update coin flip model

low-shot-task-specific-500-ex/coin_flip/best_model/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:206bdaf3cf034deec6de2394ba6a7b29d0b637ab2d1925332e8f1abb76025dd6
+oid sha256:52d3183a841c92cbf2569e43d73e61d3182b7acfd37ae4c1925fbcdfb5ae5037
 size 104973389

low-shot-task-specific-500-ex/coin_flip/best_model/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d53c7e4694c2cfcf4dd0eb9bfe3b05a7cd7809c6a8e6ff871c4c99d6ddfefaf4
+oid sha256:8b393456323a196b768aebc7f56020f0818e4b80d7ac3a55132cfb4b72dbd0e0
 size 209984517

low-shot-task-specific-500-ex/coin_flip/best_model/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3055502c9e3004eb987550db217f6677d695763c959badae25d773f1d985ab91
+oid sha256:e8edb68dd6c81386559d1e69774ba987c7891d658260174f825deb48eee022d8
 size 14575

low-shot-task-specific-500-ex/coin_flip/best_model/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8aea4ff6d6c72e86d24e872bf7765995d2e2e0abda70fdf4dff06ed25a492666
+oid sha256:72784f3df358e4b01284895b23305a7f44b47671c221043eb9da1558ee751bbc
 size 627

low-shot-task-specific-500-ex/coin_flip/best_model/trainer_state.json

@@ -1,135 +1,107 @@
 {
-  "best_metric": 0.17182409763336182,
-  "best_model_checkpoint": "checkpoints/instrucode/low-shot-task-specific-500-ex/coin_flip/checkpoint-60",
-  "epoch": 9.795918367346939,
+  "best_metric": 0.16547438502311707,
+  "best_model_checkpoint": "checkpoints/instrucode/low-shot-task-specific-500-ex/coin_flip/checkpoint-49",
+  "epoch": 8.0,
   "eval_steps": 500,
-  "global_step": 60,
+  "global_step": 49,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.98,
-      "eval_loss": 3.2136309146881104,
-      "eval_runtime": 1.7971,
-      "eval_samples_per_second": 27.266,
-      "eval_steps_per_second": 3.895,
+      "eval_loss": 3.1955792903900146,
+      "eval_runtime": 1.7606,
+      "eval_samples_per_second": 27.832,
+      "eval_steps_per_second": 3.976,
       "step": 6
     },
     {
       "epoch": 1.63,
       "learning_rate": 6.666666666666667e-05,
-      "loss": 3.5659,
+      "loss": 3.5618,
       "step": 10
     },
     {
       "epoch": 1.96,
-      "eval_loss": 1.1381325721740723,
-      "eval_runtime": 1.8028,
-      "eval_samples_per_second": 27.18,
-      "eval_steps_per_second": 3.883,
+      "eval_loss": 1.2319456338882446,
+      "eval_runtime": 1.775,
+      "eval_samples_per_second": 27.605,
+      "eval_steps_per_second": 3.944,
       "step": 12
     },
     {
       "epoch": 2.94,
-      "eval_loss": 0.39599937200546265,
-      "eval_runtime": 1.7938,
-      "eval_samples_per_second": 27.316,
-      "eval_steps_per_second": 3.902,
+      "eval_loss": 0.3878885507583618,
+      "eval_runtime": 1.7784,
+      "eval_samples_per_second": 27.553,
+      "eval_steps_per_second": 3.936,
       "step": 18
     },
     {
       "epoch": 3.27,
       "learning_rate": 5.333333333333333e-05,
-      "loss": 0.8239,
+      "loss": 0.8597,
       "step": 20
     },
     {
       "epoch": 3.92,
-      "eval_loss": 0.23788291215896606,
-      "eval_runtime": 1.8071,
-      "eval_samples_per_second": 27.115,
-      "eval_steps_per_second": 3.874,
+      "eval_loss": 0.21831320226192474,
+      "eval_runtime": 1.7786,
+      "eval_samples_per_second": 27.549,
+      "eval_steps_per_second": 3.936,
       "step": 24
     },
     {
       "epoch": 4.9,
       "learning_rate": 4e-05,
-      "loss": 0.2375,
+      "loss": 0.2398,
