Upload folder using huggingface_hub

stellaris-races-2/adapter_config.json

@@ -0,0 +1,17 @@
+{
+  "base_model_name_or_path": "",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

stellaris-races-2/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b8a02415852f6ca5a4989e22aa797cca0201c1ff30660d1849cada36dab4fa1
+size 67154893

stellaris-races-2/checkpoint-650/adapter_config.json

@@ -0,0 +1,17 @@
+{
+  "base_model_name_or_path": "",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

stellaris-races-2/checkpoint-650/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39af94937e17a7fc51b0675882ba8867dbc20746362389245c2754efc53d332b
+size 67154893

stellaris-races-2/checkpoint-650/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aada27d8cedc93dd532d3ae3ce6d1a16ab64961b66d3950a43d1c28cfd03ab29
+size 134325445

stellaris-races-2/checkpoint-650/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cf33b502dfcf467fc287c562683fa4971e712ed1dc3e4f6d79fe9bc7a6e901a
+size 14575

stellaris-races-2/checkpoint-650/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b7bcb04975557f954a1f6641b49df2abd8cabc03d60787ac29eef3e47161a7c
+size 627

stellaris-races-2/checkpoint-650/trainer_state.json

@@ -0,0 +1,1836 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 59.09090909090909,
+  "global_step": 650,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.45,
+      "learning_rate": 7e-06,
+      "loss": 3.0049,
+      "step": 5
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 3.0305967330932617,
+      "eval_runtime": 0.4174,
+      "eval_samples_per_second": 14.374,
+      "eval_steps_per_second": 2.396,
+      "step": 5
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 7e-06,
+      "loss": 3.0817,
+      "step": 10
+    },
+    {
+      "epoch": 0.91,
+      "eval_loss": 3.026319742202759,
+      "eval_runtime": 0.44,
+      "eval_samples_per_second": 13.636,
+      "eval_steps_per_second": 2.273,
+      "step": 10
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 7e-06,
+      "loss": 3.0438,
+      "step": 15
+    },
+    {
+      "epoch": 1.36,
+      "eval_loss": 3.022075891494751,
+      "eval_runtime": 0.4273,
+      "eval_samples_per_second": 14.04,
+      "eval_steps_per_second": 2.34,
+      "step": 15
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7e-06,
+      "loss": 2.9792,
+      "step": 20
+    },
+    {
+      "epoch": 1.82,
+      "eval_loss": 3.016573667526245,
+      "eval_runtime": 0.4072,
+      "eval_samples_per_second": 14.733,
+      "eval_steps_per_second": 2.456,
+      "step": 20
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 7e-06,
+      "loss": 2.934,
+      "step": 25
+    },
+    {
+      "epoch": 2.27,
+      "eval_loss": 3.0110604763031006,
+      "eval_runtime": 0.4045,
+      "eval_samples_per_second": 14.832,
+      "eval_steps_per_second": 2.472,
+      "step": 25
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 7e-06,
+      "loss": 3.1707,
+      "step": 30
+    },
+    {
+      "epoch": 2.73,
+      "eval_loss": 3.004523277282715,
+      "eval_runtime": 0.408,
+      "eval_samples_per_second": 14.706,
+      "eval_steps_per_second": 2.451,
+      "step": 30
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 7e-06,
+      "loss": 2.9016,
+      "step": 35
+    },
+    {
+      "epoch": 3.18,
+      "eval_loss": 2.997664213180542,
+      "eval_runtime": 0.4355,
+      "eval_samples_per_second": 13.778,
+      "eval_steps_per_second": 2.296,
+      "step": 35
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 7e-06,
+      "loss": 3.0052,
+      "step": 40
+    },
+    {
+      "epoch": 3.64,
+      "eval_loss": 2.990278482437134,
+      "eval_runtime": 0.4213,
+      "eval_samples_per_second": 14.243,
+      "eval_steps_per_second": 2.374,
+      "step": 40
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 7e-06,
+      "loss": 3.0642,
+      "step": 45
+    },
+    {
+      "epoch": 4.09,
+      "eval_loss": 2.982116937637329,
+      "eval_runtime": 0.4131,
+      "eval_samples_per_second": 14.525,
+      "eval_steps_per_second": 2.421,
+      "step": 45
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 7e-06,
+      "loss": 2.8696,
+      "step": 50
+    },
+    {
+      "epoch": 4.55,
+      "eval_loss": 2.973396062850952,
+      "eval_runtime": 0.4071,
+      "eval_samples_per_second": 14.737,
+      "eval_steps_per_second": 2.456,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 7e-06,
+      "loss": 3.0592,
+      "step": 55
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.964240312576294,
+      "eval_runtime": 0.4248,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 2.354,
+      "step": 55
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 7e-06,
+      "loss": 2.9686,
+      "step": 60
+    },
+    {
+      "epoch": 5.45,
+      "eval_loss": 2.9537107944488525,
+      "eval_runtime": 0.434,
+      "eval_samples_per_second": 13.825,
+      "eval_steps_per_second": 2.304,
+      "step": 60
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 7e-06,
+      "loss": 2.92,
+      "step": 65
+    },
+    {
+      "epoch": 5.91,
+      "eval_loss": 2.942852258682251,
+      "eval_runtime": 0.4106,
+      "eval_samples_per_second": 14.614,
+      "eval_steps_per_second": 2.436,
+      "step": 65
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 7e-06,
+      "loss": 2.9857,
+      "step": 70
+    },
+    {
+      "epoch": 6.36,
+      "eval_loss": 2.9314286708831787,
+      "eval_runtime": 0.4051,
+      "eval_samples_per_second": 14.812,
+      "eval_steps_per_second": 2.469,
+      "step": 70
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 7e-06,
+      "loss": 2.864,
+      "step": 75
+    },
+    {
+      "epoch": 6.82,
+      "eval_loss": 2.9195594787597656,
+      "eval_runtime": 0.4379,
+      "eval_samples_per_second": 13.701,
+      "eval_steps_per_second": 2.284,
+      "step": 75
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 7e-06,
+      "loss": 2.8973,
+      "step": 80
+    },
+    {
+      "epoch": 7.27,
+      "eval_loss": 2.906179189682007,
+      "eval_runtime": 0.4044,
+      "eval_samples_per_second": 14.837,
+      "eval_steps_per_second": 2.473,
+      "step": 80
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 7e-06,
+      "loss": 2.9251,
+      "step": 85
+    },
+    {
+      "epoch": 7.73,
+      "eval_loss": 2.892069101333618,
+      "eval_runtime": 0.4132,
+      "eval_samples_per_second": 14.519,
+      "eval_steps_per_second": 2.42,
+      "step": 85
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 7e-06,
+      "loss": 2.8072,
+      "step": 90
+    },
+    {
+      "epoch": 8.18,
+      "eval_loss": 2.8766021728515625,
+      "eval_runtime": 0.4169,
+      "eval_samples_per_second": 14.391,
+      "eval_steps_per_second": 2.399,
+      "step": 90
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 7e-06,
+      "loss": 2.7725,
+      "step": 95
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 2.8613014221191406,
+      "eval_runtime": 0.4218,
+      "eval_samples_per_second": 14.223,
+      "eval_steps_per_second": 2.371,
+      "step": 95
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 7e-06,
+      "loss": 2.8727,
+      "step": 100
+    },
+    {
+      "epoch": 9.09,
+      "eval_loss": 2.845301866531372,
+      "eval_runtime": 0.4068,
+      "eval_samples_per_second": 14.748,
+      "eval_steps_per_second": 2.458,
+      "step": 100
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 7e-06,
+      "loss": 2.8022,
+      "step": 105
+    },
+    {
+      "epoch": 9.55,
+      "eval_loss": 2.8308088779449463,
+      "eval_runtime": 0.405,
+      "eval_samples_per_second": 14.814,
+      "eval_steps_per_second": 2.469,
+      "step": 105
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7e-06,
+      "loss": 2.8153,
+      "step": 110
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.8145711421966553,
+      "eval_runtime": 0.4061,
+      "eval_samples_per_second": 14.775,
+      "eval_steps_per_second": 2.463,
+      "step": 110
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 7e-06,
+      "loss": 2.8341,
+      "step": 115
+    },
+    {
+      "epoch": 10.45,
+      "eval_loss": 2.7991654872894287,
+      "eval_runtime": 0.4419,
+      "eval_samples_per_second": 13.577,
+      "eval_steps_per_second": 2.263,
+      "step": 115
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 7e-06,
+      "loss": 2.5969,
+      "step": 120
+    },
+    {
+      "epoch": 10.91,
+      "eval_loss": 2.783989906311035,
+      "eval_runtime": 0.4292,
+      "eval_samples_per_second": 13.98,
+      "eval_steps_per_second": 2.33,
+      "step": 120
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 7e-06,
+      "loss": 2.8274,
+      "step": 125
+    },
+    {
+      "epoch": 11.36,
+      "eval_loss": 2.7690820693969727,
+      "eval_runtime": 0.4123,
+      "eval_samples_per_second": 14.554,
+      "eval_steps_per_second": 2.426,
+      "step": 125
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 7e-06,
+      "loss": 2.7421,
+      "step": 130
+    },
+    {
+      "epoch": 11.82,
+      "eval_loss": 2.755321741104126,
+      "eval_runtime": 0.4081,
+      "eval_samples_per_second": 14.703,
+      "eval_steps_per_second": 2.451,
+      "step": 130
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 7e-06,
+      "loss": 2.635,
+      "step": 135
+    },
+    {
+      "epoch": 12.27,
+      "eval_loss": 2.742081642150879,
+      "eval_runtime": 0.4061,
+      "eval_samples_per_second": 14.774,
+      "eval_steps_per_second": 2.462,
+      "step": 135
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 7e-06,
+      "loss": 2.7192,
+      "step": 140
+    },
+    {
+      "epoch": 12.73,
+      "eval_loss": 2.729800224304199,
+      "eval_runtime": 0.4106,
+      "eval_samples_per_second": 14.613,
+      "eval_steps_per_second": 2.436,
+      "step": 140
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 7e-06,
+      "loss": 2.5953,
+      "step": 145
+    },
+    {
+      "epoch": 13.18,
+      "eval_loss": 2.7182958126068115,
+      "eval_runtime": 0.4151,
+      "eval_samples_per_second": 14.454,
+      "eval_steps_per_second": 2.409,
+      "step": 145
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 7e-06,
+      "loss": 2.687,
+      "step": 150
+    },
+    {
+      "epoch": 13.64,
+      "eval_loss": 2.7074530124664307,
+      "eval_runtime": 0.4084,
+      "eval_samples_per_second": 14.692,
+      "eval_steps_per_second": 2.449,
+      "step": 150
+    },
+    {
+      "epoch": 14.09,
+      "learning_rate": 7e-06,
+      "loss": 2.6218,
+      "step": 155
+    },
+    {
+      "epoch": 14.09,
+      "eval_loss": 2.6966331005096436,
+      "eval_runtime": 0.4099,
+      "eval_samples_per_second": 14.638,
+      "eval_steps_per_second": 2.44,
+      "step": 155
+    },
+    {
+      "epoch": 14.55,
+      "learning_rate": 7e-06,
+      "loss": 2.56,
+      "step": 160
+    },
+    {
+      "epoch": 14.55,
+      "eval_loss": 2.685837745666504,
+      "eval_runtime": 0.4082,
+      "eval_samples_per_second": 14.7,
+      "eval_steps_per_second": 2.45,
+      "step": 160
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 7e-06,
+      "loss": 2.6864,
+      "step": 165
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.675248384475708,
+      "eval_runtime": 0.4084,
+      "eval_samples_per_second": 14.69,
+      "eval_steps_per_second": 2.448,
+      "step": 165
+    },
+    {
+      "epoch": 15.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4397,
+      "step": 170
+    },
+    {
+      "epoch": 15.45,
+      "eval_loss": 2.6641385555267334,
+      "eval_runtime": 0.4077,
+      "eval_samples_per_second": 14.718,
+      "eval_steps_per_second": 2.453,
+      "step": 170
+    },
+    {
+      "epoch": 15.91,
+      "learning_rate": 7e-06,
+      "loss": 2.709,
+      "step": 175
+    },
+    {
+      "epoch": 15.91,
+      "eval_loss": 2.652723550796509,
+      "eval_runtime": 0.4138,
+      "eval_samples_per_second": 14.498,
+      "eval_steps_per_second": 2.416,
+      "step": 175
+    },
+    {
+      "epoch": 16.36,
+      "learning_rate": 7e-06,
+      "loss": 2.5046,
+      "step": 180
+    },
+    {
+      "epoch": 16.36,
+      "eval_loss": 2.6419966220855713,
+      "eval_runtime": 0.4086,
+      "eval_samples_per_second": 14.684,
+      "eval_steps_per_second": 2.447,
+      "step": 180
+    },
+    {
+      "epoch": 16.82,
+      "learning_rate": 7e-06,
+      "loss": 2.5968,
+      "step": 185
+    },
+    {
+      "epoch": 16.82,
+      "eval_loss": 2.6306352615356445,
+      "eval_runtime": 0.4096,
+      "eval_samples_per_second": 14.647,
+      "eval_steps_per_second": 2.441,
+      "step": 185
+    },
+    {
+      "epoch": 17.27,
+      "learning_rate": 7e-06,
+      "loss": 2.6083,
+      "step": 190
+    },
+    {
+      "epoch": 17.27,
+      "eval_loss": 2.619220733642578,
+      "eval_runtime": 0.4137,
+      "eval_samples_per_second": 14.502,
+      "eval_steps_per_second": 2.417,
+      "step": 190
+    },
+    {
+      "epoch": 17.73,
+      "learning_rate": 7e-06,
+      "loss": 2.51,
+      "step": 195
+    },
+    {
+      "epoch": 17.73,
+      "eval_loss": 2.6068971157073975,
+      "eval_runtime": 0.4084,
+      "eval_samples_per_second": 14.693,
+      "eval_steps_per_second": 2.449,
+      "step": 195
+    },
+    {
+      "epoch": 18.18,
+      "learning_rate": 7e-06,
+      "loss": 2.5301,
+      "step": 200
+    },
+    {
+      "epoch": 18.18,
+      "eval_loss": 2.5943899154663086,
+      "eval_runtime": 0.4198,
+      "eval_samples_per_second": 14.291,
+      "eval_steps_per_second": 2.382,
+      "step": 200
+    },
+    {
+      "epoch": 18.64,
+      "learning_rate": 7e-06,
+      "loss": 2.5177,
+      "step": 205
+    },
+    {
+      "epoch": 18.64,
+      "eval_loss": 2.58101487159729,
+      "eval_runtime": 0.4131,
+      "eval_samples_per_second": 14.523,
+      "eval_steps_per_second": 2.421,
+      "step": 205
+    },
+    {
+      "epoch": 19.09,
+      "learning_rate": 7e-06,
+      "loss": 2.3431,
+      "step": 210
+    },
+    {
+      "epoch": 19.09,
+      "eval_loss": 2.568328619003296,
+      "eval_runtime": 0.4161,
+      "eval_samples_per_second": 14.42,
+      "eval_steps_per_second": 2.403,
+      "step": 210
+    },
+    {
+      "epoch": 19.55,
+      "learning_rate": 7e-06,
+      "loss": 2.358,
+      "step": 215
+    },
+    {
+      "epoch": 19.55,
+      "eval_loss": 2.556185007095337,
+      "eval_runtime": 0.4167,
+      "eval_samples_per_second": 14.399,
+      "eval_steps_per_second": 2.4,
+      "step": 215
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 7e-06,
+      "loss": 2.4723,
+      "step": 220
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 2.54498553276062,
+      "eval_runtime": 0.4187,
+      "eval_samples_per_second": 14.331,
+      "eval_steps_per_second": 2.388,
+      "step": 220
+    },
+    {
+      "epoch": 20.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4351,
+      "step": 225
+    },
+    {
+      "epoch": 20.45,
+      "eval_loss": 2.5338480472564697,
+      "eval_runtime": 0.4148,
+      "eval_samples_per_second": 14.463,
+      "eval_steps_per_second": 2.411,
+      "step": 225
+    },
+    {
+      "epoch": 20.91,
+      "learning_rate": 7e-06,
+      "loss": 2.3407,
+      "step": 230
+    },
+    {
+      "epoch": 20.91,
+      "eval_loss": 2.522146463394165,
+      "eval_runtime": 0.4166,
+      "eval_samples_per_second": 14.401,
+      "eval_steps_per_second": 2.4,
+      "step": 230
+    },
+    {
+      "epoch": 21.36,
+      "learning_rate": 7e-06,
+      "loss": 2.4691,
+      "step": 235
+    },
+    {
+      "epoch": 21.36,
+      "eval_loss": 2.5105435848236084,
+      "eval_runtime": 0.4117,
+      "eval_samples_per_second": 14.572,
+      "eval_steps_per_second": 2.429,
+      "step": 235
+    },
+    {
+      "epoch": 21.82,
+      "learning_rate": 7e-06,
+      "loss": 2.3515,
+      "step": 240
+    },
+    {
+      "epoch": 21.82,
+      "eval_loss": 2.498257637023926,
+      "eval_runtime": 0.4513,
+      "eval_samples_per_second": 13.296,
+      "eval_steps_per_second": 2.216,
+      "step": 240
+    },
+    {
+      "epoch": 22.27,
+      "learning_rate": 7e-06,
+      "loss": 2.2231,
+      "step": 245
+    },
+    {
+      "epoch": 22.27,
+      "eval_loss": 2.4853975772857666,
+      "eval_runtime": 0.4211,
+      "eval_samples_per_second": 14.247,
+      "eval_steps_per_second": 2.375,
+      "step": 245
+    },
+    {
+      "epoch": 22.73,
+      "learning_rate": 7e-06,
+      "loss": 2.2849,
+      "step": 250
+    },
+    {
+      "epoch": 22.73,
+      "eval_loss": 2.474109411239624,
+      "eval_runtime": 0.422,
+      "eval_samples_per_second": 14.218,
+      "eval_steps_per_second": 2.37,
+      "step": 250
+    },
+    {
+      "epoch": 23.18,
+      "learning_rate": 7e-06,
+      "loss": 2.4086,
+      "step": 255
+    },
+    {
+      "epoch": 23.18,
+      "eval_loss": 2.4653780460357666,
+      "eval_runtime": 0.4081,
+      "eval_samples_per_second": 14.702,
+      "eval_steps_per_second": 2.45,
+      "step": 255
+    },
+    {
+      "epoch": 23.64,
+      "learning_rate": 7e-06,
+      "loss": 2.1498,
+      "step": 260
+    },
+    {
+      "epoch": 23.64,
+      "eval_loss": 2.4591386318206787,
+      "eval_runtime": 0.4098,
+      "eval_samples_per_second": 14.64,
+      "eval_steps_per_second": 2.44,
+      "step": 260
+    },
+    {
+      "epoch": 24.09,
+      "learning_rate": 7e-06,
+      "loss": 2.3414,
+      "step": 265
+    },
+    {
+      "epoch": 24.09,
+      "eval_loss": 2.454268455505371,
+      "eval_runtime": 0.4114,
+      "eval_samples_per_second": 14.586,
+      "eval_steps_per_second": 2.431,
+      "step": 265
+    },
+    {
+      "epoch": 24.55,
+      "learning_rate": 7e-06,
+      "loss": 2.1754,
+      "step": 270
+    },
+    {
+      "epoch": 24.55,
+      "eval_loss": 2.4511959552764893,
+      "eval_runtime": 0.4144,
+      "eval_samples_per_second": 14.479,
+      "eval_steps_per_second": 2.413,
+      "step": 270
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3577,
+      "step": 275
+    },
+    {
+      "epoch": 25.0,
+      "eval_loss": 2.4483206272125244,
+      "eval_runtime": 0.4191,
+      "eval_samples_per_second": 14.316,
+      "eval_steps_per_second": 2.386,
+      "step": 275
+    },
+    {
+      "epoch": 25.45,
+      "learning_rate": 7e-06,
+      "loss": 2.1878,
+      "step": 280
+    },
+    {
+      "epoch": 25.45,
+      "eval_loss": 2.4461658000946045,
+      "eval_runtime": 0.4103,
+      "eval_samples_per_second": 14.624,
+      "eval_steps_per_second": 2.437,
+      "step": 280
+    },
+    {
+      "epoch": 25.91,
+      "learning_rate": 7e-06,
+      "loss": 2.3283,
+      "step": 285
+    },
+    {
+      "epoch": 25.91,
+      "eval_loss": 2.4441139698028564,
+      "eval_runtime": 0.4091,
+      "eval_samples_per_second": 14.667,
+      "eval_steps_per_second": 2.444,
+      "step": 285
+    },
+    {
+      "epoch": 26.36,
+      "learning_rate": 7e-06,
+      "loss": 2.2045,
+      "step": 290
+    },
+    {
+      "epoch": 26.36,
+      "eval_loss": 2.44203782081604,
+      "eval_runtime": 0.4117,
+      "eval_samples_per_second": 14.574,
+      "eval_steps_per_second": 2.429,
+      "step": 290
+    },
+    {
+      "epoch": 26.82,
+      "learning_rate": 7e-06,
+      "loss": 2.2802,
+      "step": 295
+    },
+    {
+      "epoch": 26.82,
+      "eval_loss": 2.4402430057525635,
+      "eval_runtime": 0.4202,
+      "eval_samples_per_second": 14.279,
+      "eval_steps_per_second": 2.38,
+      "step": 295
+    },
+    {
+      "epoch": 27.27,
+      "learning_rate": 7e-06,
+      "loss": 2.0865,
+      "step": 300
+    },
+    {
+      "epoch": 27.27,
+      "eval_loss": 2.438746690750122,
+      "eval_runtime": 0.4215,
+      "eval_samples_per_second": 14.235,
+      "eval_steps_per_second": 2.373,
+      "step": 300
+    },
+    {
+      "epoch": 27.73,
+      "learning_rate": 7e-06,
+      "loss": 2.275,
+      "step": 305
+    },
+    {
+      "epoch": 27.73,
+      "eval_loss": 2.4373252391815186,
+      "eval_runtime": 0.4173,
+      "eval_samples_per_second": 14.377,
+      "eval_steps_per_second": 2.396,
+      "step": 305
+    },
+    {
+      "epoch": 28.18,
+      "learning_rate": 7e-06,
+      "loss": 2.277,
+      "step": 310
+    },
+    {
+      "epoch": 28.18,
+      "eval_loss": 2.4359142780303955,
+      "eval_runtime": 0.4204,
+      "eval_samples_per_second": 14.273,
+      "eval_steps_per_second": 2.379,
+      "step": 310
+    },
+    {
+      "epoch": 28.64,
+      "learning_rate": 7e-06,
+      "loss": 2.2201,
+      "step": 315
+    },
+    {
+      "epoch": 28.64,
+      "eval_loss": 2.435053825378418,
+      "eval_runtime": 0.4099,
+      "eval_samples_per_second": 14.637,
+      "eval_steps_per_second": 2.44,
+      "step": 315
+    },
+    {
+      "epoch": 29.09,
+      "learning_rate": 7e-06,
+      "loss": 2.1064,
+      "step": 320
+    },
+    {
+      "epoch": 29.09,
+      "eval_loss": 2.4337093830108643,
+      "eval_runtime": 0.4202,
+      "eval_samples_per_second": 14.277,
+      "eval_steps_per_second": 2.38,
+      "step": 320
+    },
+    {
+      "epoch": 29.55,
+      "learning_rate": 7e-06,
+      "loss": 2.1089,
+      "step": 325
+    },
+    {
+      "epoch": 29.55,
+      "eval_loss": 2.4328536987304688,
+      "eval_runtime": 0.4206,
+      "eval_samples_per_second": 14.265,
+      "eval_steps_per_second": 2.377,
+      "step": 325
+    },
+    {
+      "epoch": 30.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3056,
+      "step": 330
+    },
+    {
+      "epoch": 30.0,
+      "eval_loss": 2.43233323097229,
+      "eval_runtime": 0.4239,
+      "eval_samples_per_second": 14.155,
+      "eval_steps_per_second": 2.359,
+      "step": 330
+    },
+    {
+      "epoch": 30.45,
+      "learning_rate": 7e-06,
+      "loss": 2.0761,
+      "step": 335
+    },
+    {
+      "epoch": 30.45,
+      "eval_loss": 2.431633949279785,
+      "eval_runtime": 0.4101,
+      "eval_samples_per_second": 14.631,
+      "eval_steps_per_second": 2.438,
+      "step": 335
+    },
+    {
+      "epoch": 30.91,
+      "learning_rate": 7e-06,
+      "loss": 2.2653,
+      "step": 340
+    },
+    {
+      "epoch": 30.91,
+      "eval_loss": 2.43125319480896,
+      "eval_runtime": 0.4102,
+      "eval_samples_per_second": 14.626,
+      "eval_steps_per_second": 2.438,
+      "step": 340
+    },
+    {
+      "epoch": 31.36,
+      "learning_rate": 7e-06,
+      "loss": 2.1188,
+      "step": 345
+    },
+    {
+      "epoch": 31.36,
+      "eval_loss": 2.430781126022339,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.715,
+      "eval_steps_per_second": 2.452,
+      "step": 345
+    },
+    {
+      "epoch": 31.82,
+      "learning_rate": 7e-06,
+      "loss": 2.23,
+      "step": 350
+    },
+    {
+      "epoch": 31.82,
+      "eval_loss": 2.4303319454193115,
+      "eval_runtime": 0.4174,
+      "eval_samples_per_second": 14.376,
+      "eval_steps_per_second": 2.396,
+      "step": 350
+    },
+    {
+      "epoch": 32.27,
+      "learning_rate": 7e-06,
+      "loss": 2.0428,
+      "step": 355
+    },
+    {
+      "epoch": 32.27,
+      "eval_loss": 2.4302732944488525,
+      "eval_runtime": 0.4196,
+      "eval_samples_per_second": 14.301,
+      "eval_steps_per_second": 2.383,
+      "step": 355
+    },
+    {
+      "epoch": 32.73,
+      "learning_rate": 7e-06,
+      "loss": 2.1743,
+      "step": 360
+    },
+    {
+      "epoch": 32.73,
+      "eval_loss": 2.4303529262542725,
+      "eval_runtime": 0.419,
+      "eval_samples_per_second": 14.32,
+      "eval_steps_per_second": 2.387,
+      "step": 360
+    },
+    {
+      "epoch": 33.18,
+      "learning_rate": 7e-06,
+      "loss": 2.2025,
+      "step": 365
+    },
+    {
+      "epoch": 33.18,
+      "eval_loss": 2.430536985397339,
+      "eval_runtime": 0.4276,
+      "eval_samples_per_second": 14.032,
+      "eval_steps_per_second": 2.339,
+      "step": 365
+    },
+    {
+      "epoch": 33.64,
+      "learning_rate": 7e-06,
+      "loss": 1.9553,
+      "step": 370
+    },
+    {
+      "epoch": 33.64,
+      "eval_loss": 2.4313228130340576,
+      "eval_runtime": 0.4148,
+      "eval_samples_per_second": 14.465,
+      "eval_steps_per_second": 2.411,
+      "step": 370
+    },
+    {
+      "epoch": 34.09,
+      "learning_rate": 7e-06,
+      "loss": 2.2841,
+      "step": 375
+    },
+    {
+      "epoch": 34.09,
+      "eval_loss": 2.431515693664551,
+      "eval_runtime": 0.4289,
+      "eval_samples_per_second": 13.989,
+      "eval_steps_per_second": 2.332,
+      "step": 375
+    },
+    {
+      "epoch": 34.55,
+      "learning_rate": 7e-06,
+      "loss": 2.0907,
+      "step": 380
+    },
+    {
+      "epoch": 34.55,
+      "eval_loss": 2.4317359924316406,
+      "eval_runtime": 0.4149,
+      "eval_samples_per_second": 14.462,
+      "eval_steps_per_second": 2.41,
+      "step": 380
+    },
+    {
+      "epoch": 35.0,
+      "learning_rate": 7e-06,
+      "loss": 2.1472,
+      "step": 385
+    },
+    {
+      "epoch": 35.0,
+      "eval_loss": 2.432173728942871,
+      "eval_runtime": 0.4348,
+      "eval_samples_per_second": 13.801,
+      "eval_steps_per_second": 2.3,
+      "step": 385
+    },
+    {
+      "epoch": 35.45,
+      "learning_rate": 7e-06,
+      "loss": 2.1642,
+      "step": 390
+    },
+    {
+      "epoch": 35.45,
+      "eval_loss": 2.433072090148926,
+      "eval_runtime": 0.4156,
+      "eval_samples_per_second": 14.438,
+      "eval_steps_per_second": 2.406,
+      "step": 390
+    },
+    {
+      "epoch": 35.91,
+      "learning_rate": 7e-06,
+      "loss": 2.0184,
+      "step": 395
+    },
+    {
+      "epoch": 35.91,
+      "eval_loss": 2.4334452152252197,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.728,
+      "eval_steps_per_second": 2.455,
+      "step": 395
+    },
+    {
+      "epoch": 36.36,
+      "learning_rate": 7e-06,
+      "loss": 2.1743,
+      "step": 400
+    },
+    {
+      "epoch": 36.36,
+      "eval_loss": 2.435070753097534,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.712,
+      "eval_steps_per_second": 2.452,
+      "step": 400
+    },
+    {
+      "epoch": 36.82,
+      "learning_rate": 7e-06,
+      "loss": 2.0438,
+      "step": 405
+    },
+    {
+      "epoch": 36.82,
+      "eval_loss": 2.436411142349243,
+      "eval_runtime": 0.4349,
+      "eval_samples_per_second": 13.797,
+      "eval_steps_per_second": 2.299,
+      "step": 405
+    },
+    {
+      "epoch": 37.27,
+      "learning_rate": 7e-06,
+      "loss": 2.1146,
+      "step": 410
+    },
+    {
+      "epoch": 37.27,
+      "eval_loss": 2.4375665187835693,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.729,
+      "eval_steps_per_second": 2.455,
+      "step": 410
+    },
+    {
+      "epoch": 37.73,
+      "learning_rate": 7e-06,
+      "loss": 2.0405,
+      "step": 415
+    },
+    {
+      "epoch": 37.73,
+      "eval_loss": 2.4385483264923096,
+      "eval_runtime": 0.4305,
+      "eval_samples_per_second": 13.938,
+      "eval_steps_per_second": 2.323,
+      "step": 415
+    },
+    {
+      "epoch": 38.18,
+      "learning_rate": 7e-06,
+      "loss": 1.9226,
+      "step": 420
+    },
+    {
+      "epoch": 38.18,
+      "eval_loss": 2.4396464824676514,
+      "eval_runtime": 0.4116,
+      "eval_samples_per_second": 14.579,
+      "eval_steps_per_second": 2.43,
+      "step": 420
+    },
+    {
+      "epoch": 38.64,
+      "learning_rate": 7e-06,
+      "loss": 2.1808,
+      "step": 425
+    },
+    {
+      "epoch": 38.64,
+      "eval_loss": 2.441523790359497,
+      "eval_runtime": 0.4292,
+      "eval_samples_per_second": 13.98,
+      "eval_steps_per_second": 2.33,
+      "step": 425
+    },
+    {
+      "epoch": 39.09,
+      "learning_rate": 7e-06,
+      "loss": 1.9993,
+      "step": 430
+    },
+    {
+      "epoch": 39.09,
+      "eval_loss": 2.4434759616851807,
+      "eval_runtime": 0.4337,
+      "eval_samples_per_second": 13.836,
+      "eval_steps_per_second": 2.306,
+      "step": 430
+    },
+    {
+      "epoch": 39.55,
+      "learning_rate": 7e-06,
+      "loss": 2.0123,
+      "step": 435
+    },
+    {
+      "epoch": 39.55,
+      "eval_loss": 2.445014715194702,
+      "eval_runtime": 0.4082,
+      "eval_samples_per_second": 14.699,
+      "eval_steps_per_second": 2.45,
+      "step": 435
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 7e-06,
+      "loss": 2.0762,
+      "step": 440
+    },
+    {
+      "epoch": 40.0,
+      "eval_loss": 2.446848154067993,
+      "eval_runtime": 0.4152,
+      "eval_samples_per_second": 14.451,
+      "eval_steps_per_second": 2.408,
+      "step": 440
+    },
+    {
+      "epoch": 40.45,
+      "learning_rate": 7e-06,
+      "loss": 2.0079,
+      "step": 445
+    },
+    {
+      "epoch": 40.45,
+      "eval_loss": 2.449875831604004,
+      "eval_runtime": 0.4092,
+      "eval_samples_per_second": 14.663,
+      "eval_steps_per_second": 2.444,
+      "step": 445
+    },
+    {
+      "epoch": 40.91,
+      "learning_rate": 7e-06,
+      "loss": 2.0856,
+      "step": 450
+    },
+    {
+      "epoch": 40.91,
+      "eval_loss": 2.4520113468170166,
+      "eval_runtime": 0.4073,
+      "eval_samples_per_second": 14.73,
+      "eval_steps_per_second": 2.455,
+      "step": 450
+    },
+    {
+      "epoch": 41.36,
+      "learning_rate": 7e-06,
+      "loss": 1.8473,
+      "step": 455
+    },
+    {
+      "epoch": 41.36,
+      "eval_loss": 2.45548415184021,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.728,
+      "eval_steps_per_second": 2.455,
+      "step": 455
+    },
+    {
+      "epoch": 41.82,
+      "learning_rate": 7e-06,
+      "loss": 2.1588,
+      "step": 460
+    },
+    {
+      "epoch": 41.82,
+      "eval_loss": 2.458113431930542,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.728,
+      "eval_steps_per_second": 2.455,
+      "step": 460
+    },
+    {
+      "epoch": 42.27,
+      "learning_rate": 7e-06,
+      "loss": 2.0635,
+      "step": 465
+    },
+    {
+      "epoch": 42.27,
+      "eval_loss": 2.457326650619507,
+      "eval_runtime": 0.4122,
+      "eval_samples_per_second": 14.556,
+      "eval_steps_per_second": 2.426,
+      "step": 465
+    },
+    {
+      "epoch": 42.73,
+      "learning_rate": 7e-06,
+      "loss": 1.8833,
+      "step": 470
+    },
+    {
+      "epoch": 42.73,
+      "eval_loss": 2.457960367202759,
+      "eval_runtime": 0.408,
+      "eval_samples_per_second": 14.707,
+      "eval_steps_per_second": 2.451,
+      "step": 470
+    },
+    {
+      "epoch": 43.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8618,
+      "step": 475
+    },
+    {
+      "epoch": 43.18,
+      "eval_loss": 2.4618539810180664,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.727,
+      "eval_steps_per_second": 2.455,
+      "step": 475
+    },
+    {
+      "epoch": 43.64,
+      "learning_rate": 7e-06,
+      "loss": 1.9884,
+      "step": 480
+    },
+    {
+      "epoch": 43.64,
+      "eval_loss": 2.4648358821868896,
+      "eval_runtime": 0.4066,
+      "eval_samples_per_second": 14.756,
+      "eval_steps_per_second": 2.459,
+      "step": 480
+    },
+    {
+      "epoch": 44.09,
+      "learning_rate": 7e-06,
+      "loss": 1.9292,
+      "step": 485
+    },
+    {
+      "epoch": 44.09,
+      "eval_loss": 2.467848777770996,
+      "eval_runtime": 0.4077,
+      "eval_samples_per_second": 14.718,
+      "eval_steps_per_second": 2.453,
+      "step": 485
+    },
+    {
+      "epoch": 44.55,
+      "learning_rate": 7e-06,
+      "loss": 1.8937,
+      "step": 490
+    },
+    {
+      "epoch": 44.55,
+      "eval_loss": 2.471271276473999,
+      "eval_runtime": 0.4098,
+      "eval_samples_per_second": 14.642,
+      "eval_steps_per_second": 2.44,
+      "step": 490
+    },
+    {
+      "epoch": 45.0,
+      "learning_rate": 7e-06,
+      "loss": 2.0734,
+      "step": 495
+    },
+    {
+      "epoch": 45.0,
+      "eval_loss": 2.4725749492645264,
+      "eval_runtime": 0.4143,
+      "eval_samples_per_second": 14.482,
+      "eval_steps_per_second": 2.414,
+      "step": 495
+    },
+    {
+      "epoch": 45.45,
+      "learning_rate": 7e-06,
+      "loss": 1.99,
+      "step": 500
+    },
+    {
+      "epoch": 45.45,
+      "eval_loss": 2.473742723464966,
+      "eval_runtime": 0.4224,
+      "eval_samples_per_second": 14.205,
+      "eval_steps_per_second": 2.367,
+      "step": 500
+    },
+    {
+      "epoch": 45.91,
+      "learning_rate": 7e-06,
+      "loss": 1.8188,
+      "step": 505
+    },
+    {
+      "epoch": 45.91,
+      "eval_loss": 2.4784109592437744,
+      "eval_runtime": 0.4294,
+      "eval_samples_per_second": 13.973,
+      "eval_steps_per_second": 2.329,
+      "step": 505
+    },
+    {
+      "epoch": 46.36,
+      "learning_rate": 7e-06,
+      "loss": 1.8354,
+      "step": 510
+    },
+    {
+      "epoch": 46.36,
+      "eval_loss": 2.485568046569824,
+      "eval_runtime": 0.4576,
+      "eval_samples_per_second": 13.112,
+      "eval_steps_per_second": 2.185,
+      "step": 510
+    },
+    {
+      "epoch": 46.82,
+      "learning_rate": 7e-06,
+      "loss": 1.9735,
+      "step": 515
+    },
+    {
+      "epoch": 46.82,
+      "eval_loss": 2.4888756275177,
+      "eval_runtime": 0.4219,
+      "eval_samples_per_second": 14.22,
+      "eval_steps_per_second": 2.37,
+      "step": 515
+    },
+    {
+      "epoch": 47.27,
+      "learning_rate": 7e-06,
+      "loss": 1.953,
+      "step": 520
+    },
+    {
+      "epoch": 47.27,
+      "eval_loss": 2.4856255054473877,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.712,
+      "eval_steps_per_second": 2.452,
+      "step": 520
+    },
+    {
+      "epoch": 47.73,
+      "learning_rate": 7e-06,
+      "loss": 1.8938,
+      "step": 525
+    },
+    {
+      "epoch": 47.73,
+      "eval_loss": 2.48818039894104,
+      "eval_runtime": 0.4106,
+      "eval_samples_per_second": 14.613,
+      "eval_steps_per_second": 2.436,
+      "step": 525
+    },
+    {
+      "epoch": 48.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8276,
+      "step": 530
+    },
+    {
+      "epoch": 48.18,
+      "eval_loss": 2.4947216510772705,
+      "eval_runtime": 0.4098,
+      "eval_samples_per_second": 14.642,
+      "eval_steps_per_second": 2.44,
+      "step": 530
+    },
+    {
+      "epoch": 48.64,
+      "learning_rate": 7e-06,
+      "loss": 1.9289,
+      "step": 535
+    },
+    {
+      "epoch": 48.64,
+      "eval_loss": 2.500380516052246,
+      "eval_runtime": 0.4688,
+      "eval_samples_per_second": 12.798,
+      "eval_steps_per_second": 2.133,
+      "step": 535
+    },
+    {
+      "epoch": 49.09,
+      "learning_rate": 7e-06,
+      "loss": 1.8077,
+      "step": 540
+    },
+    {
+      "epoch": 49.09,
+      "eval_loss": 2.50510573387146,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.712,
+      "eval_steps_per_second": 2.452,
+      "step": 540
+    },
+    {
+      "epoch": 49.55,
+      "learning_rate": 7e-06,
+      "loss": 1.9091,
+      "step": 545
+    },
+    {
+      "epoch": 49.55,
+      "eval_loss": 2.5079917907714844,
+      "eval_runtime": 0.4339,
+      "eval_samples_per_second": 13.829,
+      "eval_steps_per_second": 2.305,
+      "step": 545
+    },
+    {
+      "epoch": 50.0,
+      "learning_rate": 7e-06,
+      "loss": 1.8086,
+      "step": 550
+    },
+    {
+      "epoch": 50.0,
+      "eval_loss": 2.5064351558685303,
+      "eval_runtime": 0.4096,
+      "eval_samples_per_second": 14.648,
+      "eval_steps_per_second": 2.441,
+      "step": 550
+    },
+    {
+      "epoch": 50.45,
+      "learning_rate": 7e-06,
+      "loss": 1.8497,
+      "step": 555
+    },
+    {
+      "epoch": 50.45,
+      "eval_loss": 2.5103464126586914,
+      "eval_runtime": 0.4116,
+      "eval_samples_per_second": 14.577,
+      "eval_steps_per_second": 2.429,
+      "step": 555
+    },
+    {
+      "epoch": 50.91,
+      "learning_rate": 7e-06,
+      "loss": 1.8287,
+      "step": 560
+    },
+    {
+      "epoch": 50.91,
+      "eval_loss": 2.5173234939575195,
+      "eval_runtime": 0.4197,
+      "eval_samples_per_second": 14.296,
+      "eval_steps_per_second": 2.383,
+      "step": 560
+    },
+    {
+      "epoch": 51.36,
+      "learning_rate": 7e-06,
+      "loss": 1.8288,
+      "step": 565
+    },
+    {
+      "epoch": 51.36,
+      "eval_loss": 2.5230062007904053,
+      "eval_runtime": 0.4091,
+      "eval_samples_per_second": 14.665,
+      "eval_steps_per_second": 2.444,
+      "step": 565
+    },
+    {
+      "epoch": 51.82,
+      "learning_rate": 7e-06,
+      "loss": 1.803,
+      "step": 570
+    },
+    {
+      "epoch": 51.82,
+      "eval_loss": 2.52628493309021,
+      "eval_runtime": 0.4316,
+      "eval_samples_per_second": 13.903,
+      "eval_steps_per_second": 2.317,
+      "step": 570
+    },
+    {
+      "epoch": 52.27,
+      "learning_rate": 7e-06,
+      "loss": 1.7498,
+      "step": 575
+    },
+    {
+      "epoch": 52.27,
+      "eval_loss": 2.531799077987671,
+      "eval_runtime": 0.4158,
+      "eval_samples_per_second": 14.431,
+      "eval_steps_per_second": 2.405,
+      "step": 575
+    },
+    {
+      "epoch": 52.73,
+      "learning_rate": 7e-06,
+      "loss": 1.7879,
+      "step": 580
+    },
+    {
+      "epoch": 52.73,
+      "eval_loss": 2.5400400161743164,
+      "eval_runtime": 0.4282,
+      "eval_samples_per_second": 14.011,
+      "eval_steps_per_second": 2.335,
+      "step": 580
+    },
+    {
+      "epoch": 53.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8234,
+      "step": 585
+    },
+    {
+      "epoch": 53.18,
+      "eval_loss": 2.5378053188323975,
+      "eval_runtime": 0.4216,
+      "eval_samples_per_second": 14.231,
+      "eval_steps_per_second": 2.372,
+      "step": 585
+    },
+    {
+      "epoch": 53.64,
+      "learning_rate": 7e-06,
+      "loss": 1.7653,
+      "step": 590
+    },
+    {
+      "epoch": 53.64,
+      "eval_loss": 2.541457176208496,
+      "eval_runtime": 0.4186,
+      "eval_samples_per_second": 14.335,
+      "eval_steps_per_second": 2.389,
+      "step": 590
+    },
+    {
+      "epoch": 54.09,
+      "learning_rate": 7e-06,
+      "loss": 1.7322,
+      "step": 595
+    },
+    {
+      "epoch": 54.09,
+      "eval_loss": 2.5493710041046143,
+      "eval_runtime": 0.411,
+      "eval_samples_per_second": 14.598,
+      "eval_steps_per_second": 2.433,
+      "step": 595
+    },
+    {
+      "epoch": 54.55,
+      "learning_rate": 7e-06,
+      "loss": 1.6075,
+      "step": 600
+    },
+    {
+      "epoch": 54.55,
+      "eval_loss": 2.5551865100860596,
+      "eval_runtime": 0.4082,
+      "eval_samples_per_second": 14.699,
+      "eval_steps_per_second": 2.45,
+      "step": 600
+    },
+    {
+      "epoch": 55.0,
+      "learning_rate": 7e-06,
+      "loss": 1.8756,
+      "step": 605
+    },
+    {
+      "epoch": 55.0,
+      "eval_loss": 2.563950300216675,
+      "eval_runtime": 0.4162,
+      "eval_samples_per_second": 14.417,
+      "eval_steps_per_second": 2.403,
+      "step": 605
+    },
+    {
+      "epoch": 55.45,
+      "learning_rate": 7e-06,
+      "loss": 1.6549,
+      "step": 610
+    },
+    {
+      "epoch": 55.45,
+      "eval_loss": 2.567089080810547,
+      "eval_runtime": 0.4089,
+      "eval_samples_per_second": 14.673,
+      "eval_steps_per_second": 2.445,
+      "step": 610
+    },
+    {
+      "epoch": 55.91,
+      "learning_rate": 7e-06,
+      "loss": 1.7606,
+      "step": 615
+    },
+    {
+      "epoch": 55.91,
+      "eval_loss": 2.5678203105926514,
+      "eval_runtime": 0.4075,
+      "eval_samples_per_second": 14.723,
+      "eval_steps_per_second": 2.454,
+      "step": 615
+    },
+    {
+      "epoch": 56.36,
+      "learning_rate": 7e-06,
+      "loss": 1.6907,
+      "step": 620
+    },
+    {
+      "epoch": 56.36,
+      "eval_loss": 2.5747501850128174,
+      "eval_runtime": 0.4086,
+      "eval_samples_per_second": 14.683,
+      "eval_steps_per_second": 2.447,
+      "step": 620
+    },
+    {
+      "epoch": 56.82,
+      "learning_rate": 7e-06,
+      "loss": 1.7504,
+      "step": 625
+    },
+    {
+      "epoch": 56.82,
+      "eval_loss": 2.5831196308135986,
+      "eval_runtime": 0.41,
+      "eval_samples_per_second": 14.635,
+      "eval_steps_per_second": 2.439,
+      "step": 625
+    },
+    {
+      "epoch": 57.27,
+      "learning_rate": 7e-06,
+      "loss": 1.6549,
+      "step": 630
+    },
+    {
+      "epoch": 57.27,
+      "eval_loss": 2.5876691341400146,
+      "eval_runtime": 0.4183,
+      "eval_samples_per_second": 14.343,
+      "eval_steps_per_second": 2.391,
+      "step": 630
+    },
+    {
+      "epoch": 57.73,
+      "learning_rate": 7e-06,
+      "loss": 1.694,
+      "step": 635
+    },
+    {
+      "epoch": 57.73,
+      "eval_loss": 2.5929791927337646,
+      "eval_runtime": 0.4201,
+      "eval_samples_per_second": 14.283,
+      "eval_steps_per_second": 2.38,
+      "step": 635
+    },
+    {
+      "epoch": 58.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8248,
+      "step": 640
+    },
+    {
+      "epoch": 58.18,
+      "eval_loss": 2.5961735248565674,
+      "eval_runtime": 0.4089,
+      "eval_samples_per_second": 14.675,
+      "eval_steps_per_second": 2.446,
+      "step": 640
+    },
+    {
+      "epoch": 58.64,
+      "learning_rate": 7e-06,
+      "loss": 1.4473,
+      "step": 645
+    },
+    {
+      "epoch": 58.64,
+      "eval_loss": 2.5989134311676025,
+      "eval_runtime": 0.4215,
+      "eval_samples_per_second": 14.237,
+      "eval_steps_per_second": 2.373,
+      "step": 645
+    },
+    {
+      "epoch": 59.09,
+      "learning_rate": 7e-06,
+      "loss": 1.7458,
+      "step": 650
+    },
+    {
+      "epoch": 59.09,
+      "eval_loss": 2.615797758102417,
+      "eval_runtime": 0.4163,
+      "eval_samples_per_second": 14.414,
+      "eval_steps_per_second": 2.402,
+      "step": 650
+    }
+  ],
+  "max_steps": 1100,
+  "num_train_epochs": 100,
+  "total_flos": 197646153179136.0,
+  "trial_name": null,
+  "trial_params": null
+}

