6af8b12c3906
+size 54599592

checkpoints-v5.6/checkpoint-4096/eval_state.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a36ee7482846bfc1ebc36030b2404f9402dc9b03d82fa9dad17a03390fa9ad99
+size 57504195

checkpoints-v5.6/checkpoint-4096/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a6dc5b97d66fc7a0a7c5f7b000f731214324c8c4ec4f0d2f740153a8d9695d8
+size 54599624

checkpoints-v5.6/checkpoint-4096/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10cff83ebeecce21ea96ec8c7d555a559e6e8651dbb5310ba8524472b597fdd8
+size 76550347

checkpoints-v5.6/checkpoint-4096/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0401bbc32a985fb9214c37475cefac4af17c87b54c40831a6e33fbd65fbdc86c
+size 14645

checkpoints-v5.6/checkpoint-4096/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90078bef0ff08e77712ec862bc2a11b4989d3477480b20822129904ef078a3a3
+size 1383

checkpoints-v5.6/checkpoint-4096/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbc93ab79de1c4761b64667cf33632ad206830e56d286204806262fb069585fe
+size 1465

checkpoints-v5.6/checkpoint-4096/trainer_state.json

@@ -0,0 +1,186 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.176525093197147,
+  "eval_steps": 1024,
+  "global_step": 4096,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04413127329928675,
+      "grad_norm": 2.024094581604004,
+      "learning_rate": 9.990234375e-05,
+      "loss": 13.011528968811035,
+      "step": 1024
+    },
+    {
+      "epoch": 0.04413127329928675,
+      "eval_bleu": 0.13973491511585304,
+      "eval_ce_clean_loss": 2.321801658886582,
+      "eval_ce_pred_loss": 5.398845981687371,
+      "eval_flow_cos_loss": 0.4359625861970092,
+      "eval_flow_mse_loss": 1.261753735003441,
+      "eval_loss": 7.471738189014037,
+      "flow/cos_sim": 0.5640374244784495,
+      "flow/improvement_ratio": 0.9125845870737836,
+      "flow/mag_ratio_mean": 0.005239322681281803,
+      "flow/mag_ratio_std": 0.0019533936416349018,
+      "step": 1024
+    },
+    {
+      "epoch": 0.04413127329928675,
+      "eval_bleu": 0.13973491511585304,
+      "eval_ce_clean_loss": 2.321801658886582,
+      "eval_ce_pred_loss": 5.398845981687371,
+      "eval_flow_cos_loss": 0.4359625861970092,
+      "eval_flow_mse_loss": 1.261753735003441,
+      "eval_loss": 7.471738189014037,
+      "eval_runtime": 200.3486,
+      "eval_samples_per_second": 149.739,
+      "eval_steps_per_second": 2.341,
+      "flow/cos_sim": 0.5640374244784495,
+      "flow/improvement_ratio": 0.9125845870737836,
+      "flow/mag_ratio_mean": 0.005239322681281803,
+      "flow/mag_ratio_std": 0.0019533936416349018,
+      "step": 1024
+    },
+    {
+      "epoch": 0.0882625465985735,
+      "grad_norm": 2.3836026191711426,
+      "learning_rate": 9.9476028157316e-05,
+      "loss": 5.076037406921387,
+      "step": 2048
+    },
+    {
+      "epoch": 0.0882625465985735,
+      "eval_bleu": 0.334789938692836,
+      "eval_ce_clean_loss": 0.34960190961355847,
+      "eval_ce_pred_loss": 3.634908381301457,
+      "eval_flow_cos_loss": 0.4467242890075326,
+      "eval_flow_mse_loss": 1.0751057558222366,
+      "eval_loss": 4.080824585611632,
+      "flow/cos_sim": 0.5532757186177951,
+      "flow/improvement_ratio": 0.988054730363492,
+      "flow/mag_ratio_mean": 0.4422806202094438,
+      "flow/mag_ratio_std": 0.11935382749416681,
+      "step": 2048
+    },
+    {
+      "epoch": 0.0882625465985735,
+      "eval_bleu": 0.334789938692836,
+      "eval_ce_clean_loss": 0.34960190961355847,
+      "eval_ce_pred_loss": 3.634908381301457,