       "step": 30
     },
     {
       "epoch": 4.9,
-      "eval_loss": 0.1869448572397232,
-      "eval_runtime": 1.8046,
-      "eval_samples_per_second": 27.154,
-      "eval_steps_per_second": 3.879,
+      "eval_loss": 0.18551723659038544,
+      "eval_runtime": 1.7792,
+      "eval_samples_per_second": 27.54,
+      "eval_steps_per_second": 3.934,
       "step": 30
     },
     {
       "epoch": 5.88,
-      "eval_loss": 0.1762770116329193,
-      "eval_runtime": 1.7955,
-      "eval_samples_per_second": 27.291,
-      "eval_steps_per_second": 3.899,
+      "eval_loss": 0.1781032681465149,
+      "eval_runtime": 1.7761,
+      "eval_samples_per_second": 27.588,
+      "eval_steps_per_second": 3.941,
       "step": 36
     },
     {
       "epoch": 6.53,
       "learning_rate": 2.6666666666666667e-05,
-      "loss": 0.1756,
+      "loss": 0.1729,
       "step": 40
     },
     {
       "epoch": 6.86,
-      "eval_loss": 0.17334015667438507,
-      "eval_runtime": 1.7994,
-      "eval_samples_per_second": 27.231,
-      "eval_steps_per_second": 3.89,
+      "eval_loss": 0.17477163672447205,
+      "eval_runtime": 1.7765,
+      "eval_samples_per_second": 27.583,
+      "eval_steps_per_second": 3.94,
       "step": 42
     },
     {
       "epoch": 8.0,
-      "eval_loss": 0.17443998157978058,
-      "eval_runtime": 1.7969,
-      "eval_samples_per_second": 27.269,
-      "eval_steps_per_second": 3.896,
+      "eval_loss": 0.16547438502311707,
+      "eval_runtime": 1.7731,
+      "eval_samples_per_second": 27.635,
+      "eval_steps_per_second": 3.948,
       "step": 49
-    },
-    {
-      "epoch": 8.16,
-      "learning_rate": 1.3333333333333333e-05,
-      "loss": 0.1626,
-      "step": 50
-    },
-    {
-      "epoch": 8.98,
-      "eval_loss": 0.17577075958251953,
-      "eval_runtime": 1.7999,
-      "eval_samples_per_second": 27.223,
-      "eval_steps_per_second": 3.889,
-      "step": 55
-    },
-    {
-      "epoch": 9.8,
-      "learning_rate": 0.0,
-      "loss": 0.1535,
-      "step": 60
-    },
-    {
-      "epoch": 9.8,
-      "eval_loss": 0.17182409763336182,
-      "eval_runtime": 1.7992,
-      "eval_samples_per_second": 27.235,
-      "eval_steps_per_second": 3.891,
-      "step": 60
     }
   ],
   "logging_steps": 10,
   "max_steps": 60,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 7168130697461760.0,
+  "total_flos": 5852695881646080.0,
   "trial_name": null,
   "trial_params": null
 }

low-shot-task-specific-500-ex/coin_flip/best_model/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5173d21d2a38d1cd1cd4daa45bed5a9f6f0d64b0897c6366683a240cd58f864
+oid sha256:d6b95880f589434d710796f9f9e970ac87dcf098561bddac80dccc8c12e9aed5
 size 4091

low-shot-task-specific-500-ex/coin_flip/checkpoint-60/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "base_model_name_or_path": "meta-llama/Llama-2-13b-hf",
+  "bias": "none",
+  "enable_lora": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "merge_weights": false,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

low-shot-task-specific-500-ex/coin_flip/checkpoint-60/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dedda326c7c619c6ee1e0863e8333c86f5010e4b0bbba306c5b8dfad1ac7a35d
+size 104973389

low-shot-task-specific-500-ex/coin_flip/checkpoint-60/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a24ada10edf0d000f276cf7a2e17965fadda7af5feae736cd25bd682cbeb4fef
+size 209984517