stellaris-races-2/checkpoint-650/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26d2b678a4d9f285f917a1b9046a7a81b7861eec33925ea0244852fdce629f6f
+size 3963

stellaris-races-2/checkpoint-700/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6332478cdf7a8712c2bec05164e4c72bc3e5dcf080ce79baca079e55566bc5d1
+size 134325445

stellaris-races-2/checkpoint-700/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd398218c018852befcb98e08d6adc91c2bd8acb8804c9aa0fa95dbcd21f7991
+size 67154893

stellaris-races-2/checkpoint-700/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13d42a8af34d95de94a838bb64cb91d6a86e05477c01335df80403312f583e2e
+size 14575

stellaris-races-2/checkpoint-700/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35b3c99f2846d66428cacd962c99b801c22421bed4d2a9ef542f5ce16d83410a
+size 627

stellaris-races-2/checkpoint-700/trainer_state.json

@@ -0,0 +1,1976 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 63.63636363636363,
+  "global_step": 700,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.45,
+      "learning_rate": 7e-06,
+      "loss": 3.0049,
+      "step": 5
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 3.0305967330932617,
+      "eval_runtime": 0.4174,
+      "eval_samples_per_second": 14.374,
+      "eval_steps_per_second": 2.396,
+      "step": 5
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 7e-06,
+      "loss": 3.0817,
+      "step": 10
+    },
+    {
+      "epoch": 0.91,
+      "eval_loss": 3.026319742202759,
+      "eval_runtime": 0.44,
+      "eval_samples_per_second": 13.636,
+      "eval_steps_per_second": 2.273,
+      "step": 10
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 7e-06,
+      "loss": 3.0438,
+      "step": 15
+    },
+    {
+      "epoch": 1.36,
+      "eval_loss": 3.022075891494751,
+      "eval_runtime": 0.4273,
+      "eval_samples_per_second": 14.04,
+      "eval_steps_per_second": 2.34,
+      "step": 15
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7e-06,
+      "loss": 2.9792,
+      "step": 20
+    },
+    {
+      "epoch": 1.82,
+      "eval_loss": 3.016573667526245,
+      "eval_runtime": 0.4072,
+      "eval_samples_per_second": 14.733,
+      "eval_steps_per_second": 2.456,
+      "step": 20
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 7e-06,
+      "loss": 2.934,
+      "step": 25
+    },
+    {
+      "epoch": 2.27,
+      "eval_loss": 3.0110604763031006,
+      "eval_runtime": 0.4045,
+      "eval_samples_per_second": 14.832,
+      "eval_steps_per_second": 2.472,
+      "step": 25
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 7e-06,
+      "loss": 3.1707,
+      "step": 30
+    },
+    {
+      "epoch": 2.73,
+      "eval_loss": 3.004523277282715,
+      "eval_runtime": 0.408,
+      "eval_samples_per_second": 14.706,
+      "eval_steps_per_second": 2.451,
+      "step": 30
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 7e-06,
+      "loss": 2.9016,
+      "step": 35
+    },
+    {
+      "epoch": 3.18,
+      "eval_loss": 2.997664213180542,
+      "eval_runtime": 0.4355,
+      "eval_samples_per_second": 13.778,
+      "eval_steps_per_second": 2.296,
+      "step": 35
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 7e-06,
+      "loss": 3.0052,
+      "step": 40
+    },
+    {
+      "epoch": 3.64,
+      "eval_loss": 2.990278482437134,
+      "eval_runtime": 0.4213,
+      "eval_samples_per_second": 14.243,
+      "eval_steps_per_second": 2.374,
+      "step": 40
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 7e-06,
+      "loss": 3.0642,
+      "step": 45
+    },
+    {
+      "epoch": 4.09,
+      "eval_loss": 2.982116937637329,
+      "eval_runtime": 0.4131,
+      "eval_samples_per_second": 14.525,
+      "eval_steps_per_second": 2.421,
+      "step": 45
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 7e-06,
+      "loss": 2.8696,
+      "step": 50
+    },
+    {
+      "epoch": 4.55,
+      "eval_loss": 2.973396062850952,
+      "eval_runtime": 0.4071,
+      "eval_samples_per_second": 14.737,
+      "eval_steps_per_second": 2.456,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 7e-06,
+      "loss": 3.0592,
+      "step": 55
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.964240312576294,
+      "eval_runtime": 0.4248,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 2.354,
+      "step": 55
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 7e-06,
+      "loss": 2.9686,
+      "step": 60
+    },
+    {
+      "epoch": 5.45,
+      "eval_loss": 2.9537107944488525,
+      "eval_runtime": 0.434,
+      "eval_samples_per_second": 13.825,
+      "eval_steps_per_second": 2.304,
+      "step": 60
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 7e-06,
+      "loss": 2.92,
+      "step": 65
+    },
+    {
+      "epoch": 5.91,
+      "eval_loss": 2.942852258682251,
+      "eval_runtime": 0.4106,
+      "eval_samples_per_second": 14.614,
+      "eval_steps_per_second": 2.436,
+      "step": 65
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 7e-06,
+      "loss": 2.9857,
+      "step": 70
+    },
+    {
+      "epoch": 6.36,
+      "eval_loss": 2.9314286708831787,
+      "eval_runtime": 0.4051,
+      "eval_samples_per_second": 14.812,
+      "eval_steps_per_second": 2.469,
+      "step": 70
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 7e-06,
+      "loss": 2.864,
+      "step": 75
+    },
+    {
+      "epoch": 6.82,
+      "eval_loss": 2.9195594787597656,
+      "eval_runtime": 0.4379,
+      "eval_samples_per_second": 13.701,
+      "eval_steps_per_second": 2.284,
+      "step": 75
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 7e-06,
+      "loss": 2.8973,
+      "step": 80
+    },
+    {
+      "epoch": 7.27,
+      "eval_loss": 2.906179189682007,
+      "eval_runtime": 0.4044,
+      "eval_samples_per_second": 14.837,
+      "eval_steps_per_second": 2.473,
+      "step": 80
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 7e-06,
+      "loss": 2.9251,
+      "step": 85
+    },
+    {
+      "epoch": 7.73,
+      "eval_loss": 2.892069101333618,
+      "eval_runtime": 0.4132,
+      "eval_samples_per_second": 14.519,
+      "eval_steps_per_second": 2.42,
+      "step": 85
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 7e-06,
+      "loss": 2.8072,
+      "step": 90
+    },
+    {
+      "epoch": 8.18,
+      "eval_loss": 2.8766021728515625,
+      "eval_runtime": 0.4169,
+      "eval_samples_per_second": 14.391,
+      "eval_steps_per_second": 2.399,
+      "step": 90
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 7e-06,
+      "loss": 2.7725,
+      "step": 95
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 2.8613014221191406,
+      "eval_runtime": 0.4218,
+      "eval_samples_per_second": 14.223,
+      "eval_steps_per_second": 2.371,
+      "step": 95
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 7e-06,
+      "loss": 2.8727,
+      "step": 100
+    },
+    {
+      "epoch": 9.09,
+      "eval_loss": 2.845301866531372,
+      "eval_runtime": 0.4068,
+      "eval_samples_per_second": 14.748,
+      "eval_steps_per_second": 2.458,
+      "step": 100
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 7e-06,
+      "loss": 2.8022,
+      "step": 105
+    },
+    {
+      "epoch": 9.55,
+      "eval_loss": 2.8308088779449463,
+      "eval_runtime": 0.405,
+      "eval_samples_per_second": 14.814,
+      "eval_steps_per_second": 2.469,
+      "step": 105
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7e-06,
+      "loss": 2.8153,
+      "step": 110
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.8145711421966553,
+      "eval_runtime": 0.4061,
+      "eval_samples_per_second": 14.775,
+      "eval_steps_per_second": 2.463,
+      "step": 110
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 7e-06,
+      "loss": 2.8341,
+      "step": 115
+    },
+    {
+      "epoch": 10.45,
+      "eval_loss": 2.7991654872894287,
+      "eval_runtime": 0.4419,
+      "eval_samples_per_second": 13.577,
+      "eval_steps_per_second": 2.263,
+      "step": 115
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 7e-06,
+      "loss": 2.5969,
+      "step": 120
+    },
+    {
+      "epoch": 10.91,
+      "eval_loss": 2.783989906311035,
+      "eval_runtime": 0.4292,
+      "eval_samples_per_second": 13.98,
+      "eval_steps_per_second": 2.33,
+      "step": 120
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 7e-06,
+      "loss": 2.8274,
+      "step": 125
+    },
+    {
+      "epoch": 11.36,
+      "eval_loss": 2.7690820693969727,
+      "eval_runtime": 0.4123,
+      "eval_samples_per_second": 14.554,
+      "eval_steps_per_second": 2.426,
+      "step": 125
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 7e-06,
+      "loss": 2.7421,
+      "step": 130
+    },
+    {
+      "epoch": 11.82,
+      "eval_loss": 2.755321741104126,
+      "eval_runtime": 0.4081,
+      "eval_samples_per_second": 14.703,
+      "eval_steps_per_second": 2.451,
+      "step": 130
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 7e-06,
+      "loss": 2.635,
+      "step": 135
+    },
+    {
+      "epoch": 12.27,
+      "eval_loss": 2.742081642150879,
+      "eval_runtime": 0.4061,
+      "eval_samples_per_second": 14.774,
+      "eval_steps_per_second": 2.462,
+      "step": 135
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 7e-06,
+      "loss": 2.7192,
+      "step": 140
+    },
+    {
+      "epoch": 12.73,
+      "eval_loss": 2.729800224304199,
+      "eval_runtime": 0.4106,
+      "eval_samples_per_second": 14.613,
+      "eval_steps_per_second": 2.436,
+      "step": 140
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 7e-06,
+      "loss": 2.5953,
+      "step": 145
+    },
+    {
+      "epoch": 13.18,
+      "eval_loss": 2.7182958126068115,
+      "eval_runtime": 0.4151,
+      "eval_samples_per_second": 14.454,
+      "eval_steps_per_second": 2.409,
+      "step": 145
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 7e-06,
+      "loss": 2.687,
+      "step": 150
+    },
+    {
+      "epoch": 13.64,
+      "eval_loss": 2.7074530124664307,
+      "eval_runtime": 0.4084,
+      "eval_samples_per_second": 14.692,
+      "eval_steps_per_second": 2.449,
+      "step": 150
+    },
+    {
+      "epoch": 14.09,
+      "learning_rate": 7e-06,
+      "loss": 2.6218,
+      "step": 155
+    },
+    {
+      "epoch": 14.09,
+      "eval_loss": 2.6966331005096436,
+      "eval_runtime": 0.4099,
+      "eval_samples_per_second": 14.638,
+      "eval_steps_per_second": 2.44,
+      "step": 155
+    },
+    {
+      "epoch": 14.55,
+      "learning_rate": 7e-06,
+      "loss": 2.56,
+      "step": 160
+    },
+    {
+      "epoch": 14.55,
+      "eval_loss": 2.685837745666504,
+      "eval_runtime": 0.4082,
+      "eval_samples_per_second": 14.7,
+      "eval_steps_per_second": 2.45,
+      "step": 160
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 7e-06,
+      "loss": 2.6864,
+      "step": 165
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.675248384475708,
+      "eval_runtime": 0.4084,
+      "eval_samples_per_second": 14.69,
+      "eval_steps_per_second": 2.448,
+      "step": 165
+    },
+    {
+      "epoch": 15.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4397,
+      "step": 170
+    },
+    {
+      "epoch": 15.45,
+      "eval_loss": 2.6641385555267334,
+      "eval_runtime": 0.4077,
+      "eval_samples_per_second": 14.718,
+      "eval_steps_per_second": 2.453,
+      "step": 170
+    },
+    {
+      "epoch": 15.91,
+      "learning_rate": 7e-06,
+      "loss": 2.709,
+      "step": 175
+    },
+    {
+      "epoch": 15.91,
+      "eval_loss": 2.652723550796509,
+      "eval_runtime": 0.4138,
+      "eval_samples_per_second": 14.498,
+      "eval_steps_per_second": 2.416,
+      "step": 175
+    },
+    {
+      "epoch": 16.36,
+      "learning_rate": 7e-06,
+      "loss": 2.5046,
+      "step": 180
+    },
+    {
+      "epoch": 16.36,
+      "eval_loss": 2.6419966220855713,
+      "eval_runtime": 0.4086,
+      "eval_samples_per_second": 14.684,
+      "eval_steps_per_second": 2.447,
+      "step": 180
+    },
+    {
+      "epoch": 16.82,
+      "learning_rate": 7e-06,
+      "loss": 2.5968,
+      "step": 185
+    },
+    {
+      "epoch": 16.82,
+      "eval_loss": 2.6306352615356445,
+      "eval_runtime": 0.4096,
+      "eval_samples_per_second": 14.647,
+      "eval_steps_per_second": 2.441,
+      "step": 185
+    },
+    {
+      "epoch": 17.27,
+      "learning_rate": 7e-06,
+      "loss": 2.6083,
+      "step": 190
+    },
+    {
+      "epoch": 17.27,
+      "eval_loss": 2.619220733642578,
+      "eval_runtime": 0.4137,
+      "eval_samples_per_second": 14.502,
+      "eval_steps_per_second": 2.417,
+      "step": 190
+    },
+    {
+      "epoch": 17.73,
+      "learning_rate": 7e-06,
+      "loss": 2.51,
+      "step": 195
+    },
+    {
+      "epoch": 17.73,
+      "eval_loss": 2.6068971157073975,
+      "eval_runtime": 0.4084,
+      "eval_samples_per_second": 14.693,
+      "eval_steps_per_second": 2.449,
+      "step": 195
+    },
+    {
+      "epoch": 18.18,
+      "learning_rate": 7e-06,
+      "loss": 2.5301,
+      "step": 200
+    },
+    {
+      "epoch": 18.18,
+      "eval_loss": 2.5943899154663086,
+      "eval_runtime": 0.4198,
+      "eval_samples_per_second": 14.291,
+      "eval_steps_per_second": 2.382,
+      "step": 200
+    },
+    {
+      "epoch": 18.64,
+      "learning_rate": 7e-06,
+      "loss": 2.5177,
+      "step": 205
+    },
+    {
+      "epoch": 18.64,
+      "eval_loss": 2.58101487159729,
+      "eval_runtime": 0.4131,
+      "eval_samples_per_second": 14.523,
+      "eval_steps_per_second": 2.421,
+      "step": 205
+    },
+    {
+      "epoch": 19.09,
+      "learning_rate": 7e-06,
+      "loss": 2.3431,
+      "step": 210
+    },
+    {
+      "epoch": 19.09,
+      "eval_loss": 2.568328619003296,
+      "eval_runtime": 0.4161,
+      "eval_samples_per_second": 14.42,
+      "eval_steps_per_second": 2.403,
+      "step": 210
+    },
+    {
+      "epoch": 19.55,
+      "learning_rate": 7e-06,
+      "loss": 2.358,
+      "step": 215
+    },
+    {
+      "epoch": 19.55,
+      "eval_loss": 2.556185007095337,
+      "eval_runtime": 0.4167,
+      "eval_samples_per_second": 14.399,
+      "eval_steps_per_second": 2.4,
+      "step": 215
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 7e-06,
+      "loss": 2.4723,
+      "step": 220
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 2.54498553276062,
+      "eval_runtime": 0.4187,
+      "eval_samples_per_second": 14.331,
+      "eval_steps_per_second": 2.388,
+      "step": 220
+    },
+    {
+      "epoch": 20.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4351,
+      "step": 225
+    },
+    {
+      "epoch": 20.45,
+      "eval_loss": 2.5338480472564697,
+      "eval_runtime": 0.4148,
+      "eval_samples_per_second": 14.463,
+      "eval_steps_per_second": 2.411,
+      "step": 225
+    },
+    {
+      "epoch": 20.91,
+      "learning_rate": 7e-06,
+      "loss": 2.3407,
+      "step": 230
+    },
+    {
+      "epoch": 20.91,
+      "eval_loss": 2.522146463394165,
+      "eval_runtime": 0.4166,
+      "eval_samples_per_second": 14.401,
+      "eval_steps_per_second": 2.4,
+      "step": 230
+    },
+    {
+      "epoch": 21.36,
+      "learning_rate": 7e-06,
+      "loss": 2.4691,
+      "step": 235
+    },
+    {
+      "epoch": 21.36,
+      "eval_loss": 2.5105435848236084,
+      "eval_runtime": 0.4117,
+      "eval_samples_per_second": 14.572,
+      "eval_steps_per_second": 2.429,
+      "step": 235
+    },
+    {
+      "epoch": 21.82,
+      "learning_rate": 7e-06,
+      "loss": 2.3515,
+      "step": 240
+    },
+    {
+      "epoch": 21.82,
+      "eval_loss": 2.498257637023926,
+      "eval_runtime": 0.4513,
+      "eval_samples_per_second": 13.296,
+      "eval_steps_per_second": 2.216,
+      "step": 240
+    },
+    {
+      "epoch": 22.27,
+      "learning_rate": 7e-06,
+      "loss": 2.2231,
+      "step": 245
+    },
+    {
+      "epoch": 22.27,
+      "eval_loss": 2.4853975772857666,
+      "eval_runtime": 0.4211,
+      "eval_samples_per_second": 14.247,
+      "eval_steps_per_second": 2.375,
+      "step": 245
+    },
+    {
+      "epoch": 22.73,
+      "learning_rate": 7e-06,
+      "loss": 2.2849,
+      "step": 250
+    },
+    {
+      "epoch": 22.73,
+      "eval_loss": 2.474109411239624,
+      "eval_runtime": 0.422,
+      "eval_samples_per_second": 14.218,
+      "eval_steps_per_second": 2.37,
+      "step": 250
+    },
+    {
+      "epoch": 23.18,
+      "learning_rate": 7e-06,
+      "loss": 2.4086,
+      "step": 255
+    },
+    {
+      "epoch": 23.18,
+      "eval_loss": 2.4653780460357666,
+      "eval_runtime": 0.4081,
+      "eval_samples_per_second": 14.702,
+      "eval_steps_per_second": 2.45,
+      "step": 255
+    },
+    {
+      "epoch": 23.64,
+      "learning_rate": 7e-06,
+      "loss": 2.1498,
+      "step": 260
+    },
+    {
+      "epoch": 23.64,
+      "eval_loss": 2.4591386318206787,
+      "eval_runtime": 0.4098,
+      "eval_samples_per_second": 14.64,
+      "eval_steps_per_second": 2.44,
+      "step": 260
+    },
+    {
+      "epoch": 24.09,
+      "learning_rate": 7e-06,
+      "loss": 2.3414,
+      "step": 265
+    },
+    {
+      "epoch": 24.09,
+      "eval_loss": 2.454268455505371,
+      "eval_runtime": 0.4114,
+      "eval_samples_per_second": 14.586,
+      "eval_steps_per_second": 2.431,
+      "step": 265
+    },
+    {
+      "epoch": 24.55,
+      "learning_rate": 7e-06,
+      "loss": 2.1754,
+      "step": 270
+    },
+    {
+      "epoch": 24.55,
+      "eval_loss": 2.4511959552764893,
+      "eval_runtime": 0.4144,
+      "eval_samples_per_second": 14.479,
+      "eval_steps_per_second": 2.413,
+      "step": 270
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3577,
+      "step": 275
+    },
+    {
+      "epoch": 25.0,
+      "eval_loss": 2.4483206272125244,
+      "eval_runtime": 0.4191,
+      "eval_samples_per_second": 14.316,
+      "eval_steps_per_second": 2.386,
+      "step": 275
+    },
+    {
+      "epoch": 25.45,
+      "learning_rate": 7e-06,
+      "loss": 2.1878,
+      "step": 280
+    },
+    {
+      "epoch": 25.45,
+      "eval_loss": 2.4461658000946045,
+      "eval_runtime": 0.4103,
+      "eval_samples_per_second": 14.624,
+      "eval_steps_per_second": 2.437,
+      "step": 280
+    },
+    {
+      "epoch": 25.91,
+      "learning_rate": 7e-06,
+      "loss": 2.3283,
+      "step": 285
+    },
+    {
+      "epoch": 25.91,
+      "eval_loss": 2.4441139698028564,
+      "eval_runtime": 0.4091,
+      "eval_samples_per_second": 14.667,
+      "eval_steps_per_second": 2.444,
+      "step": 285
+    },
+    {
+      "epoch": 26.36,
+      "learning_rate": 7e-06,
+      "loss": 2.2045,
+      "step": 290
+    },
+    {
+      "epoch": 26.36,
+      "eval_loss": 2.44203782081604,
+      "eval_runtime": 0.4117,
+      "eval_samples_per_second": 14.574,
+      "eval_steps_per_second": 2.429,
+      "step": 290
+    },
+    {
+      "epoch": 26.82,
+      "learning_rate": 7e-06,
+      "loss": 2.2802,
+      "step": 295
+    },
+    {
+      "epoch": 26.82,
+      "eval_loss": 2.4402430057525635,
+      "eval_runtime": 0.4202,
+      "eval_samples_per_second": 14.279,
+      "eval_steps_per_second": 2.38,
+      "step": 295
+    },
+    {
+      "epoch": 27.27,
+      "learning_rate": 7e-06,
+      "loss": 2.0865,
+      "step": 300
+    },
+    {
+      "epoch": 27.27,
+      "eval_loss": 2.438746690750122,
+      "eval_runtime": 0.4215,
+      "eval_samples_per_second": 14.235,
+      "eval_steps_per_second": 2.373,
+      "step": 300
+    },
+    {
+      "epoch": 27.73,
+      "learning_rate": 7e-06,
+      "loss": 2.275,
+      "step": 305
+    },
+    {
+      "epoch": 27.73,
+      "eval_loss": 2.4373252391815186,
+      "eval_runtime": 0.4173,
+      "eval_samples_per_second": 14.377,
+      "eval_steps_per_second": 2.396,
+      "step": 305
+    },
+    {
+      "epoch": 28.18,
+      "learning_rate": 7e-06,
+      "loss": 2.277,
+      "step": 310
+    },
+    {
+      "epoch": 28.18,
+      "eval_loss": 2.4359142780303955,
+      "eval_runtime": 0.4204,
+      "eval_samples_per_second": 14.273,
+      "eval_steps_per_second": 2.379,
+      "step": 310
+    },
+    {
+      "epoch": 28.64,
+      "learning_rate": 7e-06,
+      "loss": 2.2201,
+      "step": 315
+    },
+    {
+      "epoch": 28.64,
+      "eval_loss": 2.435053825378418,
+      "eval_runtime": 0.4099,
+      "eval_samples_per_second": 14.637,
+      "eval_steps_per_second": 2.44,
+      "step": 315
+    },
+    {
+      "epoch": 29.09,
+      "learning_rate": 7e-06,
+      "loss": 2.1064,
+      "step": 320
+    },
+    {
+      "epoch": 29.09,
+      "eval_loss": 2.4337093830108643,
+      "eval_runtime": 0.4202,
+      "eval_samples_per_second": 14.277,
+      "eval_steps_per_second": 2.38,
+      "step": 320
+    },
+    {
+      "epoch": 29.55,
+      "learning_rate": 7e-06,
+      "loss": 2.1089,
+      "step": 325
+    },
+    {
+      "epoch": 29.55,
+      "eval_loss": 2.4328536987304688,
+      "eval_runtime": 0.4206,
+      "eval_samples_per_second": 14.265,
+      "eval_steps_per_second": 2.377,
+      "step": 325
+    },
+    {
+      "epoch": 30.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3056,
+      "step": 330
+    },
+    {
+      "epoch": 30.0,
+      "eval_loss": 2.43233323097229,
+      "eval_runtime": 0.4239,
+      "eval_samples_per_second": 14.155,
+      "eval_steps_per_second": 2.359,
+      "step": 330
+    },
+    {
+      "epoch": 30.45,
+      "learning_rate": 7e-06,
+      "loss": 2.0761,
+      "step": 335
+    },
+    {
+      "epoch": 30.45,
+      "eval_loss": 2.431633949279785,
+      "eval_runtime": 0.4101,
+      "eval_samples_per_second": 14.631,
+      "eval_steps_per_second": 2.438,
+      "step": 335
+    },
+    {
+      "epoch": 30.91,
+      "learning_rate": 7e-06,
+      "loss": 2.2653,
+      "step": 340
+    },
+    {
+      "epoch": 30.91,
+      "eval_loss": 2.43125319480896,
+      "eval_runtime": 0.4102,
+      "eval_samples_per_second": 14.626,
+      "eval_steps_per_second": 2.438,
+      "step": 340
+    },
+    {
+      "epoch": 31.36,
+      "learning_rate": 7e-06,
+      "loss": 2.1188,
+      "step": 345
+    },
+    {
+      "epoch": 31.36,
+      "eval_loss": 2.430781126022339,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.715,
+      "eval_steps_per_second": 2.452,
+      "step": 345
+    },
+    {
+      "epoch": 31.82,
