feat: add missing files

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f57c04341fd5297c78ad083da03808ea8935e2b27bf84ee6cb21ef68b02adfcb
+size 4481473896

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7620a3db81553dd539823e6897c170f6edbe0cf10134328481958b6a2df6c393
+size 14180

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65419197ae68c73458d70bdece839226d629ae7646f303ac3a0298bea3c1f850
+size 1064

trainer_state.json

@@ -0,0 +1,153 @@
+{
+  "best_metric": 0.8743219971656799,
+  "best_model_checkpoint": "/disco/guillem.garcia/bs32lr2e5/train_save/checkpoint-2800",
+  "epoch": 1.6928657799274487,
+  "eval_steps": 350,
+  "global_step": 2800,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.21160822249093109,
+      "grad_norm": 11.506365776062012,
+      "learning_rate": 9.936177359758147e-06,
+      "loss": 1.5267,
+      "step": 350
+    },
+    {
+      "epoch": 0.21160822249093109,
+      "eval_loss": 1.0800126791000366,
+      "eval_runtime": 230.2683,
+      "eval_samples_per_second": 24.775,
+      "eval_steps_per_second": 1.55,
+      "step": 350
+    },
+    {
+      "epoch": 0.42321644498186217,
+      "grad_norm": 4.511447429656982,
+      "learning_rate": 8.760497144776622e-06,
+      "loss": 1.107,
+      "step": 700
+    },
+    {
+      "epoch": 0.42321644498186217,
+      "eval_loss": 0.9888215661048889,
+      "eval_runtime": 230.2255,
+      "eval_samples_per_second": 24.78,
+      "eval_steps_per_second": 1.551,
+      "step": 700
+    },
+    {
+      "epoch": 0.6348246674727932,
+      "grad_norm": 4.709233283996582,
+      "learning_rate": 7.584816929795097e-06,
+      "loss": 1.042,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6348246674727932,
+      "eval_loss": 0.9507893919944763,
+      "eval_runtime": 229.7622,
+      "eval_samples_per_second": 24.83,
+      "eval_steps_per_second": 1.554,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8464328899637243,
+      "grad_norm": 4.256916522979736,
+      "learning_rate": 6.409136714813571e-06,
+      "loss": 1.0089,
+      "step": 1400
+    },
+    {
+      "epoch": 0.8464328899637243,
+      "eval_loss": 0.9195001125335693,
+      "eval_runtime": 229.7223,
+      "eval_samples_per_second": 24.834,
+      "eval_steps_per_second": 1.554,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0580411124546554,
+      "grad_norm": 3.999547004699707,
+      "learning_rate": 5.233456499832046e-06,
+      "loss": 0.9749,
+      "step": 1750
+    },
+    {
+      "epoch": 1.0580411124546554,
+      "eval_loss": 0.9013524651527405,
+      "eval_runtime": 229.8561,
+      "eval_samples_per_second": 24.82,
+      "eval_steps_per_second": 1.553,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2696493349455864,
+      "grad_norm": 4.245368003845215,
+      "learning_rate": 4.0577762848505205e-06,
+      "loss": 0.9585,
+      "step": 2100
+    },
+    {
+      "epoch": 1.2696493349455864,
+      "eval_loss": 0.8950515389442444,
+      "eval_runtime": 244.5641,
+      "eval_samples_per_second": 23.327,
+      "eval_steps_per_second": 1.46,
+      "step": 2100
+    },
+    {
+      "epoch": 1.4812575574365177,
+      "grad_norm": 4.251204013824463,
+      "learning_rate": 2.882096069868996e-06,
+      "loss": 0.9381,
+      "step": 2450
+    },
+    {
+      "epoch": 1.4812575574365177,
+      "eval_loss": 0.8771164417266846,
+      "eval_runtime": 229.912,
+      "eval_samples_per_second": 24.814,
+      "eval_steps_per_second": 1.553,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6928657799274487,
+      "grad_norm": 3.8351638317108154,
+      "learning_rate": 1.7064158548874707e-06,
+      "loss": 0.9412,
+      "step": 2800
+    },
+    {
+      "epoch": 1.6928657799274487,
+      "eval_loss": 0.8743219971656799,
+      "eval_runtime": 230.0215,
+      "eval_samples_per_second": 24.802,
+      "eval_steps_per_second": 1.552,
+      "step": 2800
+    }
+  ],
+  "logging_steps": 350,
+  "max_steps": 3308,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 350,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.35596032611922e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bba6c70f7c1dab2f8354d16b0b243b22c9cf2a4831cd3899cb223cc397cafa4
+size 5240