+      "eval_flow_cos_loss": 0.4467242890075326,
+      "eval_flow_mse_loss": 1.0751057558222366,
+      "eval_loss": 4.080824585611632,
+      "eval_runtime": 198.1742,
+      "eval_samples_per_second": 151.382,
+      "eval_steps_per_second": 2.367,
+      "flow/cos_sim": 0.5532757186177951,
+      "flow/improvement_ratio": 0.988054730363492,
+      "flow/mag_ratio_mean": 0.4422806202094438,
+      "flow/mag_ratio_std": 0.11935382749416681,
+      "step": 2048
+    },
+    {
+      "epoch": 0.13239381989786023,
+      "grad_norm": 0.8187930583953857,
+      "learning_rate": 9.791307026072513e-05,
+      "loss": 3.823514938354492,
+      "step": 3072
+    },
+    {
+      "epoch": 0.13239381989786023,
+      "eval_bleu": 0.372073011692056,
+      "eval_ce_clean_loss": 0.13431392060414052,
+      "eval_ce_pred_loss": 3.167514971832731,
+      "eval_flow_cos_loss": 0.3423818138870857,
+      "eval_flow_mse_loss": 1.1092233459578393,
+      "eval_loss": 3.5463931677438048,
+      "flow/cos_sim": 0.6576181853503815,
+      "flow/improvement_ratio": 0.9944826788993787,
+      "flow/mag_ratio_mean": 0.6025184103166625,
+      "flow/mag_ratio_std": 0.11320645424094536,
+      "step": 3072
+    },
+    {
+      "epoch": 0.13239381989786023,
+      "eval_bleu": 0.372073011692056,
+      "eval_ce_clean_loss": 0.13431392060414052,
+      "eval_ce_pred_loss": 3.167514971832731,
+      "eval_flow_cos_loss": 0.3423818138870857,
+      "eval_flow_mse_loss": 1.1092233459578393,
+      "eval_loss": 3.5463931677438048,
+      "eval_runtime": 199.1274,
+      "eval_samples_per_second": 150.657,
+      "eval_steps_per_second": 2.355,
+      "flow/cos_sim": 0.6576181853503815,
+      "flow/improvement_ratio": 0.9944826788993787,
+      "flow/mag_ratio_mean": 0.6025184103166625,
+      "flow/mag_ratio_std": 0.11320645424094536,
+      "step": 3072
+    },
+    {
+      "epoch": 0.176525093197147,
+      "grad_norm": 1.596978783607483,
+      "learning_rate": 9.53439476074686e-05,
+      "loss": 3.4521546363830566,
+      "step": 4096
+    },
+    {
+      "epoch": 0.176525093197147,
+      "eval_bleu": 0.3922768378761595,
+      "eval_ce_clean_loss": 0.06992586965023327,
+      "eval_ce_pred_loss": 2.9707689447951977,
+      "eval_flow_cos_loss": 0.18835138766241988,
+      "eval_flow_mse_loss": 1.08429173365839,
+      "eval_loss": 3.2808436879725345,
+      "flow/cos_sim": 0.8116486336884976,
+      "flow/improvement_ratio": 0.9945465602091889,
+      "flow/mag_ratio_mean": 0.7910831776509153,
+      "flow/mag_ratio_std": 0.09716444489544135,
+      "step": 4096
+    },
+    {
+      "epoch": 0.176525093197147,
+      "eval_bleu": 0.3922768378761595,
+      "eval_ce_clean_loss": 0.06992586965023327,
+      "eval_ce_pred_loss": 2.9707689447951977,
+      "eval_flow_cos_loss": 0.18835138766241988,
+      "eval_flow_mse_loss": 1.08429173365839,
+      "eval_loss": 3.2808436879725345,
+      "eval_runtime": 199.2079,
+      "eval_samples_per_second": 150.596,
+      "eval_steps_per_second": 2.354,
+      "flow/cos_sim": 0.8116486336884976,
+      "flow/improvement_ratio": 0.9945465602091889,
+      "flow/mag_ratio_mean": 0.7910831776509153,
+      "flow/mag_ratio_std": 0.09716444489544135,
+      "step": 4096
+    }
+  ],
+  "logging_steps": 1024,
+  "max_steps": 23204,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1024,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints-v5.6/checkpoint-4096/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8469bbc81a2ba0be2f5b44007faafd15c75615abe30f4f4e56171816d31caa5b
+size 5137