+      "learning_rate": 7e-06,
+      "loss": 2.23,
+      "step": 350
+    },
+    {
+      "epoch": 31.82,
+      "eval_loss": 2.4303319454193115,
+      "eval_runtime": 0.4174,
+      "eval_samples_per_second": 14.376,
+      "eval_steps_per_second": 2.396,
+      "step": 350
+    },
+    {
+      "epoch": 32.27,
+      "learning_rate": 7e-06,
+      "loss": 2.0428,
+      "step": 355
+    },
+    {
+      "epoch": 32.27,
+      "eval_loss": 2.4302732944488525,
+      "eval_runtime": 0.4196,
+      "eval_samples_per_second": 14.301,
+      "eval_steps_per_second": 2.383,
+      "step": 355
+    },
+    {
+      "epoch": 32.73,
+      "learning_rate": 7e-06,
+      "loss": 2.1743,
+      "step": 360
+    },
+    {
+      "epoch": 32.73,
+      "eval_loss": 2.4303529262542725,
+      "eval_runtime": 0.419,
+      "eval_samples_per_second": 14.32,
+      "eval_steps_per_second": 2.387,
+      "step": 360
+    },
+    {
+      "epoch": 33.18,
+      "learning_rate": 7e-06,
+      "loss": 2.2025,
+      "step": 365
+    },
+    {
+      "epoch": 33.18,
+      "eval_loss": 2.430536985397339,
+      "eval_runtime": 0.4276,
+      "eval_samples_per_second": 14.032,
+      "eval_steps_per_second": 2.339,
+      "step": 365
+    },
+    {
+      "epoch": 33.64,
+      "learning_rate": 7e-06,
+      "loss": 1.9553,
+      "step": 370
+    },
+    {
+      "epoch": 33.64,
+      "eval_loss": 2.4313228130340576,
+      "eval_runtime": 0.4148,
+      "eval_samples_per_second": 14.465,
+      "eval_steps_per_second": 2.411,
+      "step": 370
+    },
+    {
+      "epoch": 34.09,
+      "learning_rate": 7e-06,
+      "loss": 2.2841,
+      "step": 375
+    },
+    {
+      "epoch": 34.09,
+      "eval_loss": 2.431515693664551,
+      "eval_runtime": 0.4289,
+      "eval_samples_per_second": 13.989,
+      "eval_steps_per_second": 2.332,
+      "step": 375
+    },
+    {
+      "epoch": 34.55,
+      "learning_rate": 7e-06,
+      "loss": 2.0907,
+      "step": 380
+    },
+    {
+      "epoch": 34.55,
+      "eval_loss": 2.4317359924316406,
+      "eval_runtime": 0.4149,
+      "eval_samples_per_second": 14.462,
+      "eval_steps_per_second": 2.41,
+      "step": 380
+    },
+    {
+      "epoch": 35.0,
+      "learning_rate": 7e-06,
+      "loss": 2.1472,
+      "step": 385
+    },
+    {
+      "epoch": 35.0,
+      "eval_loss": 2.432173728942871,
+      "eval_runtime": 0.4348,
+      "eval_samples_per_second": 13.801,
+      "eval_steps_per_second": 2.3,
+      "step": 385
+    },
+    {
+      "epoch": 35.45,
+      "learning_rate": 7e-06,
+      "loss": 2.1642,
+      "step": 390
+    },
+    {
+      "epoch": 35.45,
+      "eval_loss": 2.433072090148926,
+      "eval_runtime": 0.4156,
+      "eval_samples_per_second": 14.438,
+      "eval_steps_per_second": 2.406,
+      "step": 390
+    },
+    {
+      "epoch": 35.91,
+      "learning_rate": 7e-06,
+      "loss": 2.0184,
+      "step": 395
+    },
+    {
+      "epoch": 35.91,
+      "eval_loss": 2.4334452152252197,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.728,
+      "eval_steps_per_second": 2.455,
+      "step": 395
+    },
+    {
+      "epoch": 36.36,
+      "learning_rate": 7e-06,
+      "loss": 2.1743,
+      "step": 400
+    },
+    {
+      "epoch": 36.36,
+      "eval_loss": 2.435070753097534,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.712,
+      "eval_steps_per_second": 2.452,
+      "step": 400
+    },
+    {
+      "epoch": 36.82,
+      "learning_rate": 7e-06,
+      "loss": 2.0438,
+      "step": 405
+    },
+    {
+      "epoch": 36.82,
+      "eval_loss": 2.436411142349243,
+      "eval_runtime": 0.4349,
+      "eval_samples_per_second": 13.797,
+      "eval_steps_per_second": 2.299,
+      "step": 405
+    },
+    {
+      "epoch": 37.27,
+      "learning_rate": 7e-06,
+      "loss": 2.1146,
+      "step": 410
+    },
+    {
+      "epoch": 37.27,
+      "eval_loss": 2.4375665187835693,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.729,
+      "eval_steps_per_second": 2.455,
+      "step": 410
+    },
+    {
+      "epoch": 37.73,
+      "learning_rate": 7e-06,
+      "loss": 2.0405,
+      "step": 415
+    },
+    {
+      "epoch": 37.73,
+      "eval_loss": 2.4385483264923096,
+      "eval_runtime": 0.4305,
+      "eval_samples_per_second": 13.938,
+      "eval_steps_per_second": 2.323,
+      "step": 415
+    },
+    {
+      "epoch": 38.18,
+      "learning_rate": 7e-06,
+      "loss": 1.9226,
+      "step": 420
+    },
+    {
+      "epoch": 38.18,
+      "eval_loss": 2.4396464824676514,
+      "eval_runtime": 0.4116,
+      "eval_samples_per_second": 14.579,
+      "eval_steps_per_second": 2.43,
+      "step": 420
+    },
+    {
+      "epoch": 38.64,
+      "learning_rate": 7e-06,
+      "loss": 2.1808,
+      "step": 425
+    },
+    {
+      "epoch": 38.64,
+      "eval_loss": 2.441523790359497,
+      "eval_runtime": 0.4292,
+      "eval_samples_per_second": 13.98,
+      "eval_steps_per_second": 2.33,
+      "step": 425
+    },
+    {
+      "epoch": 39.09,
+      "learning_rate": 7e-06,
+      "loss": 1.9993,
+      "step": 430
+    },
+    {
+      "epoch": 39.09,
+      "eval_loss": 2.4434759616851807,
+      "eval_runtime": 0.4337,
+      "eval_samples_per_second": 13.836,
+      "eval_steps_per_second": 2.306,
+      "step": 430
+    },
+    {
+      "epoch": 39.55,
+      "learning_rate": 7e-06,
+      "loss": 2.0123,
+      "step": 435
+    },
+    {
+      "epoch": 39.55,
+      "eval_loss": 2.445014715194702,
+      "eval_runtime": 0.4082,
+      "eval_samples_per_second": 14.699,
+      "eval_steps_per_second": 2.45,
+      "step": 435
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 7e-06,
+      "loss": 2.0762,
+      "step": 440
+    },
+    {
+      "epoch": 40.0,
+      "eval_loss": 2.446848154067993,
+      "eval_runtime": 0.4152,
+      "eval_samples_per_second": 14.451,
+      "eval_steps_per_second": 2.408,
+      "step": 440
+    },
+    {
+      "epoch": 40.45,
+      "learning_rate": 7e-06,
+      "loss": 2.0079,
+      "step": 445
+    },
+    {
+      "epoch": 40.45,
+      "eval_loss": 2.449875831604004,
+      "eval_runtime": 0.4092,
+      "eval_samples_per_second": 14.663,
+      "eval_steps_per_second": 2.444,
+      "step": 445
+    },
+    {
+      "epoch": 40.91,
+      "learning_rate": 7e-06,
+      "loss": 2.0856,
+      "step": 450
+    },
+    {
+      "epoch": 40.91,
+      "eval_loss": 2.4520113468170166,
+      "eval_runtime": 0.4073,
+      "eval_samples_per_second": 14.73,
+      "eval_steps_per_second": 2.455,
+      "step": 450
+    },
+    {
+      "epoch": 41.36,
+      "learning_rate": 7e-06,
+      "loss": 1.8473,
+      "step": 455
+    },
+    {
+      "epoch": 41.36,
+      "eval_loss": 2.45548415184021,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.728,
+      "eval_steps_per_second": 2.455,
+      "step": 455
+    },
+    {
+      "epoch": 41.82,
+      "learning_rate": 7e-06,
+      "loss": 2.1588,
+      "step": 460
+    },
+    {
+      "epoch": 41.82,
+      "eval_loss": 2.458113431930542,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.728,
+      "eval_steps_per_second": 2.455,
+      "step": 460
+    },
+    {
+      "epoch": 42.27,
+      "learning_rate": 7e-06,
+      "loss": 2.0635,
+      "step": 465
+    },
+    {
+      "epoch": 42.27,
+      "eval_loss": 2.457326650619507,
+      "eval_runtime": 0.4122,
+      "eval_samples_per_second": 14.556,
+      "eval_steps_per_second": 2.426,
+      "step": 465
+    },
+    {
+      "epoch": 42.73,
+      "learning_rate": 7e-06,
+      "loss": 1.8833,
+      "step": 470
+    },
+    {
+      "epoch": 42.73,
+      "eval_loss": 2.457960367202759,
+      "eval_runtime": 0.408,
+      "eval_samples_per_second": 14.707,
+      "eval_steps_per_second": 2.451,
+      "step": 470
+    },
+    {
+      "epoch": 43.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8618,
+      "step": 475
+    },
+    {
+      "epoch": 43.18,
+      "eval_loss": 2.4618539810180664,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.727,
+      "eval_steps_per_second": 2.455,
+      "step": 475
+    },
+    {
+      "epoch": 43.64,
+      "learning_rate": 7e-06,
+      "loss": 1.9884,
+      "step": 480
+    },
+    {
+      "epoch": 43.64,
+      "eval_loss": 2.4648358821868896,
+      "eval_runtime": 0.4066,
+      "eval_samples_per_second": 14.756,
+      "eval_steps_per_second": 2.459,
+      "step": 480
+    },
+    {
+      "epoch": 44.09,
+      "learning_rate": 7e-06,
+      "loss": 1.9292,
+      "step": 485
+    },
+    {
+      "epoch": 44.09,
+      "eval_loss": 2.467848777770996,
+      "eval_runtime": 0.4077,
+      "eval_samples_per_second": 14.718,
+      "eval_steps_per_second": 2.453,
+      "step": 485
+    },
+    {
+      "epoch": 44.55,
+      "learning_rate": 7e-06,
+      "loss": 1.8937,
+      "step": 490
+    },
+    {
+      "epoch": 44.55,
+      "eval_loss": 2.471271276473999,
+      "eval_runtime": 0.4098,
+      "eval_samples_per_second": 14.642,
+      "eval_steps_per_second": 2.44,
+      "step": 490
+    },
+    {
+      "epoch": 45.0,
+      "learning_rate": 7e-06,
+      "loss": 2.0734,
+      "step": 495
+    },
+    {
+      "epoch": 45.0,
+      "eval_loss": 2.4725749492645264,
+      "eval_runtime": 0.4143,
+      "eval_samples_per_second": 14.482,
+      "eval_steps_per_second": 2.414,
+      "step": 495
+    },
+    {
+      "epoch": 45.45,
+      "learning_rate": 7e-06,
+      "loss": 1.99,
+      "step": 500
+    },
+    {
+      "epoch": 45.45,
+      "eval_loss": 2.473742723464966,
+      "eval_runtime": 0.4224,
+      "eval_samples_per_second": 14.205,
+      "eval_steps_per_second": 2.367,
+      "step": 500
+    },
+    {
+      "epoch": 45.91,
+      "learning_rate": 7e-06,
+      "loss": 1.8188,
+      "step": 505
+    },
+    {
+      "epoch": 45.91,
+      "eval_loss": 2.4784109592437744,
+      "eval_runtime": 0.4294,
+      "eval_samples_per_second": 13.973,
+      "eval_steps_per_second": 2.329,
+      "step": 505
+    },
+    {
+      "epoch": 46.36,
+      "learning_rate": 7e-06,
+      "loss": 1.8354,
+      "step": 510
+    },
+    {
+      "epoch": 46.36,
+      "eval_loss": 2.485568046569824,
+      "eval_runtime": 0.4576,
+      "eval_samples_per_second": 13.112,
+      "eval_steps_per_second": 2.185,
+      "step": 510
+    },
+    {
+      "epoch": 46.82,
+      "learning_rate": 7e-06,
+      "loss": 1.9735,
+      "step": 515
+    },
+    {
+      "epoch": 46.82,
+      "eval_loss": 2.4888756275177,
+      "eval_runtime": 0.4219,
+      "eval_samples_per_second": 14.22,
+      "eval_steps_per_second": 2.37,
+      "step": 515
+    },
+    {
+      "epoch": 47.27,
+      "learning_rate": 7e-06,
+      "loss": 1.953,
+      "step": 520
+    },
+    {
+      "epoch": 47.27,
+      "eval_loss": 2.4856255054473877,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.712,
+      "eval_steps_per_second": 2.452,
+      "step": 520
+    },
+    {
+      "epoch": 47.73,
+      "learning_rate": 7e-06,
+      "loss": 1.8938,
+      "step": 525
+    },
+    {
+      "epoch": 47.73,
+      "eval_loss": 2.48818039894104,
+      "eval_runtime": 0.4106,
+      "eval_samples_per_second": 14.613,
+      "eval_steps_per_second": 2.436,
+      "step": 525
+    },
+    {
+      "epoch": 48.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8276,
+      "step": 530
+    },
+    {
+      "epoch": 48.18,
+      "eval_loss": 2.4947216510772705,
+      "eval_runtime": 0.4098,
+      "eval_samples_per_second": 14.642,
+      "eval_steps_per_second": 2.44,
+      "step": 530
+    },
+    {
+      "epoch": 48.64,
+      "learning_rate": 7e-06,
+      "loss": 1.9289,
+      "step": 535
+    },
+    {
+      "epoch": 48.64,
+      "eval_loss": 2.500380516052246,
+      "eval_runtime": 0.4688,
+      "eval_samples_per_second": 12.798,
+      "eval_steps_per_second": 2.133,
+      "step": 535
+    },
+    {
+      "epoch": 49.09,
+      "learning_rate": 7e-06,
+      "loss": 1.8077,
+      "step": 540
+    },
+    {
+      "epoch": 49.09,
+      "eval_loss": 2.50510573387146,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.712,
+      "eval_steps_per_second": 2.452,
+      "step": 540
+    },
+    {
+      "epoch": 49.55,
+      "learning_rate": 7e-06,
+      "loss": 1.9091,
+      "step": 545
+    },
+    {
+      "epoch": 49.55,
+      "eval_loss": 2.5079917907714844,
+      "eval_runtime": 0.4339,
+      "eval_samples_per_second": 13.829,
+      "eval_steps_per_second": 2.305,
+      "step": 545
+    },
+    {
+      "epoch": 50.0,
+      "learning_rate": 7e-06,
+      "loss": 1.8086,
+      "step": 550
+    },
+    {
+      "epoch": 50.0,
+      "eval_loss": 2.5064351558685303,
+      "eval_runtime": 0.4096,
+      "eval_samples_per_second": 14.648,
+      "eval_steps_per_second": 2.441,
+      "step": 550
+    },
+    {
+      "epoch": 50.45,
+      "learning_rate": 7e-06,
+      "loss": 1.8497,
+      "step": 555
+    },
+    {
+      "epoch": 50.45,
+      "eval_loss": 2.5103464126586914,
+      "eval_runtime": 0.4116,
+      "eval_samples_per_second": 14.577,
+      "eval_steps_per_second": 2.429,
+      "step": 555
+    },
+    {
+      "epoch": 50.91,
+      "learning_rate": 7e-06,
+      "loss": 1.8287,
+      "step": 560
+    },
+    {
+      "epoch": 50.91,
+      "eval_loss": 2.5173234939575195,
+      "eval_runtime": 0.4197,
+      "eval_samples_per_second": 14.296,
+      "eval_steps_per_second": 2.383,
+      "step": 560
+    },
+    {
+      "epoch": 51.36,
+      "learning_rate": 7e-06,
+      "loss": 1.8288,
+      "step": 565
+    },
+    {
+      "epoch": 51.36,
+      "eval_loss": 2.5230062007904053,
+      "eval_runtime": 0.4091,
+      "eval_samples_per_second": 14.665,
+      "eval_steps_per_second": 2.444,
+      "step": 565
+    },
+    {
+      "epoch": 51.82,
+      "learning_rate": 7e-06,
+      "loss": 1.803,
+      "step": 570
+    },
+    {
+      "epoch": 51.82,
+      "eval_loss": 2.52628493309021,
+      "eval_runtime": 0.4316,
+      "eval_samples_per_second": 13.903,
+      "eval_steps_per_second": 2.317,
+      "step": 570
+    },
+    {
+      "epoch": 52.27,
+      "learning_rate": 7e-06,
+      "loss": 1.7498,
+      "step": 575
+    },
+    {
+      "epoch": 52.27,
+      "eval_loss": 2.531799077987671,
+      "eval_runtime": 0.4158,
+      "eval_samples_per_second": 14.431,
+      "eval_steps_per_second": 2.405,
+      "step": 575
+    },
+    {
+      "epoch": 52.73,
+      "learning_rate": 7e-06,
+      "loss": 1.7879,
+      "step": 580
+    },
+    {
+      "epoch": 52.73,
+      "eval_loss": 2.5400400161743164,
+      "eval_runtime": 0.4282,
+      "eval_samples_per_second": 14.011,
+      "eval_steps_per_second": 2.335,
+      "step": 580
+    },
+    {
+      "epoch": 53.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8234,
+      "step": 585
+    },
+    {
+      "epoch": 53.18,
+      "eval_loss": 2.5378053188323975,
+      "eval_runtime": 0.4216,
+      "eval_samples_per_second": 14.231,
+      "eval_steps_per_second": 2.372,
+      "step": 585
+    },
+    {
+      "epoch": 53.64,
+      "learning_rate": 7e-06,
+      "loss": 1.7653,
+      "step": 590
+    },
+    {
+      "epoch": 53.64,
+      "eval_loss": 2.541457176208496,
+      "eval_runtime": 0.4186,
+      "eval_samples_per_second": 14.335,
+      "eval_steps_per_second": 2.389,
+      "step": 590
+    },
+    {
+      "epoch": 54.09,
+      "learning_rate": 7e-06,
+      "loss": 1.7322,
+      "step": 595
+    },
+    {
+      "epoch": 54.09,
+      "eval_loss": 2.5493710041046143,
+      "eval_runtime": 0.411,
+      "eval_samples_per_second": 14.598,
+      "eval_steps_per_second": 2.433,
+      "step": 595
+    },
+    {
+      "epoch": 54.55,
+      "learning_rate": 7e-06,
+      "loss": 1.6075,
+      "step": 600
+    },
+    {
+      "epoch": 54.55,
+      "eval_loss": 2.5551865100860596,
+      "eval_runtime": 0.4082,
+      "eval_samples_per_second": 14.699,
+      "eval_steps_per_second": 2.45,
+      "step": 600
+    },
+    {
+      "epoch": 55.0,
+      "learning_rate": 7e-06,
+      "loss": 1.8756,
+      "step": 605
+    },
+    {
+      "epoch": 55.0,
+      "eval_loss": 2.563950300216675,
+      "eval_runtime": 0.4162,
+      "eval_samples_per_second": 14.417,
+      "eval_steps_per_second": 2.403,
+      "step": 605
+    },
+    {
+      "epoch": 55.45,
+      "learning_rate": 7e-06,
+      "loss": 1.6549,
+      "step": 610
+    },
+    {
+      "epoch": 55.45,
+      "eval_loss": 2.567089080810547,
+      "eval_runtime": 0.4089,
+      "eval_samples_per_second": 14.673,
+      "eval_steps_per_second": 2.445,
+      "step": 610
+    },
+    {
+      "epoch": 55.91,
+      "learning_rate": 7e-06,
+      "loss": 1.7606,
+      "step": 615
+    },
+    {
+      "epoch": 55.91,
+      "eval_loss": 2.5678203105926514,
+      "eval_runtime": 0.4075,
+      "eval_samples_per_second": 14.723,
+      "eval_steps_per_second": 2.454,
+      "step": 615
+    },
+    {
+      "epoch": 56.36,
+      "learning_rate": 7e-06,
+      "loss": 1.6907,
+      "step": 620
+    },
+    {
+      "epoch": 56.36,
+      "eval_loss": 2.5747501850128174,
+      "eval_runtime": 0.4086,
+      "eval_samples_per_second": 14.683,
+      "eval_steps_per_second": 2.447,
+      "step": 620
+    },
+    {
+      "epoch": 56.82,
+      "learning_rate": 7e-06,
+      "loss": 1.7504,
+      "step": 625
+    },
+    {
+      "epoch": 56.82,
+      "eval_loss": 2.5831196308135986,
+      "eval_runtime": 0.41,
+      "eval_samples_per_second": 14.635,
+      "eval_steps_per_second": 2.439,
+      "step": 625
+    },
+    {
+      "epoch": 57.27,
+      "learning_rate": 7e-06,
+      "loss": 1.6549,
+      "step": 630
+    },
+    {
+      "epoch": 57.27,
+      "eval_loss": 2.5876691341400146,
+      "eval_runtime": 0.4183,
+      "eval_samples_per_second": 14.343,
+      "eval_steps_per_second": 2.391,
+      "step": 630
+    },
+    {
+      "epoch": 57.73,
+      "learning_rate": 7e-06,
+      "loss": 1.694,
+      "step": 635
+    },
+    {
+      "epoch": 57.73,
+      "eval_loss": 2.5929791927337646,
+      "eval_runtime": 0.4201,
+      "eval_samples_per_second": 14.283,
+      "eval_steps_per_second": 2.38,
+      "step": 635
+    },
+    {
+      "epoch": 58.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8248,
+      "step": 640
+    },
+    {
+      "epoch": 58.18,
+      "eval_loss": 2.5961735248565674,
+      "eval_runtime": 0.4089,
+      "eval_samples_per_second": 14.675,
+      "eval_steps_per_second": 2.446,
+      "step": 640
+    },
+    {
+      "epoch": 58.64,
+      "learning_rate": 7e-06,
+      "loss": 1.4473,
+      "step": 645
+    },
+    {
+      "epoch": 58.64,
+      "eval_loss": 2.5989134311676025,
+      "eval_runtime": 0.4215,
+      "eval_samples_per_second": 14.237,
+      "eval_steps_per_second": 2.373,
+      "step": 645
+    },
+    {
+      "epoch": 59.09,
+      "learning_rate": 7e-06,
+      "loss": 1.7458,
+      "step": 650
+    },
+    {
+      "epoch": 59.09,
+      "eval_loss": 2.615797758102417,
+      "eval_runtime": 0.4163,
+      "eval_samples_per_second": 14.414,
+      "eval_steps_per_second": 2.402,
+      "step": 650
+    },
+    {
+      "epoch": 59.55,
+      "learning_rate": 7e-06,
+      "loss": 1.5373,
+      "step": 655
+    },
+    {
+      "epoch": 59.55,
+      "eval_loss": 2.6261303424835205,
+      "eval_runtime": 0.4089,
+      "eval_samples_per_second": 14.675,
+      "eval_steps_per_second": 2.446,
+      "step": 655
+    },
+    {
+      "epoch": 60.0,
+      "learning_rate": 7e-06,
+      "loss": 1.7489,
+      "step": 660
+    },
+    {
+      "epoch": 60.0,
+      "eval_loss": 2.61822509765625,
+      "eval_runtime": 0.4279,
+      "eval_samples_per_second": 14.022,
+      "eval_steps_per_second": 2.337,
+      "step": 660
+    },
+    {
+      "epoch": 60.45,
+      "learning_rate": 7e-06,
+      "loss": 1.6411,
+      "step": 665
+    },
+    {
+      "epoch": 60.45,
+      "eval_loss": 2.618873357772827,
+      "eval_runtime": 0.409,
+      "eval_samples_per_second": 14.668,
+      "eval_steps_per_second": 2.445,
+      "step": 665
+    },
+    {
+      "epoch": 60.91,
+      "learning_rate": 7e-06,
+      "loss": 1.6075,
+      "step": 670
+    },
+    {
+      "epoch": 60.91,
+      "eval_loss": 2.6396188735961914,
+      "eval_runtime": 0.4237,
+      "eval_samples_per_second": 14.16,
+      "eval_steps_per_second": 2.36,
+      "step": 670
+    },
+    {
+      "epoch": 61.36,
+      "learning_rate": 7e-06,
+      "loss": 1.7576,
+      "step": 675
+    },
+    {
+      "epoch": 61.36,
+      "eval_loss": 2.644122362136841,
+      "eval_runtime": 0.4237,
+      "eval_samples_per_second": 14.163,
+      "eval_steps_per_second": 2.36,
+      "step": 675
+    },
+    {
+      "epoch": 61.82,
+      "learning_rate": 7e-06,
+      "loss": 1.5172,
+      "step": 680
+    },
+    {
+      "epoch": 61.82,
+      "eval_loss": 2.6421959400177,
+      "eval_runtime": 0.4428,
+      "eval_samples_per_second": 13.549,
+      "eval_steps_per_second": 2.258,
+      "step": 680
+    },
+    {
+      "epoch": 62.27,
+      "learning_rate": 7e-06,
+      "loss": 1.4829,
+      "step": 685
+    },
+    {
+      "epoch": 62.27,
+      "eval_loss": 2.654413938522339,
+      "eval_runtime": 0.4274,
+      "eval_samples_per_second": 14.037,
+      "eval_steps_per_second": 2.34,
+      "step": 685
+    },
+    {
+      "epoch": 62.73,
+      "learning_rate": 7e-06,
+      "loss": 1.4743,
+      "step": 690
+    },
+    {
+      "epoch": 62.73,
+      "eval_loss": 2.6696255207061768,
+      "eval_runtime": 0.4519,
+      "eval_samples_per_second": 13.278,
+      "eval_steps_per_second": 2.213,
+      "step": 690
+    },
+    {
+      "epoch": 63.18,
+      "learning_rate": 7e-06,
+      "loss": 1.7067,
+      "step": 695
+    },
+    {
+      "epoch": 63.18,
+      "eval_loss": 2.6701173782348633,
+      "eval_runtime": 0.4269,
+      "eval_samples_per_second": 14.055,
+      "eval_steps_per_second": 2.343,
+      "step": 695
+    },
+    {
+      "epoch": 63.64,
+      "learning_rate": 7e-06,
+      "loss": 1.5196,
+      "step": 700
+    },
+    {
+      "epoch": 63.64,
+      "eval_loss": 2.6708524227142334,
+      "eval_runtime": 0.4296,
+      "eval_samples_per_second": 13.965,
+      "eval_steps_per_second": 2.328,
+      "step": 700
+    }
+  ],
+  "max_steps": 1100,
+  "num_train_epochs": 100,
+  "total_flos": 212855240663040.0,
+  "trial_name": null,
+  "trial_params": null
+}