low-shot-task-specific-500-ex/coin_flip/checkpoint-60/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3055502c9e3004eb987550db217f6677d695763c959badae25d773f1d985ab91
+size 14575

low-shot-task-specific-500-ex/coin_flip/checkpoint-60/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8aea4ff6d6c72e86d24e872bf7765995d2e2e0abda70fdf4dff06ed25a492666
+size 627

low-shot-task-specific-500-ex/coin_flip/checkpoint-60/trainer_state.json

@@ -0,0 +1,135 @@
+{
+  "best_metric": 0.16547438502311707,
+  "best_model_checkpoint": "checkpoints/instrucode/low-shot-task-specific-500-ex/coin_flip/checkpoint-49",
+  "epoch": 9.795918367346939,
+  "eval_steps": 500,
+  "global_step": 60,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.98,
+      "eval_loss": 3.1955792903900146,
+      "eval_runtime": 1.7606,
+      "eval_samples_per_second": 27.832,
+      "eval_steps_per_second": 3.976,
+      "step": 6
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 3.5618,
+      "step": 10
+    },
+    {
+      "epoch": 1.96,
+      "eval_loss": 1.2319456338882446,
+      "eval_runtime": 1.775,
+      "eval_samples_per_second": 27.605,
+      "eval_steps_per_second": 3.944,
+      "step": 12
+    },
+    {
+      "epoch": 2.94,
+      "eval_loss": 0.3878885507583618,
+      "eval_runtime": 1.7784,
+      "eval_samples_per_second": 27.553,
+      "eval_steps_per_second": 3.936,
+      "step": 18
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 5.333333333333333e-05,
+      "loss": 0.8597,
+      "step": 20
+    },
+    {
+      "epoch": 3.92,
+      "eval_loss": 0.21831320226192474,
+      "eval_runtime": 1.7786,
+      "eval_samples_per_second": 27.549,
+      "eval_steps_per_second": 3.936,
+      "step": 24
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 4e-05,
+      "loss": 0.2398,
+      "step": 30
+    },
+    {
+      "epoch": 4.9,
+      "eval_loss": 0.18551723659038544,
+      "eval_runtime": 1.7792,
+      "eval_samples_per_second": 27.54,
+      "eval_steps_per_second": 3.934,
+      "step": 30
+    },
+    {
+      "epoch": 5.88,
+      "eval_loss": 0.1781032681465149,
+      "eval_runtime": 1.7761,
+      "eval_samples_per_second": 27.588,
+      "eval_steps_per_second": 3.941,
+      "step": 36
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.1729,
+      "step": 40
+    },
+    {
+      "epoch": 6.86,
+      "eval_loss": 0.17477163672447205,
+      "eval_runtime": 1.7765,
+      "eval_samples_per_second": 27.583,
+      "eval_steps_per_second": 3.94,
+      "step": 42
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.16547438502311707,
+      "eval_runtime": 1.7731,
+      "eval_samples_per_second": 27.635,
+      "eval_steps_per_second": 3.948,
+      "step": 49
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.1579,
+      "step": 50
+    },
+    {
+      "epoch": 8.98,
+      "eval_loss": 0.18015137314796448,
+      "eval_runtime": 1.7751,
+      "eval_samples_per_second": 27.604,
+      "eval_steps_per_second": 3.943,
+      "step": 55
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 0.0,
+      "loss": 0.1442,
+      "step": 60
+    },
+    {
+      "epoch": 9.8,
+      "eval_loss": 0.1699230819940567,
+      "eval_runtime": 1.7773,
+      "eval_samples_per_second": 27.57,
+      "eval_steps_per_second": 3.939,
+      "step": 60
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 60,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 7168130697461760.0,
+  "trial_name": null,
+  "trial_params": null
+}

low-shot-task-specific-500-ex/coin_flip/checkpoint-60/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6b95880f589434d710796f9f9e970ac87dcf098561bddac80dccc8c12e9aed5
+size 4091