stellaris-races-2/checkpoint-700/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26d2b678a4d9f285f917a1b9046a7a81b7861eec33925ea0244852fdce629f6f
+size 3963

stellaris-races-2/checkpoint-750/adapter_config.json

@@ -0,0 +1,17 @@
+{
+  "base_model_name_or_path": "",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

stellaris-races-2/checkpoint-750/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:917a8e22f0462fc04622885055061e9e45fb9efdfa5227b4bd8856df15eee137
+size 67154893

stellaris-races-2/checkpoint-750/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac876cac4e803309da058a5e0fa7e4a5d8c89b2a1c5dd9914f7ba577beb02a4f
+size 134325445

stellaris-races-2/checkpoint-750/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef70c26415e1f430bad66c337251accead6ee9902f6b3b691392d4f130cc090d
+size 14575

stellaris-races-2/checkpoint-750/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b43bda764475f9805d3316d95bc72cdd96cbed30e8825ff9d3627a7a91cc121
+size 627

stellaris-races-2/checkpoint-750/trainer_state.json

@@ -0,0 +1,2116 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 68.18181818181819,
+  "global_step": 750,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.45,
+      "learning_rate": 7e-06,
+      "loss": 3.0049,
+      "step": 5
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 3.0305967330932617,
+      "eval_runtime": 0.4174,
+      "eval_samples_per_second": 14.374,
+      "eval_steps_per_second": 2.396,
+      "step": 5
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 7e-06,
+      "loss": 3.0817,
+      "step": 10
+    },
+    {
+      "epoch": 0.91,
+      "eval_loss": 3.026319742202759,
+      "eval_runtime": 0.44,
+      "eval_samples_per_second": 13.636,
+      "eval_steps_per_second": 2.273,
+      "step": 10
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 7e-06,
+      "loss": 3.0438,
+      "step": 15
+    },
+    {
+      "epoch": 1.36,
+      "eval_loss": 3.022075891494751,
+      "eval_runtime": 0.4273,
+      "eval_samples_per_second": 14.04,
+      "eval_steps_per_second": 2.34,
+      "step": 15
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7e-06,
+      "loss": 2.9792,
+      "step": 20
+    },
+    {
+      "epoch": 1.82,
+      "eval_loss": 3.016573667526245,
+      "eval_runtime": 0.4072,
+      "eval_samples_per_second": 14.733,
+      "eval_steps_per_second": 2.456,
+      "step": 20
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 7e-06,
+      "loss": 2.934,
+      "step": 25
+    },
+    {
+      "epoch": 2.27,
+      "eval_loss": 3.0110604763031006,
+      "eval_runtime": 0.4045,
+      "eval_samples_per_second": 14.832,
+      "eval_steps_per_second": 2.472,
+      "step": 25
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 7e-06,
+      "loss": 3.1707,
+      "step": 30
+    },
+    {
+      "epoch": 2.73,
+      "eval_loss": 3.004523277282715,
+      "eval_runtime": 0.408,
+      "eval_samples_per_second": 14.706,
+      "eval_steps_per_second": 2.451,
+      "step": 30
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 7e-06,
+      "loss": 2.9016,
+      "step": 35
+    },
+    {
+      "epoch": 3.18,
+      "eval_loss": 2.997664213180542,
+      "eval_runtime": 0.4355,
+      "eval_samples_per_second": 13.778,
+      "eval_steps_per_second": 2.296,
+      "step": 35
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 7e-06,
+      "loss": 3.0052,
+      "step": 40
+    },
+    {
+      "epoch": 3.64,
+      "eval_loss": 2.990278482437134,
+      "eval_runtime": 0.4213,
+      "eval_samples_per_second": 14.243,
+      "eval_steps_per_second": 2.374,
+      "step": 40
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 7e-06,
+      "loss": 3.0642,
+      "step": 45
+    },
+    {
+      "epoch": 4.09,
+      "eval_loss": 2.982116937637329,
+      "eval_runtime": 0.4131,
+      "eval_samples_per_second": 14.525,
+      "eval_steps_per_second": 2.421,
+      "step": 45
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 7e-06,
+      "loss": 2.8696,
+      "step": 50
+    },
+    {
+      "epoch": 4.55,
+      "eval_loss": 2.973396062850952,
+      "eval_runtime": 0.4071,
+      "eval_samples_per_second": 14.737,
+      "eval_steps_per_second": 2.456,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 7e-06,
+      "loss": 3.0592,
+      "step": 55
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.964240312576294,
+      "eval_runtime": 0.4248,
+      "eval_samples_per_second": 14.126,
+      "eval_steps_per_second": 2.354,
+      "step": 55
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 7e-06,
+      "loss": 2.9686,
+      "step": 60
+    },
+    {
+      "epoch": 5.45,
+      "eval_loss": 2.9537107944488525,
+      "eval_runtime": 0.434,
+      "eval_samples_per_second": 13.825,
+      "eval_steps_per_second": 2.304,
+      "step": 60
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 7e-06,
+      "loss": 2.92,
+      "step": 65
+    },
+    {
+      "epoch": 5.91,
+      "eval_loss": 2.942852258682251,
+      "eval_runtime": 0.4106,
+      "eval_samples_per_second": 14.614,
+      "eval_steps_per_second": 2.436,
+      "step": 65
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 7e-06,
+      "loss": 2.9857,
+      "step": 70
+    },
+    {
+      "epoch": 6.36,
+      "eval_loss": 2.9314286708831787,
+      "eval_runtime": 0.4051,
+      "eval_samples_per_second": 14.812,
+      "eval_steps_per_second": 2.469,
+      "step": 70
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 7e-06,
+      "loss": 2.864,
+      "step": 75
+    },
+    {
+      "epoch": 6.82,
+      "eval_loss": 2.9195594787597656,
+      "eval_runtime": 0.4379,
+      "eval_samples_per_second": 13.701,
+      "eval_steps_per_second": 2.284,
+      "step": 75
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 7e-06,
+      "loss": 2.8973,
+      "step": 80
+    },
+    {
+      "epoch": 7.27,
+      "eval_loss": 2.906179189682007,
+      "eval_runtime": 0.4044,
+      "eval_samples_per_second": 14.837,
+      "eval_steps_per_second": 2.473,
+      "step": 80
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 7e-06,
+      "loss": 2.9251,
+      "step": 85
+    },
+    {
+      "epoch": 7.73,
+      "eval_loss": 2.892069101333618,
+      "eval_runtime": 0.4132,
+      "eval_samples_per_second": 14.519,
+      "eval_steps_per_second": 2.42,
+      "step": 85
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 7e-06,
+      "loss": 2.8072,
+      "step": 90
+    },
+    {
+      "epoch": 8.18,
+      "eval_loss": 2.8766021728515625,
+      "eval_runtime": 0.4169,
+      "eval_samples_per_second": 14.391,
+      "eval_steps_per_second": 2.399,
+      "step": 90
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 7e-06,
+      "loss": 2.7725,
+      "step": 95
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 2.8613014221191406,
+      "eval_runtime": 0.4218,
+      "eval_samples_per_second": 14.223,
+      "eval_steps_per_second": 2.371,
+      "step": 95
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 7e-06,
+      "loss": 2.8727,
+      "step": 100
+    },
+    {
+      "epoch": 9.09,
+      "eval_loss": 2.845301866531372,
+      "eval_runtime": 0.4068,
+      "eval_samples_per_second": 14.748,
+      "eval_steps_per_second": 2.458,
+      "step": 100
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 7e-06,
+      "loss": 2.8022,
+      "step": 105
+    },
+    {
+      "epoch": 9.55,
+      "eval_loss": 2.8308088779449463,
+      "eval_runtime": 0.405,
+      "eval_samples_per_second": 14.814,
+      "eval_steps_per_second": 2.469,
+      "step": 105
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7e-06,
+      "loss": 2.8153,
+      "step": 110
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.8145711421966553,
+      "eval_runtime": 0.4061,
+      "eval_samples_per_second": 14.775,
+      "eval_steps_per_second": 2.463,
+      "step": 110
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 7e-06,
+      "loss": 2.8341,
+      "step": 115
+    },
+    {
+      "epoch": 10.45,
+      "eval_loss": 2.7991654872894287,
+      "eval_runtime": 0.4419,
+      "eval_samples_per_second": 13.577,
+      "eval_steps_per_second": 2.263,
+      "step": 115
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 7e-06,
+      "loss": 2.5969,
+      "step": 120
+    },
+    {
+      "epoch": 10.91,
+      "eval_loss": 2.783989906311035,
+      "eval_runtime": 0.4292,
+      "eval_samples_per_second": 13.98,
+      "eval_steps_per_second": 2.33,
+      "step": 120
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 7e-06,
+      "loss": 2.8274,
+      "step": 125
+    },
+    {
+      "epoch": 11.36,
+      "eval_loss": 2.7690820693969727,
+      "eval_runtime": 0.4123,
+      "eval_samples_per_second": 14.554,
+      "eval_steps_per_second": 2.426,
+      "step": 125
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 7e-06,
+      "loss": 2.7421,
+      "step": 130
+    },
+    {
+      "epoch": 11.82,
+      "eval_loss": 2.755321741104126,
+      "eval_runtime": 0.4081,
+      "eval_samples_per_second": 14.703,
+      "eval_steps_per_second": 2.451,
+      "step": 130
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 7e-06,
+      "loss": 2.635,
+      "step": 135
+    },
+    {
+      "epoch": 12.27,
+      "eval_loss": 2.742081642150879,
+      "eval_runtime": 0.4061,
+      "eval_samples_per_second": 14.774,
+      "eval_steps_per_second": 2.462,
+      "step": 135
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 7e-06,
+      "loss": 2.7192,
+      "step": 140
+    },
+    {
+      "epoch": 12.73,
+      "eval_loss": 2.729800224304199,
+      "eval_runtime": 0.4106,
+      "eval_samples_per_second": 14.613,
+      "eval_steps_per_second": 2.436,
+      "step": 140
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 7e-06,
+      "loss": 2.5953,
+      "step": 145
+    },
+    {
+      "epoch": 13.18,
+      "eval_loss": 2.7182958126068115,
+      "eval_runtime": 0.4151,
+      "eval_samples_per_second": 14.454,
+      "eval_steps_per_second": 2.409,
+      "step": 145
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 7e-06,
+      "loss": 2.687,
+      "step": 150
+    },
+    {
+      "epoch": 13.64,
+      "eval_loss": 2.7074530124664307,
+      "eval_runtime": 0.4084,
+      "eval_samples_per_second": 14.692,
+      "eval_steps_per_second": 2.449,
+      "step": 150
+    },
+    {
+      "epoch": 14.09,
+      "learning_rate": 7e-06,
+      "loss": 2.6218,
+      "step": 155
+    },
+    {
+      "epoch": 14.09,
+      "eval_loss": 2.6966331005096436,
+      "eval_runtime": 0.4099,
+      "eval_samples_per_second": 14.638,
+      "eval_steps_per_second": 2.44,
+      "step": 155
+    },
+    {
+      "epoch": 14.55,
+      "learning_rate": 7e-06,
+      "loss": 2.56,
+      "step": 160
+    },
+    {
+      "epoch": 14.55,
+      "eval_loss": 2.685837745666504,
+      "eval_runtime": 0.4082,
+      "eval_samples_per_second": 14.7,
+      "eval_steps_per_second": 2.45,
+      "step": 160
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 7e-06,
+      "loss": 2.6864,
+      "step": 165
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.675248384475708,
+      "eval_runtime": 0.4084,
+      "eval_samples_per_second": 14.69,
+      "eval_steps_per_second": 2.448,
+      "step": 165
+    },
+    {
+      "epoch": 15.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4397,
+      "step": 170
+    },
+    {
+      "epoch": 15.45,
+      "eval_loss": 2.6641385555267334,
+      "eval_runtime": 0.4077,
+      "eval_samples_per_second": 14.718,
+      "eval_steps_per_second": 2.453,
+      "step": 170
+    },
+    {
+      "epoch": 15.91,
+      "learning_rate": 7e-06,
+      "loss": 2.709,
+      "step": 175
+    },
+    {
+      "epoch": 15.91,
+      "eval_loss": 2.652723550796509,
+      "eval_runtime": 0.4138,
+      "eval_samples_per_second": 14.498,
+      "eval_steps_per_second": 2.416,
+      "step": 175
+    },
+    {
+      "epoch": 16.36,
+      "learning_rate": 7e-06,
+      "loss": 2.5046,
+      "step": 180
+    },
+    {
+      "epoch": 16.36,
+      "eval_loss": 2.6419966220855713,
+      "eval_runtime": 0.4086,
+      "eval_samples_per_second": 14.684,
+      "eval_steps_per_second": 2.447,
+      "step": 180
+    },
+    {
+      "epoch": 16.82,
+      "learning_rate": 7e-06,
+      "loss": 2.5968,
+      "step": 185
+    },
+    {
+      "epoch": 16.82,
+      "eval_loss": 2.6306352615356445,
+      "eval_runtime": 0.4096,
+      "eval_samples_per_second": 14.647,
+      "eval_steps_per_second": 2.441,
+      "step": 185
+    },
+    {
+      "epoch": 17.27,
+      "learning_rate": 7e-06,
+      "loss": 2.6083,
+      "step": 190
+    },
+    {
+      "epoch": 17.27,
+      "eval_loss": 2.619220733642578,
+      "eval_runtime": 0.4137,
+      "eval_samples_per_second": 14.502,
+      "eval_steps_per_second": 2.417,
+      "step": 190
+    },
+    {
+      "epoch": 17.73,
+      "learning_rate": 7e-06,
+      "loss": 2.51,
+      "step": 195
+    },
+    {
+      "epoch": 17.73,
+      "eval_loss": 2.6068971157073975,
+      "eval_runtime": 0.4084,
+      "eval_samples_per_second": 14.693,
+      "eval_steps_per_second": 2.449,
+      "step": 195
+    },
+    {
+      "epoch": 18.18,
+      "learning_rate": 7e-06,
+      "loss": 2.5301,
+      "step": 200
+    },
+    {
+      "epoch": 18.18,
+      "eval_loss": 2.5943899154663086,
+      "eval_runtime": 0.4198,
+      "eval_samples_per_second": 14.291,
+      "eval_steps_per_second": 2.382,
+      "step": 200
+    },
+    {
+      "epoch": 18.64,
+      "learning_rate": 7e-06,
+      "loss": 2.5177,
+      "step": 205
+    },
+    {
+      "epoch": 18.64,
+      "eval_loss": 2.58101487159729,
+      "eval_runtime": 0.4131,
+      "eval_samples_per_second": 14.523,
+      "eval_steps_per_second": 2.421,
+      "step": 205
+    },
+    {
+      "epoch": 19.09,
+      "learning_rate": 7e-06,
+      "loss": 2.3431,
+      "step": 210
+    },
+    {
+      "epoch": 19.09,
+      "eval_loss": 2.568328619003296,
+      "eval_runtime": 0.4161,
+      "eval_samples_per_second": 14.42,
+      "eval_steps_per_second": 2.403,
+      "step": 210
+    },
+    {
+      "epoch": 19.55,
+      "learning_rate": 7e-06,
+      "loss": 2.358,
+      "step": 215
+    },
+    {
+      "epoch": 19.55,
+      "eval_loss": 2.556185007095337,
+      "eval_runtime": 0.4167,
+      "eval_samples_per_second": 14.399,
+      "eval_steps_per_second": 2.4,
+      "step": 215
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 7e-06,
+      "loss": 2.4723,
+      "step": 220
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 2.54498553276062,
+      "eval_runtime": 0.4187,
+      "eval_samples_per_second": 14.331,
+      "eval_steps_per_second": 2.388,
+      "step": 220
+    },
+    {
+      "epoch": 20.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4351,
+      "step": 225
+    },
+    {
+      "epoch": 20.45,
+      "eval_loss": 2.5338480472564697,
+      "eval_runtime": 0.4148,
+      "eval_samples_per_second": 14.463,
+      "eval_steps_per_second": 2.411,
+      "step": 225
+    },
+    {
+      "epoch": 20.91,
+      "learning_rate": 7e-06,
+      "loss": 2.3407,
+      "step": 230
+    },
+    {
+      "epoch": 20.91,
+      "eval_loss": 2.522146463394165,
+      "eval_runtime": 0.4166,
+      "eval_samples_per_second": 14.401,
+      "eval_steps_per_second": 2.4,
+      "step": 230
+    },
+    {
+      "epoch": 21.36,
+      "learning_rate": 7e-06,
+      "loss": 2.4691,
+      "step": 235
+    },
+    {
+      "epoch": 21.36,
+      "eval_loss": 2.5105435848236084,
+      "eval_runtime": 0.4117,
+      "eval_samples_per_second": 14.572,
+      "eval_steps_per_second": 2.429,
+      "step": 235
+    },
+    {
+      "epoch": 21.82,
+      "learning_rate": 7e-06,
+      "loss": 2.3515,
+      "step": 240
+    },
+    {
+      "epoch": 21.82,
+      "eval_loss": 2.498257637023926,
+      "eval_runtime": 0.4513,
+      "eval_samples_per_second": 13.296,
+      "eval_steps_per_second": 2.216,
+      "step": 240
+    },
+    {
+      "epoch": 22.27,
+      "learning_rate": 7e-06,
+      "loss": 2.2231,
+      "step": 245
+    },
+    {
+      "epoch": 22.27,
+      "eval_loss": 2.4853975772857666,
+      "eval_runtime": 0.4211,
+      "eval_samples_per_second": 14.247,
+      "eval_steps_per_second": 2.375,
+      "step": 245
+    },
+    {
+      "epoch": 22.73,
+      "learning_rate": 7e-06,
+      "loss": 2.2849,
+      "step": 250
+    },
+    {
+      "epoch": 22.73,
+      "eval_loss": 2.474109411239624,
+      "eval_runtime": 0.422,
+      "eval_samples_per_second": 14.218,
+      "eval_steps_per_second": 2.37,
+      "step": 250
+    },
+    {
+      "epoch": 23.18,
+      "learning_rate": 7e-06,
+      "loss": 2.4086,
+      "step": 255
+    },
+    {
+      "epoch": 23.18,
+      "eval_loss": 2.4653780460357666,
+      "eval_runtime": 0.4081,
+      "eval_samples_per_second": 14.702,
+      "eval_steps_per_second": 2.45,
+      "step": 255
+    },
+    {
+      "epoch": 23.64,
+      "learning_rate": 7e-06,
+      "loss": 2.1498,
+      "step": 260
+    },
+    {
+      "epoch": 23.64,
+      "eval_loss": 2.4591386318206787,
+      "eval_runtime": 0.4098,
+      "eval_samples_per_second": 14.64,
+      "eval_steps_per_second": 2.44,
+      "step": 260
+    },
+    {
+      "epoch": 24.09,
+      "learning_rate": 7e-06,
+      "loss": 2.3414,
+      "step": 265
+    },
+    {
+      "epoch": 24.09,
+      "eval_loss": 2.454268455505371,
+      "eval_runtime": 0.4114,
+      "eval_samples_per_second": 14.586,
+      "eval_steps_per_second": 2.431,
+      "step": 265
+    },
+    {
+      "epoch": 24.55,
+      "learning_rate": 7e-06,
+      "loss": 2.1754,
+      "step": 270
+    },
+    {
+      "epoch": 24.55,
+      "eval_loss": 2.4511959552764893,
+      "eval_runtime": 0.4144,
+      "eval_samples_per_second": 14.479,
+      "eval_steps_per_second": 2.413,
+      "step": 270
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3577,
+      "step": 275
+    },
+    {
+      "epoch": 25.0,
+      "eval_loss": 2.4483206272125244,
+      "eval_runtime": 0.4191,
+      "eval_samples_per_second": 14.316,
+      "eval_steps_per_second": 2.386,
+      "step": 275
+    },
+    {
+      "epoch": 25.45,
+      "learning_rate": 7e-06,
+      "loss": 2.1878,
+      "step": 280
+    },
+    {
+      "epoch": 25.45,
+      "eval_loss": 2.4461658000946045,
+      "eval_runtime": 0.4103,
+      "eval_samples_per_second": 14.624,
+      "eval_steps_per_second": 2.437,
+      "step": 280
+    },
+    {
+      "epoch": 25.91,
+      "learning_rate": 7e-06,
+      "loss": 2.3283,
+      "step": 285
+    },
+    {
+      "epoch": 25.91,
+      "eval_loss": 2.4441139698028564,
+      "eval_runtime": 0.4091,
+      "eval_samples_per_second": 14.667,
+      "eval_steps_per_second": 2.444,
+      "step": 285
+    },
+    {
+      "epoch": 26.36,
+      "learning_rate": 7e-06,
+      "loss": 2.2045,
+      "step": 290
+    },
+    {
+      "epoch": 26.36,
+      "eval_loss": 2.44203782081604,
+      "eval_runtime": 0.4117,
+      "eval_samples_per_second": 14.574,
+      "eval_steps_per_second": 2.429,
+      "step": 290
+    },
+    {
+      "epoch": 26.82,
+      "learning_rate": 7e-06,
+      "loss": 2.2802,
+      "step": 295
+    },
+    {
+      "epoch": 26.82,
+      "eval_loss": 2.4402430057525635,
+      "eval_runtime": 0.4202,
+      "eval_samples_per_second": 14.279,
+      "eval_steps_per_second": 2.38,
+      "step": 295
+    },
+    {
+      "epoch": 27.27,
+      "learning_rate": 7e-06,
+      "loss": 2.0865,
+      "step": 300
+    },
+    {
+      "epoch": 27.27,
+      "eval_loss": 2.438746690750122,
+      "eval_runtime": 0.4215,
+      "eval_samples_per_second": 14.235,
+      "eval_steps_per_second": 2.373,
+      "step": 300
+    },
+    {
+      "epoch": 27.73,
+      "learning_rate": 7e-06,
+      "loss": 2.275,
+      "step": 305
+    },
+    {
+      "epoch": 27.73,
+      "eval_loss": 2.4373252391815186,
+      "eval_runtime": 0.4173,
+      "eval_samples_per_second": 14.377,
+      "eval_steps_per_second": 2.396,
+      "step": 305
+    },
+    {
+      "epoch": 28.18,
+      "learning_rate": 7e-06,
+      "loss": 2.277,
+      "step": 310
+    },
+    {
+      "epoch": 28.18,
+      "eval_loss": 2.4359142780303955,
+      "eval_runtime": 0.4204,
+      "eval_samples_per_second": 14.273,
+      "eval_steps_per_second": 2.379,
+      "step": 310
+    },
+    {
+      "epoch": 28.64,
+      "learning_rate": 7e-06,
+      "loss": 2.2201,
+      "step": 315
+    },
+    {
+      "epoch": 28.64,
+      "eval_loss": 2.435053825378418,
+      "eval_runtime": 0.4099,
+      "eval_samples_per_second": 14.637,
+      "eval_steps_per_second": 2.44,
+      "step": 315
+    },
+    {
+      "epoch": 29.09,
+      "learning_rate": 7e-06,
+      "loss": 2.1064,
+      "step": 320
+    },
+    {
+      "epoch": 29.09,
+      "eval_loss": 2.4337093830108643,
+      "eval_runtime": 0.4202,
+      "eval_samples_per_second": 14.277,
+      "eval_steps_per_second": 2.38,
+      "step": 320
+    },
+    {
+      "epoch": 29.55,
+      "learning_rate": 7e-06,
+      "loss": 2.1089,
+      "step": 325
+    },
+    {
+      "epoch": 29.55,
+      "eval_loss": 2.4328536987304688,
+      "eval_runtime": 0.4206,
+      "eval_samples_per_second": 14.265,
+      "eval_steps_per_second": 2.377,
+      "step": 325
+    },
+    {
+      "epoch": 30.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3056,
+      "step": 330
+    },
+    {
+      "epoch": 30.0,
+      "eval_loss": 2.43233323097229,
+      "eval_runtime": 0.4239,
+      "eval_samples_per_second": 14.155,
+      "eval_steps_per_second": 2.359,
+      "step": 330
+    },
+    {
+      "epoch": 30.45,
+      "learning_rate": 7e-06,
+      "loss": 2.0761,
+      "step": 335
+    },
+    {
+      "epoch": 30.45,
+      "eval_loss": 2.431633949279785,
+      "eval_runtime": 0.4101,
+      "eval_samples_per_second": 14.631,
+      "eval_steps_per_second": 2.438,
+      "step": 335
+    },
+    {
+      "epoch": 30.91,
+      "learning_rate": 7e-06,
+      "loss": 2.2653,
+      "step": 340
+    },
+    {
+      "epoch": 30.91,
+      "eval_loss": 2.43125319480896,
+      "eval_runtime": 0.4102,
+      "eval_samples_per_second": 14.626,
+      "eval_steps_per_second": 2.438,
+      "step": 340
+    },
+    {
+      "epoch": 31.36,
+      "learning_rate": 7e-06,
+      "loss": 2.1188,
+      "step": 345
+    },
+    {
+      "epoch": 31.36,
+      "eval_loss": 2.430781126022339,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.715,
+      "eval_steps_per_second": 2.452,
+      "step": 345
+    },
+    {
+      "epoch": 31.82,
+      "learning_rate": 7e-06,
+      "loss": 2.23,
+      "step": 350
+    },
+    {
+      "epoch": 31.82,
+      "eval_loss": 2.4303319454193115,
+      "eval_runtime": 0.4174,
+      "eval_samples_per_second": 14.376,
+      "eval_steps_per_second": 2.396,
+      "step": 350
+    },
+    {
+      "epoch": 32.27,
+      "learning_rate": 7e-06,
+      "loss": 2.0428,
+      "step": 355
+    },
+    {
+      "epoch": 32.27,
+      "eval_loss": 2.4302732944488525,
+      "eval_runtime": 0.4196,
+      "eval_samples_per_second": 14.301,
+      "eval_steps_per_second": 2.383,
+      "step": 355
+    },
+    {
+      "epoch": 32.73,
+      "learning_rate": 7e-06,
+      "loss": 2.1743,
+      "step": 360
+    },
+    {
+      "epoch": 32.73,
+      "eval_loss": 2.4303529262542725,
+      "eval_runtime": 0.419,
+      "eval_samples_per_second": 14.32,
+      "eval_steps_per_second": 2.387,
+      "step": 360
+    },
+    {
+      "epoch": 33.18,
+      "learning_rate": 7e-06,
+      "loss": 2.2025,
+      "step": 365
+    },
+    {
+      "epoch": 33.18,
+      "eval_loss": 2.430536985397339,
+      "eval_runtime": 0.4276,
+      "eval_samples_per_second": 14.032,
+      "eval_steps_per_second": 2.339,
+      "step": 365
+    },
+    {
+      "epoch": 33.64,
+      "learning_rate": 7e-06,
+      "loss": 1.9553,
+      "step": 370
+    },
+    {
+      "epoch": 33.64,
+      "eval_loss": 2.4313228130340576,
+      "eval_runtime": 0.4148,
+      "eval_samples_per_second": 14.465,
+      "eval_steps_per_second": 2.411,
+      "step": 370
+    },
+    {
+      "epoch": 34.09,
+      "learning_rate": 7e-06,
+      "loss": 2.2841,
+      "step": 375
+    },
+    {
+      "epoch": 34.09,
+      "eval_loss": 2.431515693664551,
+      "eval_runtime": 0.4289,
+      "eval_samples_per_second": 13.989,
+      "eval_steps_per_second": 2.332,
+      "step": 375
+    },
+    {
+      "epoch": 34.55,
+      "learning_rate": 7e-06,
+      "loss": 2.0907,
+      "step": 380
+    },
+    {
+      "epoch": 34.55,
+      "eval_loss": 2.4317359924316406,
+      "eval_runtime": 0.4149,
+      "eval_samples_per_second": 14.462,
+      "eval_steps_per_second": 2.41,
+      "step": 380
+    },
+    {
+      "epoch": 35.0,
+      "learning_rate": 7e-06,
+      "loss": 2.1472,
+      "step": 385
+    },
+    {
+      "epoch": 35.0,
+      "eval_loss": 2.432173728942871,
+      "eval_runtime": 0.4348,
+      "eval_samples_per_second": 13.801,
+      "eval_steps_per_second": 2.3,
+      "step": 385
+    },
+    {
+      "epoch": 35.45,
+      "learning_rate": 7e-06,
+      "loss": 2.1642,
+      "step": 390
+    },
+    {
+      "epoch": 35.45,
+      "eval_loss": 2.433072090148926,
+      "eval_runtime": 0.4156,
+      "eval_samples_per_second": 14.438,
+      "eval_steps_per_second": 2.406,
+      "step": 390
+    },
+    {
+      "epoch": 35.91,
+      "learning_rate": 7e-06,
+      "loss": 2.0184,
+      "step": 395
+    },
+    {
+      "epoch": 35.91,
+      "eval_loss": 2.4334452152252197,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.728,
+      "eval_steps_per_second": 2.455,
+      "step": 395
+    },
+    {
+      "epoch": 36.36,
+      "learning_rate": 7e-06,
+      "loss": 2.1743,
+      "step": 400
+    },
+    {
+      "epoch": 36.36,
+      "eval_loss": 2.435070753097534,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.712,
+      "eval_steps_per_second": 2.452,
+      "step": 400
+    },
+    {
+      "epoch": 36.82,
+      "learning_rate": 7e-06,
+      "loss": 2.0438,
+      "step": 405
+    },
+    {
+      "epoch": 36.82,
+      "eval_loss": 2.436411142349243,
+      "eval_runtime": 0.4349,
+      "eval_samples_per_second": 13.797,
+      "eval_steps_per_second": 2.299,
+      "step": 405
+    },
+    {
+      "epoch": 37.27,
+      "learning_rate": 7e-06,
+      "loss": 2.1146,
+      "step": 410
+    },
+    {
+      "epoch": 37.27,
+      "eval_loss": 2.4375665187835693,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.729,
+      "eval_steps_per_second": 2.455,
+      "step": 410
+    },
+    {
+      "epoch": 37.73,
+      "learning_rate": 7e-06,
+      "loss": 2.0405,
+      "step": 415
+    },
+    {
+      "epoch": 37.73,
+      "eval_loss": 2.4385483264923096,
+      "eval_runtime": 0.4305,
+      "eval_samples_per_second": 13.938,
+      "eval_steps_per_second": 2.323,
+      "step": 415
+    },
+    {
+      "epoch": 38.18,
+      "learning_rate": 7e-06,
+      "loss": 1.9226,
+      "step": 420
+    },
+    {
+      "epoch": 38.18,
+      "eval_loss": 2.4396464824676514,
+      "eval_runtime": 0.4116,
+      "eval_samples_per_second": 14.579,
+      "eval_steps_per_second": 2.43,
+      "step": 420
+    },
+    {
+      "epoch": 38.64,
+      "learning_rate": 7e-06,
+      "loss": 2.1808,
+      "step": 425
+    },
+    {
+      "epoch": 38.64,
+      "eval_loss": 2.441523790359497,
+      "eval_runtime": 0.4292,
+      "eval_samples_per_second": 13.98,
+      "eval_steps_per_second": 2.33,
+      "step": 425
+    },
+    {
+      "epoch": 39.09,
+      "learning_rate": 7e-06,
+      "loss": 1.9993,
+      "step": 430
+    },
+    {
+      "epoch": 39.09,
+      "eval_loss": 2.4434759616851807,
+      "eval_runtime": 0.4337,
+      "eval_samples_per_second": 13.836,
+      "eval_steps_per_second": 2.306,
+      "step": 430
+    },
+    {
+      "epoch": 39.55,
+      "learning_rate": 7e-06,
+      "loss": 2.0123,
+      "step": 435
+    },
+    {
+      "epoch": 39.55,
+      "eval_loss": 2.445014715194702,
+      "eval_runtime": 0.4082,
+      "eval_samples_per_second": 14.699,
+      "eval_steps_per_second": 2.45,
+      "step": 435
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 7e-06,
+      "loss": 2.0762,
+      "step": 440
+    },
+    {
+      "epoch": 40.0,
+      "eval_loss": 2.446848154067993,
+      "eval_runtime": 0.4152,
+      "eval_samples_per_second": 14.451,
+      "eval_steps_per_second": 2.408,
+      "step": 440
+    },
+    {
+      "epoch": 40.45,
+      "learning_rate": 7e-06,
+      "loss": 2.0079,
+      "step": 445
+    },
+    {
+      "epoch": 40.45,
+      "eval_loss": 2.449875831604004,
+      "eval_runtime": 0.4092,
+      "eval_samples_per_second": 14.663,
+      "eval_steps_per_second": 2.444,
+      "step": 445
+    },
+    {
+      "epoch": 40.91,
+      "learning_rate": 7e-06,
+      "loss": 2.0856,
+      "step": 450
+    },
+    {
+      "epoch": 40.91,
+      "eval_loss": 2.4520113468170166,
+      "eval_runtime": 0.4073,
+      "eval_samples_per_second": 14.73,
+      "eval_steps_per_second": 2.455,
+      "step": 450
+    },
+    {
+      "epoch": 41.36,
+      "learning_rate": 7e-06,
+      "loss": 1.8473,
+      "step": 455
+    },
+    {
+      "epoch": 41.36,
+      "eval_loss": 2.45548415184021,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.728,
+      "eval_steps_per_second": 2.455,
+      "step": 455
+    },
+    {
+      "epoch": 41.82,
+      "learning_rate": 7e-06,
+      "loss": 2.1588,
+      "step": 460
+    },
+    {
+      "epoch": 41.82,
+      "eval_loss": 2.458113431930542,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.728,
+      "eval_steps_per_second": 2.455,
+      "step": 460
+    },
+    {
+      "epoch": 42.27,
+      "learning_rate": 7e-06,
+      "loss": 2.0635,
+      "step": 465
+    },
+    {
+      "epoch": 42.27,
+      "eval_loss": 2.457326650619507,
+      "eval_runtime": 0.4122,
+      "eval_samples_per_second": 14.556,
+      "eval_steps_per_second": 2.426,
+      "step": 465
+    },
+    {
+      "epoch": 42.73,
+      "learning_rate": 7e-06,
+      "loss": 1.8833,
+      "step": 470
+    },
+    {
+      "epoch": 42.73,
+      "eval_loss": 2.457960367202759,
+      "eval_runtime": 0.408,
+      "eval_samples_per_second": 14.707,
+      "eval_steps_per_second": 2.451,
+      "step": 470
+    },
+    {
+      "epoch": 43.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8618,
+      "step": 475
+    },
+    {
+      "epoch": 43.18,
+      "eval_loss": 2.4618539810180664,
+      "eval_runtime": 0.4074,
+      "eval_samples_per_second": 14.727,
+      "eval_steps_per_second": 2.455,
+      "step": 475
+    },
+    {
+      "epoch": 43.64,
+      "learning_rate": 7e-06,
+      "loss": 1.9884,
+      "step": 480
+    },
+    {
+      "epoch": 43.64,
+      "eval_loss": 2.4648358821868896,
+      "eval_runtime": 0.4066,
+      "eval_samples_per_second": 14.756,
+      "eval_steps_per_second": 2.459,
+      "step": 480
+    },
+    {
+      "epoch": 44.09,
+      "learning_rate": 7e-06,
+      "loss": 1.9292,
+      "step": 485
+    },
+    {
+      "epoch": 44.09,
+      "eval_loss": 2.467848777770996,
+      "eval_runtime": 0.4077,
+      "eval_samples_per_second": 14.718,
+      "eval_steps_per_second": 2.453,
+      "step": 485
+    },
+    {
+      "epoch": 44.55,
+      "learning_rate": 7e-06,
+      "loss": 1.8937,
+      "step": 490
+    },
+    {
+      "epoch": 44.55,
+      "eval_loss": 2.471271276473999,
+      "eval_runtime": 0.4098,
+      "eval_samples_per_second": 14.642,
+      "eval_steps_per_second": 2.44,
+      "step": 490
+    },
+    {
+      "epoch": 45.0,
+      "learning_rate": 7e-06,
+      "loss": 2.0734,
+      "step": 495
+    },
+    {
+      "epoch": 45.0,
+      "eval_loss": 2.4725749492645264,
+      "eval_runtime": 0.4143,
+      "eval_samples_per_second": 14.482,
+      "eval_steps_per_second": 2.414,
+      "step": 495
+    },
+    {
+      "epoch": 45.45,
+      "learning_rate": 7e-06,
+      "loss": 1.99,
+      "step": 500
+    },
+    {
+      "epoch": 45.45,
+      "eval_loss": 2.473742723464966,
+      "eval_runtime": 0.4224,
+      "eval_samples_per_second": 14.205,
+      "eval_steps_per_second": 2.367,
+      "step": 500
+    },
+    {
+      "epoch": 45.91,
+      "learning_rate": 7e-06,
+      "loss": 1.8188,
+      "step": 505
+    },
+    {
+      "epoch": 45.91,
+      "eval_loss": 2.4784109592437744,
+      "eval_runtime": 0.4294,
+      "eval_samples_per_second": 13.973,
+      "eval_steps_per_second": 2.329,
+      "step": 505
+    },
+    {
+      "epoch": 46.36,
+      "learning_rate": 7e-06,
+      "loss": 1.8354,
+      "step": 510
+    },
+    {
+      "epoch": 46.36,
+      "eval_loss": 2.485568046569824,
+      "eval_runtime": 0.4576,
+      "eval_samples_per_second": 13.112,
+      "eval_steps_per_second": 2.185,
+      "step": 510
+    },
+    {
+      "epoch": 46.82,
+      "learning_rate": 7e-06,
+      "loss": 1.9735,
+      "step": 515
+    },
+    {
+      "epoch": 46.82,
+      "eval_loss": 2.4888756275177,
+      "eval_runtime": 0.4219,
+      "eval_samples_per_second": 14.22,
+      "eval_steps_per_second": 2.37,
+      "step": 515
+    },
+    {
+      "epoch": 47.27,
+      "learning_rate": 7e-06,
+      "loss": 1.953,
+      "step": 520
+    },
+    {
+      "epoch": 47.27,
+      "eval_loss": 2.4856255054473877,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.712,
+      "eval_steps_per_second": 2.452,
+      "step": 520
+    },
+    {
+      "epoch": 47.73,
+      "learning_rate": 7e-06,
+      "loss": 1.8938,
+      "step": 525
+    },
+    {
+      "epoch": 47.73,
+      "eval_loss": 2.48818039894104,
+      "eval_runtime": 0.4106,
+      "eval_samples_per_second": 14.613,
+      "eval_steps_per_second": 2.436,
+      "step": 525
+    },
+    {
+      "epoch": 48.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8276,
+      "step": 530
+    },
+    {
+      "epoch": 48.18,
+      "eval_loss": 2.4947216510772705,
+      "eval_runtime": 0.4098,
+      "eval_samples_per_second": 14.642,
+      "eval_steps_per_second": 2.44,
+      "step": 530
+    },
+    {
+      "epoch": 48.64,
+      "learning_rate": 7e-06,
+      "loss": 1.9289,
+      "step": 535
+    },
+    {
+      "epoch": 48.64,
+      "eval_loss": 2.500380516052246,
+      "eval_runtime": 0.4688,
+      "eval_samples_per_second": 12.798,
+      "eval_steps_per_second": 2.133,
+      "step": 535
+    },
+    {
+      "epoch": 49.09,
+      "learning_rate": 7e-06,
+      "loss": 1.8077,
+      "step": 540
+    },
+    {
+      "epoch": 49.09,
+      "eval_loss": 2.50510573387146,
+      "eval_runtime": 0.4078,
+      "eval_samples_per_second": 14.712,
+      "eval_steps_per_second": 2.452,
+      "step": 540
+    },
+    {
+      "epoch": 49.55,
+      "learning_rate": 7e-06,
+      "loss": 1.9091,
+      "step": 545
+    },
+    {
+      "epoch": 49.55,
+      "eval_loss": 2.5079917907714844,
+      "eval_runtime": 0.4339,
+      "eval_samples_per_second": 13.829,
+      "eval_steps_per_second": 2.305,
+      "step": 545
+    },
+    {
+      "epoch": 50.0,
+      "learning_rate": 7e-06,
+      "loss": 1.8086,
+      "step": 550
+    },
+    {
+      "epoch": 50.0,
+      "eval_loss": 2.5064351558685303,
+      "eval_runtime": 0.4096,
+      "eval_samples_per_second": 14.648,
+      "eval_steps_per_second": 2.441,
+      "step": 550
+    },
+    {
+      "epoch": 50.45,
+      "learning_rate": 7e-06,
+      "loss": 1.8497,
+      "step": 555
+    },
+    {
+      "epoch": 50.45,
+      "eval_loss": 2.5103464126586914,
+      "eval_runtime": 0.4116,
+      "eval_samples_per_second": 14.577,
+      "eval_steps_per_second": 2.429,
+      "step": 555
+    },
+    {
+      "epoch": 50.91,
+      "learning_rate": 7e-06,
+      "loss": 1.8287,
+      "step": 560
+    },
+    {
+      "epoch": 50.91,
+      "eval_loss": 2.5173234939575195,
+      "eval_runtime": 0.4197,
+      "eval_samples_per_second": 14.296,
+      "eval_steps_per_second": 2.383,
+      "step": 560
+    },
+    {
+      "epoch": 51.36,
+      "learning_rate": 7e-06,
+      "loss": 1.8288,
+      "step": 565
+    },
+    {
+      "epoch": 51.36,
+      "eval_loss": 2.5230062007904053,
+      "eval_runtime": 0.4091,
+      "eval_samples_per_second": 14.665,
+      "eval_steps_per_second": 2.444,
+      "step": 565
+    },
+    {
+      "epoch": 51.82,
+      "learning_rate": 7e-06,
+      "loss": 1.803,
+      "step": 570
+    },
+    {
+      "epoch": 51.82,
+      "eval_loss": 2.52628493309021,
+      "eval_runtime": 0.4316,
+      "eval_samples_per_second": 13.903,
+      "eval_steps_per_second": 2.317,
+      "step": 570
+    },
+    {
+      "epoch": 52.27,
+      "learning_rate": 7e-06,
+      "loss": 1.7498,
+      "step": 575
+    },
+    {
+      "epoch": 52.27,
+      "eval_loss": 2.531799077987671,
+      "eval_runtime": 0.4158,
+      "eval_samples_per_second": 14.431,
+      "eval_steps_per_second": 2.405,
+      "step": 575
+    },
+    {
+      "epoch": 52.73,
+      "learning_rate": 7e-06,
+      "loss": 1.7879,
+      "step": 580
+    },
+    {
+      "epoch": 52.73,
+      "eval_loss": 2.5400400161743164,
+      "eval_runtime": 0.4282,
+      "eval_samples_per_second": 14.011,
+      "eval_steps_per_second": 2.335,
+      "step": 580
+    },
+    {
+      "epoch": 53.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8234,
+      "step": 585
+    },
+    {
+      "epoch": 53.18,
+      "eval_loss": 2.5378053188323975,
+      "eval_runtime": 0.4216,
+      "eval_samples_per_second": 14.231,
+      "eval_steps_per_second": 2.372,
+      "step": 585
+    },
+    {
+      "epoch": 53.64,
+      "learning_rate": 7e-06,
+      "loss": 1.7653,
+      "step": 590
+    },
+    {
+      "epoch": 53.64,
+      "eval_loss": 2.541457176208496,
+      "eval_runtime": 0.4186,
+      "eval_samples_per_second": 14.335,
+      "eval_steps_per_second": 2.389,
+      "step": 590
+    },
+    {
+      "epoch": 54.09,
+      "learning_rate": 7e-06,
+      "loss": 1.7322,
+      "step": 595
+    },
+    {
+      "epoch": 54.09,
+      "eval_loss": 2.5493710041046143,
+      "eval_runtime": 0.411,
+      "eval_samples_per_second": 14.598,
+      "eval_steps_per_second": 2.433,
+      "step": 595
+    },
+    {
+      "epoch": 54.55,
+      "learning_rate": 7e-06,
+      "loss": 1.6075,
+      "step": 600
+    },
+    {
+      "epoch": 54.55,
+      "eval_loss": 2.5551865100860596,
+      "eval_runtime": 0.4082,
+      "eval_samples_per_second": 14.699,
+      "eval_steps_per_second": 2.45,
+      "step": 600
+    },
+    {
+      "epoch": 55.0,
+      "learning_rate": 7e-06,
+      "loss": 1.8756,
+      "step": 605
+    },
+    {
+      "epoch": 55.0,
+      "eval_loss": 2.563950300216675,
+      "eval_runtime": 0.4162,
+      "eval_samples_per_second": 14.417,
+      "eval_steps_per_second": 2.403,
+      "step": 605
+    },
+    {
+      "epoch": 55.45,
+      "learning_rate": 7e-06,
+      "loss": 1.6549,
+      "step": 610
+    },
+    {
+      "epoch": 55.45,
+      "eval_loss": 2.567089080810547,
+      "eval_runtime": 0.4089,
+      "eval_samples_per_second": 14.673,
+      "eval_steps_per_second": 2.445,
+      "step": 610
+    },
+    {
+      "epoch": 55.91,
+      "learning_rate": 7e-06,
+      "loss": 1.7606,
+      "step": 615
+    },
+    {
+      "epoch": 55.91,
+      "eval_loss": 2.5678203105926514,
+      "eval_runtime": 0.4075,
+      "eval_samples_per_second": 14.723,
+      "eval_steps_per_second": 2.454,
+      "step": 615
+    },
+    {
+      "epoch": 56.36,
+      "learning_rate": 7e-06,
+      "loss": 1.6907,
+      "step": 620
+    },
+    {
+      "epoch": 56.36,
+      "eval_loss": 2.5747501850128174,
+      "eval_runtime": 0.4086,
+      "eval_samples_per_second": 14.683,
+      "eval_steps_per_second": 2.447,
+      "step": 620
+    },
+    {
+      "epoch": 56.82,
+      "learning_rate": 7e-06,
+      "loss": 1.7504,
+      "step": 625
+    },
+    {
+      "epoch": 56.82,
+      "eval_loss": 2.5831196308135986,
+      "eval_runtime": 0.41,
+      "eval_samples_per_second": 14.635,
+      "eval_steps_per_second": 2.439,
+      "step": 625
+    },
+    {
+      "epoch": 57.27,
+      "learning_rate": 7e-06,
+      "loss": 1.6549,
+      "step": 630
+    },
+    {
+      "epoch": 57.27,
+      "eval_loss": 2.5876691341400146,
+      "eval_runtime": 0.4183,
+      "eval_samples_per_second": 14.343,
+      "eval_steps_per_second": 2.391,
+      "step": 630
+    },
+    {
+      "epoch": 57.73,
+      "learning_rate": 7e-06,
+      "loss": 1.694,
+      "step": 635
+    },
+    {
+      "epoch": 57.73,
+      "eval_loss": 2.5929791927337646,
+      "eval_runtime": 0.4201,
+      "eval_samples_per_second": 14.283,
+      "eval_steps_per_second": 2.38,
+      "step": 635
+    },
+    {
+      "epoch": 58.18,
+      "learning_rate": 7e-06,
+      "loss": 1.8248,
+      "step": 640
+    },
+    {
+      "epoch": 58.18,
+      "eval_loss": 2.5961735248565674,
+      "eval_runtime": 0.4089,
+      "eval_samples_per_second": 14.675,
+      "eval_steps_per_second": 2.446,
+      "step": 640
+    },
+    {
+      "epoch": 58.64,
+      "learning_rate": 7e-06,
+      "loss": 1.4473,
+      "step": 645
+    },
+    {
+      "epoch": 58.64,
+      "eval_loss": 2.5989134311676025,
+      "eval_runtime": 0.4215,
+      "eval_samples_per_second": 14.237,
+      "eval_steps_per_second": 2.373,
+      "step": 645
+    },
+    {
+      "epoch": 59.09,
+      "learning_rate": 7e-06,
+      "loss": 1.7458,
+      "step": 650
+    },
+    {
+      "epoch": 59.09,
+      "eval_loss": 2.615797758102417,
+      "eval_runtime": 0.4163,
+      "eval_samples_per_second": 14.414,
+      "eval_steps_per_second": 2.402,
+      "step": 650
+    },
+    {
+      "epoch": 59.55,
+      "learning_rate": 7e-06,
+      "loss": 1.5373,
+      "step": 655
+    },
+    {
+      "epoch": 59.55,
+      "eval_loss": 2.6261303424835205,
+      "eval_runtime": 0.4089,
+      "eval_samples_per_second": 14.675,
+      "eval_steps_per_second": 2.446,
+      "step": 655
+    },
+    {
+      "epoch": 60.0,
+      "learning_rate": 7e-06,
+      "loss": 1.7489,
+      "step": 660
+    },
+    {
+      "epoch": 60.0,
+      "eval_loss": 2.61822509765625,
+      "eval_runtime": 0.4279,
+      "eval_samples_per_second": 14.022,
+      "eval_steps_per_second": 2.337,
+      "step": 660
+    },
+    {
+      "epoch": 60.45,
+      "learning_rate": 7e-06,
+      "loss": 1.6411,
+      "step": 665
+    },
+    {
+      "epoch": 60.45,
+      "eval_loss": 2.618873357772827,
+      "eval_runtime": 0.409,
+      "eval_samples_per_second": 14.668,
+      "eval_steps_per_second": 2.445,
+      "step": 665
+    },
+    {
+      "epoch": 60.91,
+      "learning_rate": 7e-06,
+      "loss": 1.6075,
+      "step": 670
+    },
+    {
+      "epoch": 60.91,
+      "eval_loss": 2.6396188735961914,
+      "eval_runtime": 0.4237,
+      "eval_samples_per_second": 14.16,
+      "eval_steps_per_second": 2.36,
+      "step": 670
+    },
+    {
+      "epoch": 61.36,
+      "learning_rate": 7e-06,
+      "loss": 1.7576,
+      "step": 675
+    },
+    {
+      "epoch": 61.36,
+      "eval_loss": 2.644122362136841,
+      "eval_runtime": 0.4237,
+      "eval_samples_per_second": 14.163,
+      "eval_steps_per_second": 2.36,
+      "step": 675
+    },
+    {
+      "epoch": 61.82,
+      "learning_rate": 7e-06,
+      "loss": 1.5172,
+      "step": 680
+    },
+    {
+      "epoch": 61.82,
+      "eval_loss": 2.6421959400177,
+      "eval_runtime": 0.4428,
+      "eval_samples_per_second": 13.549,
+      "eval_steps_per_second": 2.258,
+      "step": 680
+    },
+    {
+      "epoch": 62.27,
+      "learning_rate": 7e-06,
+      "loss": 1.4829,
+      "step": 685
+    },
+    {
+      "epoch": 62.27,
+      "eval_loss": 2.654413938522339,
+      "eval_runtime": 0.4274,
+      "eval_samples_per_second": 14.037,
+      "eval_steps_per_second": 2.34,
+      "step": 685
+    },
+    {
+      "epoch": 62.73,
+      "learning_rate": 7e-06,
+      "loss": 1.4743,
+      "step": 690
+    },
+    {
+      "epoch": 62.73,
+      "eval_loss": 2.6696255207061768,
+      "eval_runtime": 0.4519,
+      "eval_samples_per_second": 13.278,
+      "eval_steps_per_second": 2.213,
+      "step": 690
+    },
+    {
+      "epoch": 63.18,
+      "learning_rate": 7e-06,
+      "loss": 1.7067,
+      "step": 695
+    },
+    {
+      "epoch": 63.18,
+      "eval_loss": 2.6701173782348633,
+      "eval_runtime": 0.4269,
+      "eval_samples_per_second": 14.055,
+      "eval_steps_per_second": 2.343,
+      "step": 695
+    },
+    {
+      "epoch": 63.64,
+      "learning_rate": 7e-06,
+      "loss": 1.5196,
+      "step": 700
+    },
+    {
+      "epoch": 63.64,
+      "eval_loss": 2.6708524227142334,
+      "eval_runtime": 0.4296,
+      "eval_samples_per_second": 13.965,
+      "eval_steps_per_second": 2.328,
+      "step": 700
+    },
+    {
+      "epoch": 64.09,
+      "learning_rate": 7e-06,
+      "loss": 1.5256,
+      "step": 705
+    },
+    {
+      "epoch": 64.09,
+      "eval_loss": 2.682978630065918,
+      "eval_runtime": 0.4249,
+      "eval_samples_per_second": 14.122,
+      "eval_steps_per_second": 2.354,
+      "step": 705
+    },
+    {
+      "epoch": 64.55,
+      "learning_rate": 7e-06,
+      "loss": 1.4826,
+      "step": 710
+    },
+    {
+      "epoch": 64.55,
+      "eval_loss": 2.7001264095306396,
+      "eval_runtime": 0.4209,
+      "eval_samples_per_second": 14.257,
+      "eval_steps_per_second": 2.376,
+      "step": 710
+    },
+    {
+      "epoch": 65.0,
+      "learning_rate": 7e-06,
+      "loss": 1.6114,
+      "step": 715
+    },
+    {
+      "epoch": 65.0,
+      "eval_loss": 2.688549280166626,
+      "eval_runtime": 0.4252,
+      "eval_samples_per_second": 14.11,
+      "eval_steps_per_second": 2.352,
+      "step": 715
+    },
+    {
+      "epoch": 65.45,
+      "learning_rate": 7e-06,
+      "loss": 1.4344,
+      "step": 720
+    },
+    {
+      "epoch": 65.45,
+      "eval_loss": 2.689279317855835,
+      "eval_runtime": 0.4107,
+      "eval_samples_per_second": 14.607,
+      "eval_steps_per_second": 2.435,
+      "step": 720
+    },
+    {
+      "epoch": 65.91,
+      "learning_rate": 7e-06,
+      "loss": 1.4616,
+      "step": 725
+    },
+    {
+      "epoch": 65.91,
+      "eval_loss": 2.714517831802368,
+      "eval_runtime": 0.409,
+      "eval_samples_per_second": 14.669,
+      "eval_steps_per_second": 2.445,
+      "step": 725
+    },
+    {
+      "epoch": 66.36,
+      "learning_rate": 7e-06,
+      "loss": 1.5904,
+      "step": 730
+    },
+    {
+      "epoch": 66.36,
+      "eval_loss": 2.7193477153778076,
+      "eval_runtime": 0.4115,
+      "eval_samples_per_second": 14.581,
+      "eval_steps_per_second": 2.43,
+      "step": 730
+    },
+    {
+      "epoch": 66.82,
+      "learning_rate": 7e-06,
+      "loss": 1.4991,
+      "step": 735
+    },
+    {
+      "epoch": 66.82,
+      "eval_loss": 2.719238042831421,
+      "eval_runtime": 0.4072,
+      "eval_samples_per_second": 14.735,
+      "eval_steps_per_second": 2.456,
+      "step": 735
+    },
+    {
+      "epoch": 67.27,
+      "learning_rate": 7e-06,
+      "loss": 1.4878,
+      "step": 740
+    },
+    {
+      "epoch": 67.27,
+      "eval_loss": 2.724534749984741,
+      "eval_runtime": 0.4272,
+      "eval_samples_per_second": 14.045,
+      "eval_steps_per_second": 2.341,
+      "step": 740
+    },
+    {
+      "epoch": 67.73,
+      "learning_rate": 7e-06,
+      "loss": 1.5637,
+      "step": 745
+    },
+    {
+      "epoch": 67.73,
+      "eval_loss": 2.733035087585449,
+      "eval_runtime": 0.4092,
+      "eval_samples_per_second": 14.663,
+      "eval_steps_per_second": 2.444,
+      "step": 745
+    },
+    {
+      "epoch": 68.18,
+      "learning_rate": 7e-06,
+      "loss": 1.4269,
+      "step": 750
+    },
+    {
+      "epoch": 68.18,
+      "eval_loss": 2.751675844192505,
+      "eval_runtime": 0.4144,
+      "eval_samples_per_second": 14.477,
+      "eval_steps_per_second": 2.413,
+      "step": 750
+    }
+  ],
+  "max_steps": 1100,
+  "num_train_epochs": 100,
+  "total_flos": 228204741599232.0,
+  "trial_name": null,
+  "trial_params": null
+}

stellaris-races-2/checkpoint-750/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26d2b678a4d9f285f917a1b9046a7a81b7861eec33925ea0244852fdce629f6f
+size 3963

stellaris-races-3/checkpoint-250/adapter_config.json

@@ -0,0 +1,17 @@
+{
+  "base_model_name_or_path": "",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

stellaris-races-3/checkpoint-250/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c532ca75b6e9a65edfc1c5e582c021ad9729566b32f4f2435a559fdb8793e77
+size 104915277

stellaris-races-3/checkpoint-250/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28602c47db6a33f476c0b4f7192d74034b02540ea3c85f8f91c09ca2009eb7bd
+size 209849861

stellaris-races-3/checkpoint-250/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09959f000a3ff786abb3087f01081f37c686d1decfb09cdeac3c3db4094548bb
+size 14575

stellaris-races-3/checkpoint-250/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d084b51ea3803fa90876f08f0895b446a798096e792b4b0ecfcf296d75084092
+size 627

stellaris-races-3/checkpoint-250/trainer_state.json

@@ -0,0 +1,716 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 22.727272727272727,
+  "global_step": 250,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.45,
+      "learning_rate": 7e-06,
+      "loss": 2.6772,
+      "step": 5
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 2.7022464275360107,
+      "eval_runtime": 0.7068,
+      "eval_samples_per_second": 8.489,
+      "eval_steps_per_second": 1.415,
+      "step": 5
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 7e-06,
+      "loss": 2.8149,
+      "step": 10
+    },
+    {
+      "epoch": 0.91,
+      "eval_loss": 2.6998846530914307,
+      "eval_runtime": 0.7138,
+      "eval_samples_per_second": 8.406,
+      "eval_steps_per_second": 1.401,
+      "step": 10
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 7e-06,
+      "loss": 2.8494,
+      "step": 15
+    },
+    {
+      "epoch": 1.36,
+      "eval_loss": 2.6970536708831787,
+      "eval_runtime": 0.7302,
+      "eval_samples_per_second": 8.216,
+      "eval_steps_per_second": 1.369,
+      "step": 15
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7e-06,
+      "loss": 2.7035,
+      "step": 20
+    },
+    {
+      "epoch": 1.82,
+      "eval_loss": 2.6939141750335693,
+      "eval_runtime": 0.7129,
+      "eval_samples_per_second": 8.416,
+      "eval_steps_per_second": 1.403,
+      "step": 20
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 7e-06,
+      "loss": 2.5667,
+      "step": 25
+    },
+    {
+      "epoch": 2.27,
+      "eval_loss": 2.6903278827667236,
+      "eval_runtime": 0.7707,
+      "eval_samples_per_second": 7.785,
+      "eval_steps_per_second": 1.298,
+      "step": 25
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 7e-06,
+      "loss": 2.7766,
+      "step": 30
+    },
+    {
+      "epoch": 2.73,
+      "eval_loss": 2.6863253116607666,
+      "eval_runtime": 0.7245,
+      "eval_samples_per_second": 8.281,
+      "eval_steps_per_second": 1.38,
+      "step": 30
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 7e-06,
+      "loss": 2.7507,
+      "step": 35
+    },
+    {
+      "epoch": 3.18,
+      "eval_loss": 2.6823132038116455,
+      "eval_runtime": 0.7678,
+      "eval_samples_per_second": 7.815,
+      "eval_steps_per_second": 1.302,
+      "step": 35
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 7e-06,
+      "loss": 2.7033,
+      "step": 40
+    },
+    {
+      "epoch": 3.64,
+      "eval_loss": 2.677713632583618,
+      "eval_runtime": 0.7464,
+      "eval_samples_per_second": 8.038,
+      "eval_steps_per_second": 1.34,
+      "step": 40
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 7e-06,
+      "loss": 2.714,
+      "step": 45
+    },
+    {
+      "epoch": 4.09,
+      "eval_loss": 2.6728098392486572,
+      "eval_runtime": 0.7176,
+      "eval_samples_per_second": 8.361,
+      "eval_steps_per_second": 1.393,
+      "step": 45
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 7e-06,
+      "loss": 2.7606,
+      "step": 50
+    },
+    {
+      "epoch": 4.55,
+      "eval_loss": 2.6675455570220947,
+      "eval_runtime": 0.7365,
+      "eval_samples_per_second": 8.147,
+      "eval_steps_per_second": 1.358,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 7e-06,
+      "loss": 2.6694,
+      "step": 55
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.6618385314941406,
+      "eval_runtime": 0.7145,
+      "eval_samples_per_second": 8.398,
+      "eval_steps_per_second": 1.4,
+      "step": 55
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 7e-06,
+      "loss": 2.6477,
+      "step": 60
+    },
+    {
+      "epoch": 5.45,
+      "eval_loss": 2.6561715602874756,
+      "eval_runtime": 0.7766,
+      "eval_samples_per_second": 7.726,
+      "eval_steps_per_second": 1.288,
+      "step": 60
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 7e-06,
+      "loss": 2.7613,
+      "step": 65
+    },
+    {
+      "epoch": 5.91,
+      "eval_loss": 2.649792432785034,
+      "eval_runtime": 0.7178,
+      "eval_samples_per_second": 8.359,
+      "eval_steps_per_second": 1.393,
+      "step": 65
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 7e-06,
+      "loss": 2.6446,
+      "step": 70
+    },
+    {
+      "epoch": 6.36,
+      "eval_loss": 2.64331316947937,
+      "eval_runtime": 0.7629,
+      "eval_samples_per_second": 7.865,
+      "eval_steps_per_second": 1.311,
+      "step": 70
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 7e-06,
+      "loss": 2.7039,
+      "step": 75
+    },
+    {
+      "epoch": 6.82,
+      "eval_loss": 2.6366422176361084,
+      "eval_runtime": 0.7438,
+      "eval_samples_per_second": 8.067,
+      "eval_steps_per_second": 1.345,
+      "step": 75
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 7e-06,
+      "loss": 2.5376,
+      "step": 80
+    },
+    {
+      "epoch": 7.27,
+      "eval_loss": 2.6297147274017334,
+      "eval_runtime": 0.7738,
+      "eval_samples_per_second": 7.754,
+      "eval_steps_per_second": 1.292,
+      "step": 80
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 7e-06,
+      "loss": 2.6866,
+      "step": 85
+    },
+    {
+      "epoch": 7.73,
+      "eval_loss": 2.622584581375122,
+      "eval_runtime": 0.7654,
+      "eval_samples_per_second": 7.839,
+      "eval_steps_per_second": 1.307,
+      "step": 85
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 7e-06,
+      "loss": 2.6996,
+      "step": 90
+    },
+    {
+      "epoch": 8.18,
+      "eval_loss": 2.615074396133423,
+      "eval_runtime": 0.7534,
+      "eval_samples_per_second": 7.964,
+      "eval_steps_per_second": 1.327,
+      "step": 90
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 7e-06,
+      "loss": 2.72,
+      "step": 95
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 2.6072492599487305,
+      "eval_runtime": 0.7627,
+      "eval_samples_per_second": 7.867,
+      "eval_steps_per_second": 1.311,
+      "step": 95
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 7e-06,
+      "loss": 2.4708,
+      "step": 100
+    },
+    {
+      "epoch": 9.09,
+      "eval_loss": 2.5988943576812744,
+      "eval_runtime": 0.7612,
+      "eval_samples_per_second": 7.882,
+      "eval_steps_per_second": 1.314,
+      "step": 100
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 7e-06,
+      "loss": 2.5311,
+      "step": 105
+    },
+    {
+      "epoch": 9.55,
+      "eval_loss": 2.5901479721069336,
+      "eval_runtime": 0.7561,
+      "eval_samples_per_second": 7.935,
+      "eval_steps_per_second": 1.322,
+      "step": 105
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7e-06,
+      "loss": 2.6912,
+      "step": 110
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.580698013305664,
+      "eval_runtime": 0.7525,
+      "eval_samples_per_second": 7.973,
+      "eval_steps_per_second": 1.329,
+      "step": 110
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 7e-06,
+      "loss": 2.6089,
+      "step": 115
+    },
+    {
+      "epoch": 10.45,
+      "eval_loss": 2.5717146396636963,
+      "eval_runtime": 0.765,
+      "eval_samples_per_second": 7.843,
+      "eval_steps_per_second": 1.307,
+      "step": 115
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 7e-06,
+      "loss": 2.4803,
+      "step": 120
+    },
+    {
+      "epoch": 10.91,
+      "eval_loss": 2.5612823963165283,
+      "eval_runtime": 0.7202,
+      "eval_samples_per_second": 8.331,
+      "eval_steps_per_second": 1.388,
+      "step": 120
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 7e-06,
+      "loss": 2.6576,
+      "step": 125
+    },
+    {
+      "epoch": 11.36,
+      "eval_loss": 2.5491273403167725,
+      "eval_runtime": 0.7237,
+      "eval_samples_per_second": 8.291,
+      "eval_steps_per_second": 1.382,
+      "step": 125
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 7e-06,
+      "loss": 2.4848,
+      "step": 130
+    },
+    {
+      "epoch": 11.82,
+      "eval_loss": 2.536881446838379,
+      "eval_runtime": 0.749,
+      "eval_samples_per_second": 8.011,
+      "eval_steps_per_second": 1.335,
+      "step": 130
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 7e-06,
+      "loss": 2.4875,
+      "step": 135
+    },
+    {
+      "epoch": 12.27,
+      "eval_loss": 2.524549722671509,
+      "eval_runtime": 0.7273,
+      "eval_samples_per_second": 8.25,
+      "eval_steps_per_second": 1.375,
+      "step": 135
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 7e-06,
+      "loss": 2.5707,
+      "step": 140
+    },
+    {
+      "epoch": 12.73,
+      "eval_loss": 2.511542558670044,
+      "eval_runtime": 0.7341,
+      "eval_samples_per_second": 8.174,
+      "eval_steps_per_second": 1.362,
+      "step": 140
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 7e-06,
+      "loss": 2.3694,
+      "step": 145
+    },
+    {
+      "epoch": 13.18,
+      "eval_loss": 2.4980533123016357,
+      "eval_runtime": 0.733,
+      "eval_samples_per_second": 8.186,
+      "eval_steps_per_second": 1.364,
+      "step": 145
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 7e-06,
+      "loss": 2.515,
+      "step": 150
+    },
+    {
+      "epoch": 13.64,
+      "eval_loss": 2.4852240085601807,
+      "eval_runtime": 0.7212,
+      "eval_samples_per_second": 8.319,
+      "eval_steps_per_second": 1.387,
+      "step": 150
+    },
+    {
+      "epoch": 14.09,
+      "learning_rate": 7e-06,
+      "loss": 2.5051,
+      "step": 155
+    },
+    {
+      "epoch": 14.09,
+      "eval_loss": 2.471349000930786,
+      "eval_runtime": 0.7219,
+      "eval_samples_per_second": 8.311,
+      "eval_steps_per_second": 1.385,
+      "step": 155
+    },
+    {
+      "epoch": 14.55,
+      "learning_rate": 7e-06,
+      "loss": 2.5314,
+      "step": 160
+    },
+    {
+      "epoch": 14.55,
+      "eval_loss": 2.456756830215454,
+      "eval_runtime": 0.7344,
+      "eval_samples_per_second": 8.17,
+      "eval_steps_per_second": 1.362,
+      "step": 160
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3409,
+      "step": 165
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.4444773197174072,
+      "eval_runtime": 0.7207,
+      "eval_samples_per_second": 8.326,
+      "eval_steps_per_second": 1.388,
+      "step": 165
+    },
+    {
+      "epoch": 15.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4272,
+      "step": 170
+    },
+    {
+      "epoch": 15.45,
+      "eval_loss": 2.4356348514556885,
+      "eval_runtime": 0.7258,
+      "eval_samples_per_second": 8.267,
+      "eval_steps_per_second": 1.378,
+      "step": 170
+    },
+    {
+      "epoch": 15.91,
+      "learning_rate": 7e-06,
+      "loss": 2.4189,
+      "step": 175
+    },
+    {
+      "epoch": 15.91,
+      "eval_loss": 2.428696393966675,
+      "eval_runtime": 0.7229,
+      "eval_samples_per_second": 8.3,
+      "eval_steps_per_second": 1.383,
+      "step": 175
+    },
+    {
+      "epoch": 16.36,
+      "learning_rate": 7e-06,
+      "loss": 2.3876,
+      "step": 180
+    },
+    {
+      "epoch": 16.36,
+      "eval_loss": 2.422900915145874,
+      "eval_runtime": 0.7211,
+      "eval_samples_per_second": 8.321,
+      "eval_steps_per_second": 1.387,
+      "step": 180
+    },
+    {
+      "epoch": 16.82,
+      "learning_rate": 7e-06,
+      "loss": 2.3212,
+      "step": 185
+    },
+    {
+      "epoch": 16.82,
+      "eval_loss": 2.4173238277435303,
+      "eval_runtime": 0.72,
+      "eval_samples_per_second": 8.334,
+      "eval_steps_per_second": 1.389,
+      "step": 185
+    },
+    {
+      "epoch": 17.27,
+      "learning_rate": 7e-06,
+      "loss": 2.3173,
+      "step": 190
+    },
+    {
+      "epoch": 17.27,
+      "eval_loss": 2.4116928577423096,
+      "eval_runtime": 0.7204,
+      "eval_samples_per_second": 8.329,
+      "eval_steps_per_second": 1.388,
+      "step": 190
+    },
+    {
+      "epoch": 17.73,
+      "learning_rate": 7e-06,
+      "loss": 2.3327,
+      "step": 195
+    },
+    {
+      "epoch": 17.73,
+      "eval_loss": 2.4064242839813232,
+      "eval_runtime": 0.7198,
+      "eval_samples_per_second": 8.335,
+      "eval_steps_per_second": 1.389,
+      "step": 195
+    },
+    {
+      "epoch": 18.18,
+      "learning_rate": 7e-06,
+      "loss": 2.5256,
+      "step": 200
+    },
+    {
+      "epoch": 18.18,
+      "eval_loss": 2.4015276432037354,
+      "eval_runtime": 0.7429,
+      "eval_samples_per_second": 8.077,
+      "eval_steps_per_second": 1.346,
+      "step": 200
+    },
+    {
+      "epoch": 18.64,
+      "learning_rate": 7e-06,
+      "loss": 2.2825,
+      "step": 205
+    },
+    {
+      "epoch": 18.64,
+      "eval_loss": 2.3970677852630615,
+      "eval_runtime": 0.7175,
+      "eval_samples_per_second": 8.363,
+      "eval_steps_per_second": 1.394,
+      "step": 205
+    },
+    {
+      "epoch": 19.09,
+      "learning_rate": 7e-06,
+      "loss": 2.3417,
+      "step": 210
+    },
+    {
+      "epoch": 19.09,
+      "eval_loss": 2.3925275802612305,
+      "eval_runtime": 0.7168,
+      "eval_samples_per_second": 8.37,
+      "eval_steps_per_second": 1.395,
+      "step": 210
+    },
+    {
+      "epoch": 19.55,
+      "learning_rate": 7e-06,
+      "loss": 2.2123,
+      "step": 215
+    },
+    {
+      "epoch": 19.55,
+      "eval_loss": 2.388213872909546,
+      "eval_runtime": 0.7288,
+      "eval_samples_per_second": 8.232,
+      "eval_steps_per_second": 1.372,
+      "step": 215
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3696,
+      "step": 220
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 2.384182929992676,
+      "eval_runtime": 0.7182,
+      "eval_samples_per_second": 8.354,
+      "eval_steps_per_second": 1.392,
+      "step": 220
+    },
+    {
+      "epoch": 20.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4359,
+      "step": 225
+    },
+    {
+      "epoch": 20.45,
+      "eval_loss": 2.380094528198242,
+      "eval_runtime": 0.718,
+      "eval_samples_per_second": 8.357,
+      "eval_steps_per_second": 1.393,
+      "step": 225
+    },
+    {
+      "epoch": 20.91,
+      "learning_rate": 7e-06,
+      "loss": 2.1279,
+      "step": 230
+    },
+    {
+      "epoch": 20.91,
+      "eval_loss": 2.3762617111206055,
+      "eval_runtime": 0.7166,
+      "eval_samples_per_second": 8.372,
+      "eval_steps_per_second": 1.395,
+      "step": 230
+    },
+    {
+      "epoch": 21.36,
+      "learning_rate": 7e-06,
+      "loss": 2.3891,
+      "step": 235
+    },
+    {
+      "epoch": 21.36,
+      "eval_loss": 2.3726041316986084,
+      "eval_runtime": 0.7167,
+      "eval_samples_per_second": 8.371,
+      "eval_steps_per_second": 1.395,
+      "step": 235
+    },
+    {
+      "epoch": 21.82,
+      "learning_rate": 7e-06,
+      "loss": 2.1994,
+      "step": 240
+    },
+    {
+      "epoch": 21.82,
+      "eval_loss": 2.36929988861084,
+      "eval_runtime": 0.7208,
+      "eval_samples_per_second": 8.324,
+      "eval_steps_per_second": 1.387,
+      "step": 240
+    },
+    {
+      "epoch": 22.27,
+      "learning_rate": 7e-06,
+      "loss": 2.2041,
+      "step": 245
+    },
+    {
+      "epoch": 22.27,
+      "eval_loss": 2.3658368587493896,
+      "eval_runtime": 0.7161,
+      "eval_samples_per_second": 8.379,
+      "eval_steps_per_second": 1.396,
+      "step": 245
+    },
+    {
+      "epoch": 22.73,
+      "learning_rate": 7e-06,
+      "loss": 2.3645,
+      "step": 250
+    },
+    {
+      "epoch": 22.73,
+      "eval_loss": 2.362556219100952,
+      "eval_runtime": 0.7162,
+      "eval_samples_per_second": 8.378,
+      "eval_steps_per_second": 1.396,
+      "step": 250
+    }
+  ],
+  "max_steps": 550,
+  "num_train_epochs": 50,
+  "total_flos": 96658126202880.0,
+  "trial_name": null,
+  "trial_params": null
+}

stellaris-races-3/checkpoint-250/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:831e1ea9ed3dc21bd34506e70712bf07dfd13c40b89ff8d8ff382c47a9827312
+size 3963

stellaris-races-3/checkpoint-300/adapter_config.json

@@ -0,0 +1,17 @@
+{
+  "base_model_name_or_path": "",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

stellaris-races-3/checkpoint-300/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ea1eec6628944a73243408f8963a9d73725464fdd8ec8b763e160797f479fa2
+size 104915277

stellaris-races-3/checkpoint-300/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37fa98a3d95eca6f5aa818efa87aef6e39cc82bf258bf810fdbf7602bc40d3b2
+size 209849861

stellaris-races-3/checkpoint-300/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28be4d5645fcc7332cbface2653e58acb3e625ba63e23204b1ba5a11db9f296e
+size 14575

stellaris-races-3/checkpoint-300/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37046c55ed70d87b1517a498636cba6e37b1fc862b4d2d81a120d489e2da361b
+size 627

stellaris-races-3/checkpoint-300/trainer_state.json

@@ -0,0 +1,856 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 27.272727272727273,
+  "global_step": 300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.45,
+      "learning_rate": 7e-06,
+      "loss": 2.6772,
+      "step": 5
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 2.7022464275360107,
+      "eval_runtime": 0.7068,
+      "eval_samples_per_second": 8.489,
+      "eval_steps_per_second": 1.415,
+      "step": 5
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 7e-06,
+      "loss": 2.8149,
+      "step": 10
+    },
+    {
+      "epoch": 0.91,
+      "eval_loss": 2.6998846530914307,
+      "eval_runtime": 0.7138,
+      "eval_samples_per_second": 8.406,
+      "eval_steps_per_second": 1.401,
+      "step": 10
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 7e-06,
+      "loss": 2.8494,
+      "step": 15
+    },
+    {
+      "epoch": 1.36,
+      "eval_loss": 2.6970536708831787,
+      "eval_runtime": 0.7302,
+      "eval_samples_per_second": 8.216,
+      "eval_steps_per_second": 1.369,
+      "step": 15
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7e-06,
+      "loss": 2.7035,
+      "step": 20
+    },
+    {
+      "epoch": 1.82,
+      "eval_loss": 2.6939141750335693,
+      "eval_runtime": 0.7129,
+      "eval_samples_per_second": 8.416,
+      "eval_steps_per_second": 1.403,
+      "step": 20
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 7e-06,
+      "loss": 2.5667,
+      "step": 25
+    },
+    {
+      "epoch": 2.27,
+      "eval_loss": 2.6903278827667236,
+      "eval_runtime": 0.7707,
+      "eval_samples_per_second": 7.785,
+      "eval_steps_per_second": 1.298,
+      "step": 25
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 7e-06,
+      "loss": 2.7766,
+      "step": 30
+    },
+    {
+      "epoch": 2.73,
+      "eval_loss": 2.6863253116607666,
+      "eval_runtime": 0.7245,
+      "eval_samples_per_second": 8.281,
+      "eval_steps_per_second": 1.38,
+      "step": 30
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 7e-06,
+      "loss": 2.7507,
+      "step": 35
+    },
+    {
+      "epoch": 3.18,
+      "eval_loss": 2.6823132038116455,
+      "eval_runtime": 0.7678,
+      "eval_samples_per_second": 7.815,
+      "eval_steps_per_second": 1.302,
+      "step": 35
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 7e-06,
+      "loss": 2.7033,
+      "step": 40
+    },
+    {
+      "epoch": 3.64,
+      "eval_loss": 2.677713632583618,
+      "eval_runtime": 0.7464,
+      "eval_samples_per_second": 8.038,
+      "eval_steps_per_second": 1.34,
+      "step": 40
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 7e-06,
+      "loss": 2.714,
+      "step": 45
+    },
+    {
+      "epoch": 4.09,
+      "eval_loss": 2.6728098392486572,
+      "eval_runtime": 0.7176,
+      "eval_samples_per_second": 8.361,
+      "eval_steps_per_second": 1.393,
+      "step": 45
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 7e-06,
+      "loss": 2.7606,
+      "step": 50
+    },
+    {
+      "epoch": 4.55,
+      "eval_loss": 2.6675455570220947,
+      "eval_runtime": 0.7365,
+      "eval_samples_per_second": 8.147,
+      "eval_steps_per_second": 1.358,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 7e-06,
+      "loss": 2.6694,
+      "step": 55
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.6618385314941406,
+      "eval_runtime": 0.7145,
+      "eval_samples_per_second": 8.398,
+      "eval_steps_per_second": 1.4,
+      "step": 55
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 7e-06,
+      "loss": 2.6477,
+      "step": 60
+    },
+    {
+      "epoch": 5.45,
+      "eval_loss": 2.6561715602874756,
+      "eval_runtime": 0.7766,
+      "eval_samples_per_second": 7.726,
+      "eval_steps_per_second": 1.288,
+      "step": 60
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 7e-06,
+      "loss": 2.7613,
+      "step": 65
+    },
+    {
+      "epoch": 5.91,
+      "eval_loss": 2.649792432785034,
+      "eval_runtime": 0.7178,
+      "eval_samples_per_second": 8.359,
+      "eval_steps_per_second": 1.393,
+      "step": 65
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 7e-06,
+      "loss": 2.6446,
+      "step": 70
+    },
+    {
+      "epoch": 6.36,
+      "eval_loss": 2.64331316947937,
+      "eval_runtime": 0.7629,
+      "eval_samples_per_second": 7.865,
+      "eval_steps_per_second": 1.311,
+      "step": 70
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 7e-06,
+      "loss": 2.7039,
+      "step": 75
+    },
+    {
+      "epoch": 6.82,
+      "eval_loss": 2.6366422176361084,
+      "eval_runtime": 0.7438,
+      "eval_samples_per_second": 8.067,
+      "eval_steps_per_second": 1.345,
+      "step": 75
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 7e-06,
+      "loss": 2.5376,
+      "step": 80
+    },
+    {
+      "epoch": 7.27,
+      "eval_loss": 2.6297147274017334,
+      "eval_runtime": 0.7738,
+      "eval_samples_per_second": 7.754,
+      "eval_steps_per_second": 1.292,
+      "step": 80
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 7e-06,
+      "loss": 2.6866,
+      "step": 85
+    },
+    {
+      "epoch": 7.73,
+      "eval_loss": 2.622584581375122,
+      "eval_runtime": 0.7654,
+      "eval_samples_per_second": 7.839,
+      "eval_steps_per_second": 1.307,
+      "step": 85
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 7e-06,
+      "loss": 2.6996,
+      "step": 90
+    },
+    {
+      "epoch": 8.18,
+      "eval_loss": 2.615074396133423,
+      "eval_runtime": 0.7534,
+      "eval_samples_per_second": 7.964,
+      "eval_steps_per_second": 1.327,
+      "step": 90
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 7e-06,
+      "loss": 2.72,
+      "step": 95
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 2.6072492599487305,
+      "eval_runtime": 0.7627,
+      "eval_samples_per_second": 7.867,
+      "eval_steps_per_second": 1.311,
+      "step": 95
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 7e-06,
+      "loss": 2.4708,
+      "step": 100
+    },
+    {
+      "epoch": 9.09,
+      "eval_loss": 2.5988943576812744,
+      "eval_runtime": 0.7612,
+      "eval_samples_per_second": 7.882,
+      "eval_steps_per_second": 1.314,
+      "step": 100
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 7e-06,
+      "loss": 2.5311,
+      "step": 105
+    },
+    {
+      "epoch": 9.55,
+      "eval_loss": 2.5901479721069336,
+      "eval_runtime": 0.7561,
+      "eval_samples_per_second": 7.935,
+      "eval_steps_per_second": 1.322,
+      "step": 105
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7e-06,
+      "loss": 2.6912,
+      "step": 110
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.580698013305664,
+      "eval_runtime": 0.7525,
+      "eval_samples_per_second": 7.973,
+      "eval_steps_per_second": 1.329,
+      "step": 110
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 7e-06,
+      "loss": 2.6089,
+      "step": 115
+    },
+    {
+      "epoch": 10.45,
+      "eval_loss": 2.5717146396636963,
+      "eval_runtime": 0.765,
+      "eval_samples_per_second": 7.843,
+      "eval_steps_per_second": 1.307,
+      "step": 115
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 7e-06,
+      "loss": 2.4803,
+      "step": 120
+    },
+    {
+      "epoch": 10.91,
+      "eval_loss": 2.5612823963165283,
+      "eval_runtime": 0.7202,
+      "eval_samples_per_second": 8.331,
+      "eval_steps_per_second": 1.388,
+      "step": 120
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 7e-06,
+      "loss": 2.6576,
+      "step": 125
+    },
+    {
+      "epoch": 11.36,
+      "eval_loss": 2.5491273403167725,
+      "eval_runtime": 0.7237,
+      "eval_samples_per_second": 8.291,
+      "eval_steps_per_second": 1.382,
+      "step": 125
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 7e-06,
+      "loss": 2.4848,
+      "step": 130
+    },
+    {
+      "epoch": 11.82,
+      "eval_loss": 2.536881446838379,
+      "eval_runtime": 0.749,
+      "eval_samples_per_second": 8.011,
+      "eval_steps_per_second": 1.335,
+      "step": 130
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 7e-06,
+      "loss": 2.4875,
+      "step": 135
+    },
+    {
+      "epoch": 12.27,
+      "eval_loss": 2.524549722671509,
+      "eval_runtime": 0.7273,
+      "eval_samples_per_second": 8.25,
+      "eval_steps_per_second": 1.375,
+      "step": 135
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 7e-06,
+      "loss": 2.5707,
+      "step": 140
+    },
+    {
+      "epoch": 12.73,
+      "eval_loss": 2.511542558670044,
+      "eval_runtime": 0.7341,
+      "eval_samples_per_second": 8.174,
+      "eval_steps_per_second": 1.362,
+      "step": 140
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 7e-06,
+      "loss": 2.3694,
+      "step": 145
+    },
+    {
+      "epoch": 13.18,
+      "eval_loss": 2.4980533123016357,
+      "eval_runtime": 0.733,
+      "eval_samples_per_second": 8.186,
+      "eval_steps_per_second": 1.364,
+      "step": 145
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 7e-06,
+      "loss": 2.515,
+      "step": 150
+    },
+    {
+      "epoch": 13.64,
+      "eval_loss": 2.4852240085601807,
+      "eval_runtime": 0.7212,
+      "eval_samples_per_second": 8.319,
+      "eval_steps_per_second": 1.387,
+      "step": 150
+    },
+    {
+      "epoch": 14.09,
+      "learning_rate": 7e-06,
+      "loss": 2.5051,
+      "step": 155
+    },
+    {
+      "epoch": 14.09,
+      "eval_loss": 2.471349000930786,
+      "eval_runtime": 0.7219,
+      "eval_samples_per_second": 8.311,
+      "eval_steps_per_second": 1.385,
+      "step": 155
+    },
+    {
+      "epoch": 14.55,
+      "learning_rate": 7e-06,
+      "loss": 2.5314,
+      "step": 160
+    },
+    {
+      "epoch": 14.55,
+      "eval_loss": 2.456756830215454,
+      "eval_runtime": 0.7344,
+      "eval_samples_per_second": 8.17,
+      "eval_steps_per_second": 1.362,
+      "step": 160
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3409,
+      "step": 165
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.4444773197174072,
+      "eval_runtime": 0.7207,
+      "eval_samples_per_second": 8.326,
+      "eval_steps_per_second": 1.388,
+      "step": 165
+    },
+    {
+      "epoch": 15.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4272,
+      "step": 170
+    },
+    {
+      "epoch": 15.45,
+      "eval_loss": 2.4356348514556885,
+      "eval_runtime": 0.7258,
+      "eval_samples_per_second": 8.267,
+      "eval_steps_per_second": 1.378,
+      "step": 170
+    },
+    {
+      "epoch": 15.91,
+      "learning_rate": 7e-06,
+      "loss": 2.4189,
+      "step": 175
+    },
+    {
+      "epoch": 15.91,
+      "eval_loss": 2.428696393966675,
+      "eval_runtime": 0.7229,
+      "eval_samples_per_second": 8.3,
+      "eval_steps_per_second": 1.383,
+      "step": 175
+    },
+    {
+      "epoch": 16.36,
+      "learning_rate": 7e-06,
+      "loss": 2.3876,
+      "step": 180
+    },
+    {
+      "epoch": 16.36,
+      "eval_loss": 2.422900915145874,
+      "eval_runtime": 0.7211,
+      "eval_samples_per_second": 8.321,
+      "eval_steps_per_second": 1.387,
+      "step": 180
+    },
+    {
+      "epoch": 16.82,
+      "learning_rate": 7e-06,
+      "loss": 2.3212,
+      "step": 185
+    },
+    {
+      "epoch": 16.82,
+      "eval_loss": 2.4173238277435303,
+      "eval_runtime": 0.72,
+      "eval_samples_per_second": 8.334,
+      "eval_steps_per_second": 1.389,
+      "step": 185
+    },
+    {
+      "epoch": 17.27,
+      "learning_rate": 7e-06,
+      "loss": 2.3173,
+      "step": 190
+    },
+    {
+      "epoch": 17.27,
+      "eval_loss": 2.4116928577423096,
+      "eval_runtime": 0.7204,
+      "eval_samples_per_second": 8.329,
+      "eval_steps_per_second": 1.388,
+      "step": 190
+    },
+    {
+      "epoch": 17.73,
+      "learning_rate": 7e-06,
+      "loss": 2.3327,
+      "step": 195
+    },
+    {
+      "epoch": 17.73,
+      "eval_loss": 2.4064242839813232,
+      "eval_runtime": 0.7198,
+      "eval_samples_per_second": 8.335,
+      "eval_steps_per_second": 1.389,
+      "step": 195
+    },
+    {
+      "epoch": 18.18,
+      "learning_rate": 7e-06,
+      "loss": 2.5256,
+      "step": 200
+    },
+    {
+      "epoch": 18.18,
+      "eval_loss": 2.4015276432037354,
+      "eval_runtime": 0.7429,
+      "eval_samples_per_second": 8.077,
+      "eval_steps_per_second": 1.346,
+      "step": 200
+    },
+    {
+      "epoch": 18.64,
+      "learning_rate": 7e-06,
+      "loss": 2.2825,
+      "step": 205
+    },
+    {
+      "epoch": 18.64,
+      "eval_loss": 2.3970677852630615,
+      "eval_runtime": 0.7175,
+      "eval_samples_per_second": 8.363,
+      "eval_steps_per_second": 1.394,
+      "step": 205
+    },
+    {
+      "epoch": 19.09,
+      "learning_rate": 7e-06,
+      "loss": 2.3417,
+      "step": 210
+    },
+    {
+      "epoch": 19.09,
+      "eval_loss": 2.3925275802612305,
+      "eval_runtime": 0.7168,
+      "eval_samples_per_second": 8.37,
+      "eval_steps_per_second": 1.395,
+      "step": 210
+    },
+    {
+      "epoch": 19.55,
+      "learning_rate": 7e-06,
+      "loss": 2.2123,
+      "step": 215
+    },
+    {
+      "epoch": 19.55,
+      "eval_loss": 2.388213872909546,
+      "eval_runtime": 0.7288,
+      "eval_samples_per_second": 8.232,
+      "eval_steps_per_second": 1.372,
+      "step": 215
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3696,
+      "step": 220
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 2.384182929992676,
+      "eval_runtime": 0.7182,
+      "eval_samples_per_second": 8.354,
+      "eval_steps_per_second": 1.392,
+      "step": 220
+    },
+    {
+      "epoch": 20.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4359,
+      "step": 225
+    },
+    {
+      "epoch": 20.45,
+      "eval_loss": 2.380094528198242,
+      "eval_runtime": 0.718,
+      "eval_samples_per_second": 8.357,
+      "eval_steps_per_second": 1.393,
+      "step": 225
+    },
+    {
+      "epoch": 20.91,
+      "learning_rate": 7e-06,
+      "loss": 2.1279,
+      "step": 230
+    },
+    {
+      "epoch": 20.91,
+      "eval_loss": 2.3762617111206055,
+      "eval_runtime": 0.7166,
+      "eval_samples_per_second": 8.372,
+      "eval_steps_per_second": 1.395,
+      "step": 230
+    },
+    {
+      "epoch": 21.36,
+      "learning_rate": 7e-06,
+      "loss": 2.3891,
+      "step": 235
+    },
+    {
+      "epoch": 21.36,
+      "eval_loss": 2.3726041316986084,
+      "eval_runtime": 0.7167,
+      "eval_samples_per_second": 8.371,
+      "eval_steps_per_second": 1.395,
+      "step": 235
+    },
+    {
+      "epoch": 21.82,
+      "learning_rate": 7e-06,
+      "loss": 2.1994,
+      "step": 240
+    },
+    {
+      "epoch": 21.82,
+      "eval_loss": 2.36929988861084,
+      "eval_runtime": 0.7208,
+      "eval_samples_per_second": 8.324,
+      "eval_steps_per_second": 1.387,
+      "step": 240
+    },
+    {
+      "epoch": 22.27,
+      "learning_rate": 7e-06,
+      "loss": 2.2041,
+      "step": 245
+    },
+    {
+      "epoch": 22.27,
+      "eval_loss": 2.3658368587493896,
+      "eval_runtime": 0.7161,
+      "eval_samples_per_second": 8.379,
+      "eval_steps_per_second": 1.396,
+      "step": 245
+    },
+    {
+      "epoch": 22.73,
+      "learning_rate": 7e-06,
+      "loss": 2.3645,
+      "step": 250
+    },
+    {
+      "epoch": 22.73,
+      "eval_loss": 2.362556219100952,
+      "eval_runtime": 0.7162,
+      "eval_samples_per_second": 8.378,
+      "eval_steps_per_second": 1.396,
+      "step": 250
+    },
+    {
+      "epoch": 23.18,
+      "learning_rate": 7e-06,
+      "loss": 2.2448,
+      "step": 255
+    },
+    {
+      "epoch": 23.18,
+      "eval_loss": 2.359351873397827,
+      "eval_runtime": 0.7198,
+      "eval_samples_per_second": 8.336,
+      "eval_steps_per_second": 1.389,
+      "step": 255
+    },
+    {
+      "epoch": 23.64,
+      "learning_rate": 7e-06,
+      "loss": 2.1683,
+      "step": 260
+    },
+    {
+      "epoch": 23.64,
+      "eval_loss": 2.3562166690826416,
+      "eval_runtime": 0.7185,
+      "eval_samples_per_second": 8.351,
+      "eval_steps_per_second": 1.392,
+      "step": 260
+    },
+    {
+      "epoch": 24.09,
+      "learning_rate": 7e-06,
+      "loss": 2.34,
+      "step": 265
+    },
+    {
+      "epoch": 24.09,
+      "eval_loss": 2.352797746658325,
+      "eval_runtime": 0.717,
+      "eval_samples_per_second": 8.368,
+      "eval_steps_per_second": 1.395,
+      "step": 265
+    },
+    {
+      "epoch": 24.55,
+      "learning_rate": 7e-06,
+      "loss": 2.2091,
+      "step": 270
+    },
+    {
+      "epoch": 24.55,
+      "eval_loss": 2.349076271057129,
+      "eval_runtime": 0.7168,
+      "eval_samples_per_second": 8.371,
+      "eval_steps_per_second": 1.395,
+      "step": 270
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 7e-06,
+      "loss": 2.2733,
+      "step": 275
+    },
+    {
+      "epoch": 25.0,
+      "eval_loss": 2.3457562923431396,
+      "eval_runtime": 0.716,
+      "eval_samples_per_second": 8.379,
+      "eval_steps_per_second": 1.397,
+      "step": 275
+    },
+    {
+      "epoch": 25.45,
+      "learning_rate": 7e-06,
+      "loss": 2.3123,
+      "step": 280
+    },
+    {
+      "epoch": 25.45,
+      "eval_loss": 2.3424742221832275,
+      "eval_runtime": 0.7162,
+      "eval_samples_per_second": 8.378,
+      "eval_steps_per_second": 1.396,
+      "step": 280
+    },
+    {
+      "epoch": 25.91,
+      "learning_rate": 7e-06,
+      "loss": 2.1236,
+      "step": 285
+    },
+    {
+      "epoch": 25.91,
+      "eval_loss": 2.3390562534332275,
+      "eval_runtime": 0.7167,
+      "eval_samples_per_second": 8.372,
+      "eval_steps_per_second": 1.395,
+      "step": 285
+    },
+    {
+      "epoch": 26.36,
+      "learning_rate": 7e-06,
+      "loss": 2.328,
+      "step": 290
+    },
+    {
+      "epoch": 26.36,
+      "eval_loss": 2.3360955715179443,
+      "eval_runtime": 0.732,
+      "eval_samples_per_second": 8.197,
+      "eval_steps_per_second": 1.366,
+      "step": 290
+    },
+    {
+      "epoch": 26.82,
+      "learning_rate": 7e-06,
+      "loss": 2.117,
+      "step": 295
+    },
+    {
+      "epoch": 26.82,
+      "eval_loss": 2.333092451095581,
+      "eval_runtime": 0.7171,
+      "eval_samples_per_second": 8.367,
+      "eval_steps_per_second": 1.395,
+      "step": 295
+    },
+    {
+      "epoch": 27.27,
+      "learning_rate": 7e-06,
+      "loss": 2.2671,
+      "step": 300
+    },
+    {
+      "epoch": 27.27,
+      "eval_loss": 2.330547571182251,
+      "eval_runtime": 0.7166,
+      "eval_samples_per_second": 8.373,
+      "eval_steps_per_second": 1.395,
+      "step": 300
+    }
+  ],
+  "max_steps": 550,
+  "num_train_epochs": 50,
+  "total_flos": 116301968424960.0,
+  "trial_name": null,
+  "trial_params": null
+}

stellaris-races-3/checkpoint-300/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:831e1ea9ed3dc21bd34506e70712bf07dfd13c40b89ff8d8ff382c47a9827312
+size 3963

stellaris-races-3/checkpoint-350/adapter_config.json

@@ -0,0 +1,17 @@
+{
+  "base_model_name_or_path": "",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

stellaris-races-3/checkpoint-350/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4706d782fb4a62b4b2481c7dafb998a1644aa5746519061db1081b38415346e9
+size 104915277

stellaris-races-3/checkpoint-350/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21f94b12156dd5a59e22c2a102b10d9ba1429163fdcb7d7abaecb5ef89c18709
+size 209849861

stellaris-races-3/checkpoint-350/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ee60a3b9e8d10857f086ee8666f0b1f7f51615202c336d7a9c41ba50aa14383
+size 14575

stellaris-races-3/checkpoint-350/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bd6d5c7430823b9274ee0b0e78fa73dcecc95840c4cddb15b48f2dbb5fd1f43
+size 627

stellaris-races-3/checkpoint-350/trainer_state.json

@@ -0,0 +1,996 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 31.818181818181817,
+  "global_step": 350,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.45,
+      "learning_rate": 7e-06,
+      "loss": 2.6772,
+      "step": 5
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 2.7022464275360107,
+      "eval_runtime": 0.7068,
+      "eval_samples_per_second": 8.489,
+      "eval_steps_per_second": 1.415,
+      "step": 5
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 7e-06,
+      "loss": 2.8149,
+      "step": 10
+    },
+    {
+      "epoch": 0.91,
+      "eval_loss": 2.6998846530914307,
+      "eval_runtime": 0.7138,
+      "eval_samples_per_second": 8.406,
+      "eval_steps_per_second": 1.401,
+      "step": 10
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 7e-06,
+      "loss": 2.8494,
+      "step": 15
+    },
+    {
+      "epoch": 1.36,
+      "eval_loss": 2.6970536708831787,
+      "eval_runtime": 0.7302,
+      "eval_samples_per_second": 8.216,
+      "eval_steps_per_second": 1.369,
+      "step": 15
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7e-06,
+      "loss": 2.7035,
+      "step": 20
+    },
+    {
+      "epoch": 1.82,
+      "eval_loss": 2.6939141750335693,
+      "eval_runtime": 0.7129,
+      "eval_samples_per_second": 8.416,
+      "eval_steps_per_second": 1.403,
+      "step": 20
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 7e-06,
+      "loss": 2.5667,
+      "step": 25
+    },
+    {
+      "epoch": 2.27,
+      "eval_loss": 2.6903278827667236,
+      "eval_runtime": 0.7707,
+      "eval_samples_per_second": 7.785,
+      "eval_steps_per_second": 1.298,
+      "step": 25
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 7e-06,
+      "loss": 2.7766,
+      "step": 30
+    },
+    {
+      "epoch": 2.73,
+      "eval_loss": 2.6863253116607666,
+      "eval_runtime": 0.7245,
+      "eval_samples_per_second": 8.281,
+      "eval_steps_per_second": 1.38,
+      "step": 30
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 7e-06,
+      "loss": 2.7507,
+      "step": 35
+    },
+    {
+      "epoch": 3.18,
+      "eval_loss": 2.6823132038116455,
+      "eval_runtime": 0.7678,
+      "eval_samples_per_second": 7.815,
+      "eval_steps_per_second": 1.302,
+      "step": 35
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 7e-06,
+      "loss": 2.7033,
+      "step": 40
+    },
+    {
+      "epoch": 3.64,
+      "eval_loss": 2.677713632583618,
+      "eval_runtime": 0.7464,
+      "eval_samples_per_second": 8.038,
+      "eval_steps_per_second": 1.34,
+      "step": 40
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 7e-06,
+      "loss": 2.714,
+      "step": 45
+    },
+    {
+      "epoch": 4.09,
+      "eval_loss": 2.6728098392486572,
+      "eval_runtime": 0.7176,
+      "eval_samples_per_second": 8.361,
+      "eval_steps_per_second": 1.393,
+      "step": 45
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 7e-06,
+      "loss": 2.7606,
+      "step": 50
+    },
+    {
+      "epoch": 4.55,
+      "eval_loss": 2.6675455570220947,
+      "eval_runtime": 0.7365,
+      "eval_samples_per_second": 8.147,
+      "eval_steps_per_second": 1.358,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 7e-06,
+      "loss": 2.6694,
+      "step": 55
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.6618385314941406,
+      "eval_runtime": 0.7145,
+      "eval_samples_per_second": 8.398,
+      "eval_steps_per_second": 1.4,
+      "step": 55
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 7e-06,
+      "loss": 2.6477,
+      "step": 60
+    },
+    {
+      "epoch": 5.45,
+      "eval_loss": 2.6561715602874756,
+      "eval_runtime": 0.7766,
+      "eval_samples_per_second": 7.726,
+      "eval_steps_per_second": 1.288,
+      "step": 60
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 7e-06,
+      "loss": 2.7613,
+      "step": 65
+    },
+    {
+      "epoch": 5.91,
+      "eval_loss": 2.649792432785034,
+      "eval_runtime": 0.7178,
+      "eval_samples_per_second": 8.359,
+      "eval_steps_per_second": 1.393,
+      "step": 65
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 7e-06,
+      "loss": 2.6446,
+      "step": 70
+    },
+    {
+      "epoch": 6.36,
+      "eval_loss": 2.64331316947937,
+      "eval_runtime": 0.7629,
+      "eval_samples_per_second": 7.865,
+      "eval_steps_per_second": 1.311,
+      "step": 70
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 7e-06,
+      "loss": 2.7039,
+      "step": 75
+    },
+    {
+      "epoch": 6.82,
+      "eval_loss": 2.6366422176361084,
+      "eval_runtime": 0.7438,
+      "eval_samples_per_second": 8.067,
+      "eval_steps_per_second": 1.345,
+      "step": 75
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 7e-06,
+      "loss": 2.5376,
+      "step": 80
+    },
+    {
+      "epoch": 7.27,
+      "eval_loss": 2.6297147274017334,
+      "eval_runtime": 0.7738,
+      "eval_samples_per_second": 7.754,
+      "eval_steps_per_second": 1.292,
+      "step": 80
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 7e-06,
+      "loss": 2.6866,
+      "step": 85
+    },
+    {
+      "epoch": 7.73,
+      "eval_loss": 2.622584581375122,
+      "eval_runtime": 0.7654,
+      "eval_samples_per_second": 7.839,
+      "eval_steps_per_second": 1.307,
+      "step": 85
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 7e-06,
+      "loss": 2.6996,
+      "step": 90
+    },
+    {
+      "epoch": 8.18,
+      "eval_loss": 2.615074396133423,
+      "eval_runtime": 0.7534,
+      "eval_samples_per_second": 7.964,
+      "eval_steps_per_second": 1.327,
+      "step": 90
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 7e-06,
+      "loss": 2.72,
+      "step": 95
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 2.6072492599487305,
+      "eval_runtime": 0.7627,
+      "eval_samples_per_second": 7.867,
+      "eval_steps_per_second": 1.311,
+      "step": 95
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 7e-06,
+      "loss": 2.4708,
+      "step": 100
+    },
+    {
+      "epoch": 9.09,
+      "eval_loss": 2.5988943576812744,
+      "eval_runtime": 0.7612,
+      "eval_samples_per_second": 7.882,
+      "eval_steps_per_second": 1.314,
+      "step": 100
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 7e-06,
+      "loss": 2.5311,
+      "step": 105
+    },
+    {
+      "epoch": 9.55,
+      "eval_loss": 2.5901479721069336,
+      "eval_runtime": 0.7561,
+      "eval_samples_per_second": 7.935,
+      "eval_steps_per_second": 1.322,
+      "step": 105
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7e-06,
+      "loss": 2.6912,
+      "step": 110
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.580698013305664,
+      "eval_runtime": 0.7525,
+      "eval_samples_per_second": 7.973,
+      "eval_steps_per_second": 1.329,
+      "step": 110
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 7e-06,
+      "loss": 2.6089,
+      "step": 115
+    },
+    {
+      "epoch": 10.45,
+      "eval_loss": 2.5717146396636963,
+      "eval_runtime": 0.765,
+      "eval_samples_per_second": 7.843,
+      "eval_steps_per_second": 1.307,
+      "step": 115
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 7e-06,
+      "loss": 2.4803,
+      "step": 120
+    },
+    {
+      "epoch": 10.91,
+      "eval_loss": 2.5612823963165283,
+      "eval_runtime": 0.7202,
+      "eval_samples_per_second": 8.331,
+      "eval_steps_per_second": 1.388,
+      "step": 120
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 7e-06,
+      "loss": 2.6576,
+      "step": 125
+    },
+    {
+      "epoch": 11.36,
+      "eval_loss": 2.5491273403167725,
+      "eval_runtime": 0.7237,
+      "eval_samples_per_second": 8.291,
+      "eval_steps_per_second": 1.382,
+      "step": 125
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 7e-06,
+      "loss": 2.4848,
+      "step": 130
+    },
+    {
+      "epoch": 11.82,
+      "eval_loss": 2.536881446838379,
+      "eval_runtime": 0.749,
+      "eval_samples_per_second": 8.011,
+      "eval_steps_per_second": 1.335,
+      "step": 130
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 7e-06,
+      "loss": 2.4875,
+      "step": 135
+    },
+    {
+      "epoch": 12.27,
+      "eval_loss": 2.524549722671509,
+      "eval_runtime": 0.7273,
+      "eval_samples_per_second": 8.25,
+      "eval_steps_per_second": 1.375,
+      "step": 135
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 7e-06,
+      "loss": 2.5707,
+      "step": 140
+    },
+    {
+      "epoch": 12.73,
+      "eval_loss": 2.511542558670044,
+      "eval_runtime": 0.7341,
+      "eval_samples_per_second": 8.174,
+      "eval_steps_per_second": 1.362,
+      "step": 140
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 7e-06,
+      "loss": 2.3694,
+      "step": 145
+    },
+    {
+      "epoch": 13.18,
+      "eval_loss": 2.4980533123016357,
+      "eval_runtime": 0.733,
+      "eval_samples_per_second": 8.186,
+      "eval_steps_per_second": 1.364,
+      "step": 145
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 7e-06,
+      "loss": 2.515,
+      "step": 150
+    },
+    {
+      "epoch": 13.64,
+      "eval_loss": 2.4852240085601807,
+      "eval_runtime": 0.7212,
+      "eval_samples_per_second": 8.319,
+      "eval_steps_per_second": 1.387,
+      "step": 150
+    },
+    {
+      "epoch": 14.09,
+      "learning_rate": 7e-06,
+      "loss": 2.5051,
+      "step": 155
+    },
+    {
+      "epoch": 14.09,
+      "eval_loss": 2.471349000930786,
+      "eval_runtime": 0.7219,
+      "eval_samples_per_second": 8.311,
+      "eval_steps_per_second": 1.385,
+      "step": 155
+    },
+    {
+      "epoch": 14.55,
+      "learning_rate": 7e-06,
+      "loss": 2.5314,
+      "step": 160
+    },
+    {
+      "epoch": 14.55,
+      "eval_loss": 2.456756830215454,
+      "eval_runtime": 0.7344,
+      "eval_samples_per_second": 8.17,
+      "eval_steps_per_second": 1.362,
+      "step": 160
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3409,
+      "step": 165
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.4444773197174072,
+      "eval_runtime": 0.7207,
+      "eval_samples_per_second": 8.326,
+      "eval_steps_per_second": 1.388,
+      "step": 165
+    },
+    {
+      "epoch": 15.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4272,
+      "step": 170
+    },
+    {
+      "epoch": 15.45,
+      "eval_loss": 2.4356348514556885,
+      "eval_runtime": 0.7258,
+      "eval_samples_per_second": 8.267,
+      "eval_steps_per_second": 1.378,
+      "step": 170
+    },
+    {
+      "epoch": 15.91,
+      "learning_rate": 7e-06,
+      "loss": 2.4189,
+      "step": 175
+    },
+    {
+      "epoch": 15.91,
+      "eval_loss": 2.428696393966675,
+      "eval_runtime": 0.7229,
+      "eval_samples_per_second": 8.3,
+      "eval_steps_per_second": 1.383,
+      "step": 175
+    },
+    {
+      "epoch": 16.36,
+      "learning_rate": 7e-06,
+      "loss": 2.3876,
+      "step": 180
+    },
+    {
+      "epoch": 16.36,
+      "eval_loss": 2.422900915145874,
+      "eval_runtime": 0.7211,
+      "eval_samples_per_second": 8.321,
+      "eval_steps_per_second": 1.387,
+      "step": 180
+    },
+    {
+      "epoch": 16.82,
+      "learning_rate": 7e-06,
+      "loss": 2.3212,
+      "step": 185
+    },
+    {
+      "epoch": 16.82,
+      "eval_loss": 2.4173238277435303,
+      "eval_runtime": 0.72,
+      "eval_samples_per_second": 8.334,
+      "eval_steps_per_second": 1.389,
+      "step": 185
+    },
+    {
+      "epoch": 17.27,
+      "learning_rate": 7e-06,
+      "loss": 2.3173,
+      "step": 190
+    },
+    {
+      "epoch": 17.27,
+      "eval_loss": 2.4116928577423096,
+      "eval_runtime": 0.7204,
+      "eval_samples_per_second": 8.329,
+      "eval_steps_per_second": 1.388,
+      "step": 190
+    },
+    {
+      "epoch": 17.73,
+      "learning_rate": 7e-06,
+      "loss": 2.3327,
+      "step": 195
+    },
+    {
+      "epoch": 17.73,
+      "eval_loss": 2.4064242839813232,
+      "eval_runtime": 0.7198,
+      "eval_samples_per_second": 8.335,
+      "eval_steps_per_second": 1.389,
+      "step": 195
+    },
+    {
+      "epoch": 18.18,
+      "learning_rate": 7e-06,
+      "loss": 2.5256,
+      "step": 200
+    },
+    {
+      "epoch": 18.18,
+      "eval_loss": 2.4015276432037354,
+      "eval_runtime": 0.7429,
+      "eval_samples_per_second": 8.077,
+      "eval_steps_per_second": 1.346,
+      "step": 200
+    },
+    {
+      "epoch": 18.64,
+      "learning_rate": 7e-06,
+      "loss": 2.2825,
+      "step": 205
+    },
+    {
+      "epoch": 18.64,
+      "eval_loss": 2.3970677852630615,
+      "eval_runtime": 0.7175,
+      "eval_samples_per_second": 8.363,
+      "eval_steps_per_second": 1.394,
+      "step": 205
+    },
+    {
+      "epoch": 19.09,
+      "learning_rate": 7e-06,
+      "loss": 2.3417,
+      "step": 210
+    },
+    {
+      "epoch": 19.09,
+      "eval_loss": 2.3925275802612305,
+      "eval_runtime": 0.7168,
+      "eval_samples_per_second": 8.37,
+      "eval_steps_per_second": 1.395,
+      "step": 210
+    },
+    {
+      "epoch": 19.55,
+      "learning_rate": 7e-06,
+      "loss": 2.2123,
+      "step": 215
+    },
+    {
+      "epoch": 19.55,
+      "eval_loss": 2.388213872909546,
+      "eval_runtime": 0.7288,
+      "eval_samples_per_second": 8.232,
+      "eval_steps_per_second": 1.372,
+      "step": 215
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 7e-06,
+      "loss": 2.3696,
+      "step": 220
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 2.384182929992676,
+      "eval_runtime": 0.7182,
+      "eval_samples_per_second": 8.354,
+      "eval_steps_per_second": 1.392,
+      "step": 220
+    },
+    {
+      "epoch": 20.45,
+      "learning_rate": 7e-06,
+      "loss": 2.4359,
+      "step": 225
+    },
+    {
+      "epoch": 20.45,
+      "eval_loss": 2.380094528198242,
+      "eval_runtime": 0.718,
+      "eval_samples_per_second": 8.357,
+      "eval_steps_per_second": 1.393,
+      "step": 225
+    },
+    {
+      "epoch": 20.91,
+      "learning_rate": 7e-06,
+      "loss": 2.1279,
+      "step": 230
+    },
+    {
+      "epoch": 20.91,
+      "eval_loss": 2.3762617111206055,
+      "eval_runtime": 0.7166,
+      "eval_samples_per_second": 8.372,
+      "eval_steps_per_second": 1.395,
+      "step": 230
+    },
+    {
+      "epoch": 21.36,
+      "learning_rate": 7e-06,
+      "loss": 2.3891,
+      "step": 235
+    },
+    {
+      "epoch": 21.36,
+      "eval_loss": 2.3726041316986084,
+      "eval_runtime": 0.7167,
+      "eval_samples_per_second": 8.371,
+      "eval_steps_per_second": 1.395,
+      "step": 235
+    },
+    {
+      "epoch": 21.82,
+      "learning_rate": 7e-06,
+      "loss": 2.1994,
+      "step": 240
+    },
+    {
+      "epoch": 21.82,
+      "eval_loss": 2.36929988861084,
+      "eval_runtime": 0.7208,
+      "eval_samples_per_second": 8.324,
+      "eval_steps_per_second": 1.387,
+      "step": 240
+    },
+    {
+      "epoch": 22.27,
+      "learning_rate": 7e-06,
+      "loss": 2.2041,
+      "step": 245
+    },
+    {
+      "epoch": 22.27,
+      "eval_loss": 2.3658368587493896,
+      "eval_runtime": 0.7161,
+      "eval_samples_per_second": 8.379,
+      "eval_steps_per_second": 1.396,
+      "step": 245
+    },
+    {
+      "epoch": 22.73,
+      "learning_rate": 7e-06,
+      "loss": 2.3645,
+      "step": 250
+    },
+    {
+      "epoch": 22.73,
+      "eval_loss": 2.362556219100952,
+      "eval_runtime": 0.7162,
+      "eval_samples_per_second": 8.378,
+      "eval_steps_per_second": 1.396,
+      "step": 250
+    },
+    {
+      "epoch": 23.18,
+      "learning_rate": 7e-06,
+      "loss": 2.2448,
+      "step": 255
+    },
+    {
+      "epoch": 23.18,
+      "eval_loss": 2.359351873397827,
+      "eval_runtime": 0.7198,
+      "eval_samples_per_second": 8.336,
+      "eval_steps_per_second": 1.389,
+      "step": 255
+    },
+    {
+      "epoch": 23.64,
+      "learning_rate": 7e-06,
+      "loss": 2.1683,
+      "step": 260
+    },
+    {
+      "epoch": 23.64,
+      "eval_loss": 2.3562166690826416,
+      "eval_runtime": 0.7185,
+      "eval_samples_per_second": 8.351,
+      "eval_steps_per_second": 1.392,
+      "step": 260
+    },
+    {
+      "epoch": 24.09,
+      "learning_rate": 7e-06,
+      "loss": 2.34,
+      "step": 265
+    },
+    {
+      "epoch": 24.09,
+      "eval_loss": 2.352797746658325,
+      "eval_runtime": 0.717,
+      "eval_samples_per_second": 8.368,
+      "eval_steps_per_second": 1.395,
+      "step": 265
+    },
+    {
+      "epoch": 24.55,
+      "learning_rate": 7e-06,
+      "loss": 2.2091,
+      "step": 270
+    },
+    {
+      "epoch": 24.55,
+      "eval_loss": 2.349076271057129,
+      "eval_runtime": 0.7168,
+      "eval_samples_per_second": 8.371,
+      "eval_steps_per_second": 1.395,
+      "step": 270
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 7e-06,
+      "loss": 2.2733,
+      "step": 275
+    },
+    {
+      "epoch": 25.0,
+      "eval_loss": 2.3457562923431396,
+      "eval_runtime": 0.716,
+      "eval_samples_per_second": 8.379,
+      "eval_steps_per_second": 1.397,
+      "step": 275
+    },
+    {
+      "epoch": 25.45,
+      "learning_rate": 7e-06,
+      "loss": 2.3123,
+      "step": 280
+    },
+    {
+      "epoch": 25.45,
+      "eval_loss": 2.3424742221832275,
+      "eval_runtime": 0.7162,
+      "eval_samples_per_second": 8.378,
+      "eval_steps_per_second": 1.396,
+      "step": 280
+    },
+    {
+      "epoch": 25.91,
+      "learning_rate": 7e-06,
+      "loss": 2.1236,
+      "step": 285
+    },
+    {
+      "epoch": 25.91,
+      "eval_loss": 2.3390562534332275,
+      "eval_runtime": 0.7167,
+      "eval_samples_per_second": 8.372,
+      "eval_steps_per_second": 1.395,
+      "step": 285
+    },
+    {
+      "epoch": 26.36,
+      "learning_rate": 7e-06,
+      "loss": 2.328,
+      "step": 290
+    },
+    {
+      "epoch": 26.36,
+      "eval_loss": 2.3360955715179443,
+      "eval_runtime": 0.732,
+      "eval_samples_per_second": 8.197,
+      "eval_steps_per_second": 1.366,
+      "step": 290
+    },
+    {
+      "epoch": 26.82,
+      "learning_rate": 7e-06,
+      "loss": 2.117,
+      "step": 295
+    },
+    {
+      "epoch": 26.82,
+      "eval_loss": 2.333092451095581,
+      "eval_runtime": 0.7171,
+      "eval_samples_per_second": 8.367,
+      "eval_steps_per_second": 1.395,
+      "step": 295
+    },
+    {
+      "epoch": 27.27,
+      "learning_rate": 7e-06,
+      "loss": 2.2671,
+      "step": 300
+    },
+    {
+      "epoch": 27.27,
+      "eval_loss": 2.330547571182251,
+      "eval_runtime": 0.7166,
+      "eval_samples_per_second": 8.373,
+      "eval_steps_per_second": 1.395,
+      "step": 300
+    },
+    {
+      "epoch": 27.73,
+      "learning_rate": 7e-06,
+      "loss": 2.1194,
+      "step": 305
+    },
+    {
+      "epoch": 27.73,
+      "eval_loss": 2.327814817428589,
+      "eval_runtime": 0.7216,
+      "eval_samples_per_second": 8.315,
+      "eval_steps_per_second": 1.386,
+      "step": 305
+    },
+    {
+      "epoch": 28.18,
+      "learning_rate": 7e-06,
+      "loss": 2.0954,
+      "step": 310
+    },
+    {
+      "epoch": 28.18,
+      "eval_loss": 2.3255960941314697,
+      "eval_runtime": 0.7172,
+      "eval_samples_per_second": 8.366,
+      "eval_steps_per_second": 1.394,
+      "step": 310
+    },
+    {
+      "epoch": 28.64,
+      "learning_rate": 7e-06,
+      "loss": 2.199,
+      "step": 315
+    },
+    {
+      "epoch": 28.64,
+      "eval_loss": 2.3238537311553955,
+      "eval_runtime": 0.7166,
+      "eval_samples_per_second": 8.373,
+      "eval_steps_per_second": 1.396,
+      "step": 315
+    },
+    {
+      "epoch": 29.09,
+      "learning_rate": 7e-06,
+      "loss": 2.2676,
+      "step": 320
+    },
+    {
+      "epoch": 29.09,
+      "eval_loss": 2.3221495151519775,
+      "eval_runtime": 0.7165,
+      "eval_samples_per_second": 8.374,
+      "eval_steps_per_second": 1.396,
+      "step": 320
+    },
+    {
+      "epoch": 29.55,
+      "learning_rate": 7e-06,
+      "loss": 2.0614,
+      "step": 325
+    },
+    {
+      "epoch": 29.55,
+      "eval_loss": 2.320758104324341,
+      "eval_runtime": 0.7157,
+      "eval_samples_per_second": 8.384,
+      "eval_steps_per_second": 1.397,
+      "step": 325
+    },
+    {
+      "epoch": 30.0,
+      "learning_rate": 7e-06,
+      "loss": 2.1697,
+      "step": 330
+    },
+    {
+      "epoch": 30.0,
+      "eval_loss": 2.319394111633301,
+      "eval_runtime": 0.7375,
+      "eval_samples_per_second": 8.136,
+      "eval_steps_per_second": 1.356,
+      "step": 330
+    },
+    {
+      "epoch": 30.45,
+      "learning_rate": 7e-06,
+      "loss": 2.0884,
+      "step": 335
+    },
+    {
+      "epoch": 30.45,
+      "eval_loss": 2.318929433822632,
+      "eval_runtime": 0.7164,
+      "eval_samples_per_second": 8.376,
+      "eval_steps_per_second": 1.396,
+      "step": 335
+    },
+    {
+      "epoch": 30.91,
+      "learning_rate": 7e-06,
+      "loss": 2.258,
+      "step": 340
+    },
+    {
+      "epoch": 30.91,
+      "eval_loss": 2.3179547786712646,
+      "eval_runtime": 0.7305,
+      "eval_samples_per_second": 8.214,
+      "eval_steps_per_second": 1.369,
+      "step": 340
+    },
+    {
+      "epoch": 31.36,
+      "learning_rate": 7e-06,
+      "loss": 2.0064,
+      "step": 345
+    },
+    {
+      "epoch": 31.36,
+      "eval_loss": 2.3172967433929443,
+      "eval_runtime": 0.7656,
+      "eval_samples_per_second": 7.837,
+      "eval_steps_per_second": 1.306,
+      "step": 345
+    },
+    {
+      "epoch": 31.82,
+      "learning_rate": 7e-06,
+      "loss": 2.152,
+      "step": 350
+    },
+    {
+      "epoch": 31.82,
+      "eval_loss": 2.316732168197632,
+      "eval_runtime": 0.7899,
+      "eval_samples_per_second": 7.596,
+      "eval_steps_per_second": 1.266,
+      "step": 350
+    }
+  ],
+  "max_steps": 550,
+  "num_train_epochs": 50,
+  "total_flos": 135390402693120.0,
+  "trial_name": null,
+  "trial_params": null
+}

stellaris-races-3/checkpoint-350/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:831e1ea9ed3dc21bd34506e70712bf07dfd13c40b89ff8d8ff382c47a9827312
+size 3963

stellaris-races-4/checkpoint-400/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:009a3b2945d4ed2eb37d294137532fc0c9cb5b42af7ec83fb31884961b23c611
+size 209849861

stellaris-races-4/checkpoint-400/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:124f71cc33a87aec1d9822f2176df3f7b3088684e96d44863c02228ba2d2bbf7
+size 104915277

stellaris-races-4/checkpoint-400/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:043999bfe313dcacec24df2c58becfd4411786048602835475edfb4ef409bcf2
+size 14575

stellaris-races-4/checkpoint-400/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fb88d396c1008d5d4603b7db373f6448263ab01ece05d0a23b474b4a6f5bf28
+size 627

stellaris-races-4/checkpoint-400/trainer_state.json

@@ -0,0 +1,1136 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 36.36363636363637,
+  "global_step": 400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.45,
+      "learning_rate": 4e-06,
+      "loss": 2.8337,
+      "step": 5
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 2.7284791469573975,
+      "eval_runtime": 0.7048,
+      "eval_samples_per_second": 8.513,
+      "eval_steps_per_second": 1.419,
+      "step": 5
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4e-06,
+      "loss": 2.6298,
+      "step": 10
+    },
+    {
+      "epoch": 0.91,
+      "eval_loss": 2.7275335788726807,
+      "eval_runtime": 0.7398,
+      "eval_samples_per_second": 8.11,
+      "eval_steps_per_second": 1.352,
+      "step": 10
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 4e-06,
+      "loss": 2.8593,
+      "step": 15
+    },
+    {
+      "epoch": 1.36,
+      "eval_loss": 2.7263944149017334,
+      "eval_runtime": 0.7233,
+      "eval_samples_per_second": 8.295,
+      "eval_steps_per_second": 1.383,
+      "step": 15
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 4e-06,
+      "loss": 2.6932,
+      "step": 20
+    },
+    {
+      "epoch": 1.82,
+      "eval_loss": 2.7251815795898438,
+      "eval_runtime": 0.7434,
+      "eval_samples_per_second": 8.071,
+      "eval_steps_per_second": 1.345,
+      "step": 20
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 4e-06,
+      "loss": 2.7785,
+      "step": 25
+    },
+    {
+      "epoch": 2.27,
+      "eval_loss": 2.7237119674682617,
+      "eval_runtime": 0.7509,
+      "eval_samples_per_second": 7.991,
+      "eval_steps_per_second": 1.332,
+      "step": 25
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4e-06,
+      "loss": 2.7437,
+      "step": 30
+    },
+    {
+      "epoch": 2.73,
+      "eval_loss": 2.722149610519409,
+      "eval_runtime": 0.7048,
+      "eval_samples_per_second": 8.513,
+      "eval_steps_per_second": 1.419,
+      "step": 30
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 4e-06,
+      "loss": 2.7843,
+      "step": 35
+    },
+    {
+      "epoch": 3.18,
+      "eval_loss": 2.7204062938690186,
+      "eval_runtime": 0.7069,
+      "eval_samples_per_second": 8.488,
+      "eval_steps_per_second": 1.415,
+      "step": 35
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 4e-06,
+      "loss": 2.7291,
+      "step": 40
+    },
+    {
+      "epoch": 3.64,
+      "eval_loss": 2.718770980834961,
+      "eval_runtime": 0.7732,
+      "eval_samples_per_second": 7.76,
+      "eval_steps_per_second": 1.293,
+      "step": 40
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 4e-06,
+      "loss": 2.6502,
+      "step": 45
+    },
+    {
+      "epoch": 4.09,
+      "eval_loss": 2.7166388034820557,
+      "eval_runtime": 0.7069,
+      "eval_samples_per_second": 8.488,
+      "eval_steps_per_second": 1.415,
+      "step": 45
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 4e-06,
+      "loss": 2.7704,
+      "step": 50
+    },
+    {
+      "epoch": 4.55,
+      "eval_loss": 2.7146835327148438,
+      "eval_runtime": 0.7575,
+      "eval_samples_per_second": 7.921,
+      "eval_steps_per_second": 1.32,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4e-06,
+      "loss": 2.7618,
+      "step": 55
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.7123842239379883,
+      "eval_runtime": 0.7262,
+      "eval_samples_per_second": 8.262,
+      "eval_steps_per_second": 1.377,
+      "step": 55
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 4e-06,
+      "loss": 2.863,
+      "step": 60
+    },
+    {
+      "epoch": 5.45,
+      "eval_loss": 2.7101686000823975,
+      "eval_runtime": 0.7411,
+      "eval_samples_per_second": 8.096,
+      "eval_steps_per_second": 1.349,
+      "step": 60
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 4e-06,
+      "loss": 2.622,
+      "step": 65
+    },
+    {
+      "epoch": 5.91,
+      "eval_loss": 2.7078440189361572,
+      "eval_runtime": 0.7364,
+      "eval_samples_per_second": 8.148,
+      "eval_steps_per_second": 1.358,
+      "step": 65
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 4e-06,
+      "loss": 2.6697,
+      "step": 70
+    },
+    {
+      "epoch": 6.36,
+      "eval_loss": 2.705427408218384,
+      "eval_runtime": 0.7112,
+      "eval_samples_per_second": 8.437,
+      "eval_steps_per_second": 1.406,
+      "step": 70
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 4e-06,
+      "loss": 2.6311,
+      "step": 75
+    },
+    {
+      "epoch": 6.82,
+      "eval_loss": 2.7028849124908447,
+      "eval_runtime": 0.7132,
+      "eval_samples_per_second": 8.413,
+      "eval_steps_per_second": 1.402,
+      "step": 75
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 4e-06,
+      "loss": 2.8271,
+      "step": 80
+    },
+    {
+      "epoch": 7.27,
+      "eval_loss": 2.700024366378784,
+      "eval_runtime": 0.7334,
+      "eval_samples_per_second": 8.181,
+      "eval_steps_per_second": 1.364,
+      "step": 80
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 4e-06,
+      "loss": 2.7099,
+      "step": 85
+    },
+    {
+      "epoch": 7.73,
+      "eval_loss": 2.6969680786132812,
+      "eval_runtime": 0.7755,
+      "eval_samples_per_second": 7.737,
+      "eval_steps_per_second": 1.289,
+      "step": 85
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 4e-06,
+      "loss": 2.7414,
+      "step": 90
+    },
+    {
+      "epoch": 8.18,
+      "eval_loss": 2.694007158279419,
+      "eval_runtime": 0.7662,
+      "eval_samples_per_second": 7.831,
+      "eval_steps_per_second": 1.305,
+      "step": 90
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 4e-06,
+      "loss": 2.7262,
+      "step": 95
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 2.690988302230835,
+      "eval_runtime": 0.7577,
+      "eval_samples_per_second": 7.919,
+      "eval_steps_per_second": 1.32,
+      "step": 95
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 4e-06,
+      "loss": 2.6965,
+      "step": 100
+    },
+    {
+      "epoch": 9.09,
+      "eval_loss": 2.68784236907959,
+      "eval_runtime": 0.7571,
+      "eval_samples_per_second": 7.925,
+      "eval_steps_per_second": 1.321,
+      "step": 100
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 4e-06,
+      "loss": 2.6607,
+      "step": 105
+    },
+    {
+      "epoch": 9.55,
+      "eval_loss": 2.6844403743743896,
+      "eval_runtime": 0.7536,
+      "eval_samples_per_second": 7.961,
+      "eval_steps_per_second": 1.327,
+      "step": 105
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-06,
+      "loss": 2.7324,
+      "step": 110
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.681208848953247,
+      "eval_runtime": 0.7212,
+      "eval_samples_per_second": 8.32,
+      "eval_steps_per_second": 1.387,
+      "step": 110
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 4e-06,
+      "loss": 2.7323,
+      "step": 115
+    },
+    {
+      "epoch": 10.45,
+      "eval_loss": 2.677574396133423,
+      "eval_runtime": 0.721,
+      "eval_samples_per_second": 8.321,
+      "eval_steps_per_second": 1.387,
+      "step": 115
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 4e-06,
+      "loss": 2.6327,
+      "step": 120
+    },
+    {
+      "epoch": 10.91,
+      "eval_loss": 2.6740548610687256,
+      "eval_runtime": 0.7372,
+      "eval_samples_per_second": 8.139,
+      "eval_steps_per_second": 1.357,
+      "step": 120
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 4e-06,
+      "loss": 2.5854,
+      "step": 125
+    },
+    {
+      "epoch": 11.36,
+      "eval_loss": 2.670290231704712,
+      "eval_runtime": 0.7162,
+      "eval_samples_per_second": 8.377,
+      "eval_steps_per_second": 1.396,
+      "step": 125
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 4e-06,
+      "loss": 2.6755,
+      "step": 130
+    },
+    {
+      "epoch": 11.82,
+      "eval_loss": 2.666276693344116,
+      "eval_runtime": 0.7575,
+      "eval_samples_per_second": 7.921,
+      "eval_steps_per_second": 1.32,
+      "step": 130
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 4e-06,
+      "loss": 2.794,
+      "step": 135
+    },
+    {
+      "epoch": 12.27,
+      "eval_loss": 2.662689447402954,
+      "eval_runtime": 0.7405,
+      "eval_samples_per_second": 8.103,
+      "eval_steps_per_second": 1.35,
+      "step": 135
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 4e-06,
+      "loss": 2.6794,
+      "step": 140
+    },
+    {
+      "epoch": 12.73,
+      "eval_loss": 2.658726453781128,
+      "eval_runtime": 0.7176,
+      "eval_samples_per_second": 8.361,
+      "eval_steps_per_second": 1.394,
+      "step": 140
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 4e-06,
+      "loss": 2.7277,
+      "step": 145
+    },
+    {
+      "epoch": 13.18,
+      "eval_loss": 2.6543197631835938,
+      "eval_runtime": 0.8223,
+      "eval_samples_per_second": 7.297,
+      "eval_steps_per_second": 1.216,
+      "step": 145
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 4e-06,
+      "loss": 2.6594,
+      "step": 150
+    },
+    {
+      "epoch": 13.64,
+      "eval_loss": 2.649885416030884,
+      "eval_runtime": 0.719,
+      "eval_samples_per_second": 8.345,
+      "eval_steps_per_second": 1.391,
+      "step": 150
+    },
+    {
+      "epoch": 14.09,
+      "learning_rate": 4e-06,
+      "loss": 2.613,
+      "step": 155
+    },
+    {
+      "epoch": 14.09,
+      "eval_loss": 2.6451940536499023,
+      "eval_runtime": 0.7287,
+      "eval_samples_per_second": 8.234,
+      "eval_steps_per_second": 1.372,
+      "step": 155
+    },
+    {
+      "epoch": 14.55,
+      "learning_rate": 4e-06,
+      "loss": 2.6155,
+      "step": 160
+    },
+    {
+      "epoch": 14.55,
+      "eval_loss": 2.6406307220458984,
+      "eval_runtime": 0.7309,
+      "eval_samples_per_second": 8.209,
+      "eval_steps_per_second": 1.368,
+      "step": 160
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 4e-06,
+      "loss": 2.6145,
+      "step": 165
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.635648488998413,
+      "eval_runtime": 0.7195,
+      "eval_samples_per_second": 8.339,
+      "eval_steps_per_second": 1.39,
+      "step": 165
+    },
+    {
+      "epoch": 15.45,
+      "learning_rate": 4e-06,
+      "loss": 2.6747,
+      "step": 170
+    },
+    {
+      "epoch": 15.45,
+      "eval_loss": 2.6302928924560547,
+      "eval_runtime": 0.776,
+      "eval_samples_per_second": 7.732,
+      "eval_steps_per_second": 1.289,
+      "step": 170
+    },
+    {
+      "epoch": 15.91,
+      "learning_rate": 4e-06,
+      "loss": 2.5972,
+      "step": 175
+    },
+    {
+      "epoch": 15.91,
+      "eval_loss": 2.62487530708313,
+      "eval_runtime": 0.7538,
+      "eval_samples_per_second": 7.96,
+      "eval_steps_per_second": 1.327,
+      "step": 175
+    },
+    {
+      "epoch": 16.36,
+      "learning_rate": 4e-06,
+      "loss": 2.6962,
+      "step": 180
+    },
+    {
+      "epoch": 16.36,
+      "eval_loss": 2.6190078258514404,
+      "eval_runtime": 0.7203,
+      "eval_samples_per_second": 8.33,
+      "eval_steps_per_second": 1.388,
+      "step": 180
+    },
+    {
+      "epoch": 16.82,
+      "learning_rate": 4e-06,
+      "loss": 2.4865,
+      "step": 185
+    },
+    {
+      "epoch": 16.82,
+      "eval_loss": 2.6129729747772217,
+      "eval_runtime": 0.797,
+      "eval_samples_per_second": 7.529,
+      "eval_steps_per_second": 1.255,
+      "step": 185
+    },
+    {
+      "epoch": 17.27,
+      "learning_rate": 4e-06,
+      "loss": 2.5526,
+      "step": 190
+    },
+    {
+      "epoch": 17.27,
+      "eval_loss": 2.605693817138672,
+      "eval_runtime": 0.7204,
+      "eval_samples_per_second": 8.328,
+      "eval_steps_per_second": 1.388,
+      "step": 190
+    },
+    {
+      "epoch": 17.73,
+      "learning_rate": 4e-06,
+      "loss": 2.6027,
+      "step": 195
+    },
+    {
+      "epoch": 17.73,
+      "eval_loss": 2.598172426223755,
+      "eval_runtime": 0.7191,
+      "eval_samples_per_second": 8.344,
+      "eval_steps_per_second": 1.391,
+      "step": 195
+    },
+    {
+      "epoch": 18.18,
+      "learning_rate": 4e-06,
+      "loss": 2.6925,
+      "step": 200
+    },
+    {
+      "epoch": 18.18,
+      "eval_loss": 2.589960813522339,
+      "eval_runtime": 0.7286,
+      "eval_samples_per_second": 8.235,
+      "eval_steps_per_second": 1.373,
+      "step": 200
+    },
+    {
+      "epoch": 18.64,
+      "learning_rate": 4e-06,
+      "loss": 2.511,
+      "step": 205
+    },
+    {
+      "epoch": 18.64,
+      "eval_loss": 2.5812413692474365,
+      "eval_runtime": 0.7197,
+      "eval_samples_per_second": 8.337,
+      "eval_steps_per_second": 1.39,
+      "step": 205
+    },
+    {
+      "epoch": 19.09,
+      "learning_rate": 4e-06,
+      "loss": 2.5648,
+      "step": 210
+    },
+    {
+      "epoch": 19.09,
+      "eval_loss": 2.572657585144043,
+      "eval_runtime": 0.7846,
+      "eval_samples_per_second": 7.648,
+      "eval_steps_per_second": 1.275,
+      "step": 210
+    },
+    {
+      "epoch": 19.55,
+      "learning_rate": 4e-06,
+      "loss": 2.5478,
+      "step": 215
+    },
+    {
+      "epoch": 19.55,
+      "eval_loss": 2.564173460006714,
+      "eval_runtime": 0.7202,
+      "eval_samples_per_second": 8.331,
+      "eval_steps_per_second": 1.389,
+      "step": 215
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 4e-06,
+      "loss": 2.5774,
+      "step": 220
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 2.5561046600341797,
+      "eval_runtime": 0.7206,
+      "eval_samples_per_second": 8.327,
+      "eval_steps_per_second": 1.388,
+      "step": 220
+    },
+    {
+      "epoch": 20.45,
+      "learning_rate": 4e-06,
+      "loss": 2.4921,
+      "step": 225
+    },
+    {
+      "epoch": 20.45,
+      "eval_loss": 2.548100233078003,
+      "eval_runtime": 0.8137,
+      "eval_samples_per_second": 7.374,
+      "eval_steps_per_second": 1.229,
+      "step": 225
+    },
+    {
+      "epoch": 20.91,
+      "learning_rate": 4e-06,
+      "loss": 2.5361,
+      "step": 230
+    },
+    {
+      "epoch": 20.91,
+      "eval_loss": 2.540703535079956,
+      "eval_runtime": 0.7196,
+      "eval_samples_per_second": 8.338,
+      "eval_steps_per_second": 1.39,
+      "step": 230
+    },
+    {
+      "epoch": 21.36,
+      "learning_rate": 4e-06,
+      "loss": 2.5863,
+      "step": 235
+    },
+    {
+      "epoch": 21.36,
+      "eval_loss": 2.5332720279693604,
+      "eval_runtime": 0.7198,
+      "eval_samples_per_second": 8.336,
+      "eval_steps_per_second": 1.389,
+      "step": 235
+    },
+    {
+      "epoch": 21.82,
+      "learning_rate": 4e-06,
+      "loss": 2.4494,
+      "step": 240
+    },
+    {
+      "epoch": 21.82,
+      "eval_loss": 2.52538800239563,
+      "eval_runtime": 0.7351,
+      "eval_samples_per_second": 8.162,
+      "eval_steps_per_second": 1.36,
+      "step": 240
+    },
+    {
+      "epoch": 22.27,
+      "learning_rate": 4e-06,
+      "loss": 2.4845,
+      "step": 245
+    },
+    {
+      "epoch": 22.27,
+      "eval_loss": 2.5179646015167236,
+      "eval_runtime": 0.7206,
+      "eval_samples_per_second": 8.326,
+      "eval_steps_per_second": 1.388,
+      "step": 245
+    },
+    {
+      "epoch": 22.73,
+      "learning_rate": 4e-06,
+      "loss": 2.4264,
+      "step": 250
+    },
+    {
+      "epoch": 22.73,
+      "eval_loss": 2.5103187561035156,
+      "eval_runtime": 0.7402,
+      "eval_samples_per_second": 8.106,
+      "eval_steps_per_second": 1.351,
+      "step": 250
+    },
+    {
+      "epoch": 23.18,
+      "learning_rate": 4e-06,
+      "loss": 2.5424,
+      "step": 255
+    },
+    {
+      "epoch": 23.18,
+      "eval_loss": 2.502636194229126,
+      "eval_runtime": 0.7202,
+      "eval_samples_per_second": 8.331,
+      "eval_steps_per_second": 1.389,
+      "step": 255
+    },
+    {
+      "epoch": 23.64,
+      "learning_rate": 4e-06,
+      "loss": 2.552,
+      "step": 260
+    },
+    {
+      "epoch": 23.64,
+      "eval_loss": 2.495004892349243,
+      "eval_runtime": 0.7201,
+      "eval_samples_per_second": 8.333,
+      "eval_steps_per_second": 1.389,
+      "step": 260
+    },
+    {
+      "epoch": 24.09,
+      "learning_rate": 4e-06,
+      "loss": 2.3589,
+      "step": 265
+    },
+    {
+      "epoch": 24.09,
+      "eval_loss": 2.487484931945801,
+      "eval_runtime": 0.8136,
+      "eval_samples_per_second": 7.375,
+      "eval_steps_per_second": 1.229,
+      "step": 265
+    },
+    {
+      "epoch": 24.55,
+      "learning_rate": 4e-06,
+      "loss": 2.4911,
+      "step": 270
+    },
+    {
+      "epoch": 24.55,
+      "eval_loss": 2.479656457901001,
+      "eval_runtime": 0.7225,
+      "eval_samples_per_second": 8.305,
+      "eval_steps_per_second": 1.384,
+      "step": 270
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 4e-06,
+      "loss": 2.4487,
+      "step": 275
+    },
+    {
+      "epoch": 25.0,
+      "eval_loss": 2.4720494747161865,
+      "eval_runtime": 0.7181,
+      "eval_samples_per_second": 8.355,
+      "eval_steps_per_second": 1.392,
+      "step": 275
+    },
+    {
+      "epoch": 25.45,
+      "learning_rate": 4e-06,
+      "loss": 2.4137,
+      "step": 280
+    },
+    {
+      "epoch": 25.45,
+      "eval_loss": 2.4644792079925537,
+      "eval_runtime": 0.7389,
+      "eval_samples_per_second": 8.12,
+      "eval_steps_per_second": 1.353,
+      "step": 280
+    },
+    {
+      "epoch": 25.91,
+      "learning_rate": 4e-06,
+      "loss": 2.4539,
+      "step": 285
+    },
+    {
+      "epoch": 25.91,
+      "eval_loss": 2.457874059677124,
+      "eval_runtime": 0.7241,
+      "eval_samples_per_second": 8.286,
+      "eval_steps_per_second": 1.381,
+      "step": 285
+    },
+    {
+      "epoch": 26.36,
+      "learning_rate": 4e-06,
+      "loss": 2.3732,
+      "step": 290
+    },
+    {
+      "epoch": 26.36,
+      "eval_loss": 2.4511220455169678,
+      "eval_runtime": 0.7242,
+      "eval_samples_per_second": 8.285,
+      "eval_steps_per_second": 1.381,
+      "step": 290
+    },
+    {
+      "epoch": 26.82,
+      "learning_rate": 4e-06,
+      "loss": 2.3741,
+      "step": 295
+    },
+    {
+      "epoch": 26.82,
+      "eval_loss": 2.4447500705718994,
+      "eval_runtime": 0.7194,
+      "eval_samples_per_second": 8.341,
+      "eval_steps_per_second": 1.39,
+      "step": 295
+    },
+    {
+      "epoch": 27.27,
+      "learning_rate": 4e-06,
+      "loss": 2.3254,
+      "step": 300
+    },
+    {
+      "epoch": 27.27,
+      "eval_loss": 2.43924880027771,
+      "eval_runtime": 0.7178,
+      "eval_samples_per_second": 8.359,
+      "eval_steps_per_second": 1.393,
+      "step": 300
+    },
+    {
+      "epoch": 27.73,
+      "learning_rate": 4e-06,
+      "loss": 2.4294,
+      "step": 305
+    },
+    {
+      "epoch": 27.73,
+      "eval_loss": 2.4342501163482666,
+      "eval_runtime": 0.81,
+      "eval_samples_per_second": 7.407,
+      "eval_steps_per_second": 1.235,
+      "step": 305
+    },
+    {
+      "epoch": 28.18,
+      "learning_rate": 4e-06,
+      "loss": 2.4231,
+      "step": 310
+    },
+    {
+      "epoch": 28.18,
+      "eval_loss": 2.429851531982422,
+      "eval_runtime": 0.7188,
+      "eval_samples_per_second": 8.347,
+      "eval_steps_per_second": 1.391,
+      "step": 310
+    },
+    {
+      "epoch": 28.64,
+      "learning_rate": 4e-06,
+      "loss": 2.3326,
+      "step": 315
+    },
+    {
+      "epoch": 28.64,
+      "eval_loss": 2.426027536392212,
+      "eval_runtime": 0.7195,
+      "eval_samples_per_second": 8.34,
+      "eval_steps_per_second": 1.39,
+      "step": 315
+    },
+    {
+      "epoch": 29.09,
+      "learning_rate": 4e-06,
+      "loss": 2.3759,
+      "step": 320
+    },
+    {
+      "epoch": 29.09,
+      "eval_loss": 2.422687292098999,
+      "eval_runtime": 0.7851,
+      "eval_samples_per_second": 7.642,
+      "eval_steps_per_second": 1.274,
+      "step": 320
+    },
+    {
+      "epoch": 29.55,
+      "learning_rate": 4e-06,
+      "loss": 2.3845,
+      "step": 325
+    },
+    {
+      "epoch": 29.55,
+      "eval_loss": 2.4192168712615967,
+      "eval_runtime": 0.7186,
+      "eval_samples_per_second": 8.35,
+      "eval_steps_per_second": 1.392,
+      "step": 325
+    },
+    {
+      "epoch": 30.0,
+      "learning_rate": 4e-06,
+      "loss": 2.4086,
+      "step": 330
+    },
+    {
+      "epoch": 30.0,
+      "eval_loss": 2.416382074356079,
+      "eval_runtime": 0.7187,
+      "eval_samples_per_second": 8.349,
+      "eval_steps_per_second": 1.391,
+      "step": 330
+    },
+    {
+      "epoch": 30.45,
+      "learning_rate": 4e-06,
+      "loss": 2.4291,
+      "step": 335
+    },
+    {
+      "epoch": 30.45,
+      "eval_loss": 2.4131243228912354,
+      "eval_runtime": 0.7308,
+      "eval_samples_per_second": 8.21,
+      "eval_steps_per_second": 1.368,
+      "step": 335
+    },
+    {
+      "epoch": 30.91,
+      "learning_rate": 4e-06,
+      "loss": 2.2839,
+      "step": 340
+    },
+    {
+      "epoch": 30.91,
+      "eval_loss": 2.410294532775879,
+      "eval_runtime": 0.7204,
+      "eval_samples_per_second": 8.329,
+      "eval_steps_per_second": 1.388,
+      "step": 340
+    },
+    {
+      "epoch": 31.36,
+      "learning_rate": 4e-06,
+      "loss": 2.3238,
+      "step": 345
+    },
+    {
+      "epoch": 31.36,
+      "eval_loss": 2.407674551010132,
+      "eval_runtime": 0.7732,
+      "eval_samples_per_second": 7.76,
+      "eval_steps_per_second": 1.293,
+      "step": 345
+    },
+    {
+      "epoch": 31.82,
+      "learning_rate": 4e-06,
+      "loss": 2.2245,
+      "step": 350
+    },
+    {
+      "epoch": 31.82,
+      "eval_loss": 2.404945135116577,
+      "eval_runtime": 0.7223,
+      "eval_samples_per_second": 8.306,
+      "eval_steps_per_second": 1.384,
+      "step": 350
+    },
+    {
+      "epoch": 32.27,
+      "learning_rate": 4e-06,
+      "loss": 2.5788,
+      "step": 355
+    },
+    {
+      "epoch": 32.27,
+      "eval_loss": 2.402589797973633,
+      "eval_runtime": 0.7809,
+      "eval_samples_per_second": 7.684,
+      "eval_steps_per_second": 1.281,
+      "step": 355
+    },
+    {
+      "epoch": 32.73,
+      "learning_rate": 4e-06,
+      "loss": 2.3155,
+      "step": 360
+    },
+    {
+      "epoch": 32.73,
+      "eval_loss": 2.3999927043914795,
+      "eval_runtime": 0.7914,
+      "eval_samples_per_second": 7.582,
+      "eval_steps_per_second": 1.264,
+      "step": 360
+    },
+    {
+      "epoch": 33.18,
+      "learning_rate": 4e-06,
+      "loss": 2.3154,
+      "step": 365
+    },
+    {
+      "epoch": 33.18,
+      "eval_loss": 2.3976738452911377,
+      "eval_runtime": 0.718,
+      "eval_samples_per_second": 8.357,
+      "eval_steps_per_second": 1.393,
+      "step": 365
+    },
+    {
+      "epoch": 33.64,
+      "learning_rate": 4e-06,
+      "loss": 2.3466,
+      "step": 370
+    },
+    {
+      "epoch": 33.64,
+      "eval_loss": 2.3953628540039062,
+      "eval_runtime": 0.717,
+      "eval_samples_per_second": 8.369,
+      "eval_steps_per_second": 1.395,
+      "step": 370
+    },
+    {
+      "epoch": 34.09,
+      "learning_rate": 4e-06,
+      "loss": 2.2118,
+      "step": 375
+    },
+    {
+      "epoch": 34.09,
+      "eval_loss": 2.392909288406372,
+      "eval_runtime": 0.7187,
+      "eval_samples_per_second": 8.348,
+      "eval_steps_per_second": 1.391,
+      "step": 375
+    },
+    {
+      "epoch": 34.55,
+      "learning_rate": 4e-06,
+      "loss": 2.2313,
+      "step": 380
+    },
+    {
+      "epoch": 34.55,
+      "eval_loss": 2.3906614780426025,
+      "eval_runtime": 0.7213,
+      "eval_samples_per_second": 8.318,
+      "eval_steps_per_second": 1.386,
+      "step": 380
+    },
+    {
+      "epoch": 35.0,
+      "learning_rate": 4e-06,
+      "loss": 2.3948,
+      "step": 385
+    },
+    {
+      "epoch": 35.0,
+      "eval_loss": 2.388585329055786,
+      "eval_runtime": 0.7204,
+      "eval_samples_per_second": 8.329,
+      "eval_steps_per_second": 1.388,
+      "step": 385
+    },
+    {
+      "epoch": 35.45,
+      "learning_rate": 4e-06,
+      "loss": 2.3867,
+      "step": 390
+    },
+    {
+      "epoch": 35.45,
+      "eval_loss": 2.3863141536712646,
+      "eval_runtime": 0.718,
+      "eval_samples_per_second": 8.357,
+      "eval_steps_per_second": 1.393,
+      "step": 390
+    },
+    {
+      "epoch": 35.91,
+      "learning_rate": 4e-06,
+      "loss": 2.2204,
+      "step": 395
+    },
+    {
+      "epoch": 35.91,
+      "eval_loss": 2.384239912033081,
+      "eval_runtime": 0.7674,
+      "eval_samples_per_second": 7.819,
+      "eval_steps_per_second": 1.303,
+      "step": 395
+    },
+    {
+      "epoch": 36.36,
+      "learning_rate": 4e-06,
+      "loss": 2.3365,
+      "step": 400
+    },
+    {
+      "epoch": 36.36,
+      "eval_loss": 2.38242506980896,
+      "eval_runtime": 0.8118,
+      "eval_samples_per_second": 7.391,
+      "eval_steps_per_second": 1.232,
+      "step": 400
+    }
+  ],
+  "max_steps": 550,
+  "num_train_epochs": 50,
+  "total_flos": 153162314403840.0,
+  "trial_name": null,
+  "trial_params": null
+}

stellaris-races-4/checkpoint-400/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93a3b0f6023b90114718fdb6d05cf7d620abcfa5e56ef74e7c66ae2d006b26f2
+size 3963

stellaris-races-4/checkpoint-450/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:034de5ba5946f249e6a0032f10d200caf745ec2834f6ffa8e77fde5261be3884
+size 209849861