Upload folder using huggingface_hub

adapter_config.json

@@ -1,6 +1,6 @@
 {
   "auto_mapping": null,
-  "base_model_name_or_path": "TinyPixel/pythia-1b",
+  "base_model_name_or_path": "EleutherAI/pythia-410m",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b764cefe431eb39be2ea102916fa8d800b347c52d860823f8b22ca08bd55747d
-size 134264397
+oid sha256:83ba5898fc17f29e89e4e62186c37829e2be4928676e3460b0612f936b60503b
+size 100733581

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a269b2a09b370d5ff9b834b7ad50bb4f5b1cc9a4b4f2914cef6c90826618422
-size 268514437
+oid sha256:00196e2e03f69797d92ee6c265ea1f184be227f27c70a771778aff76ebd1235a
+size 201442181

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3b60cfa0010d2dc843abda4bf592429e0146eda3f69771df5e72928c1bd85f1
+oid sha256:4babfd4a3d3b22e6223f179a08b4cfa2ae518b4d44d2c70a19f39410076f73f0
 size 14575

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a03bd55ec967989abd72e176bb55c4192fe56bafe1ccfdaea2dc4dd09d252069
+oid sha256:39f44b759122c3c879d266cd63b8d0b929867b61b815abaa9f3e43a6f98624ee
 size 627

special_tokens_map.json

@@ -1,12 +1,11 @@
 {
   "additional_special_tokens": [
-    "<|end_of_turn|>",
-    "<|start_of_turn|>",
-    "### Human:",
-    "### Assistant:"
+    "<|prompter|>",
+    "<|assistant|>",
+    "<|end|>"
   ],
   "bos_token": "<|endoftext|>",
   "eos_token": "<|endoftext|>",
-  "pad_token": "[PAD]",
+  "pad_token": "<|endoftext|>",
   "unk_token": "<|endoftext|>"
 }

tokenizer.json

@@ -235,7 +235,7 @@
     },
     {
       "id": 50277,
-      "content": "PROMPT:",
+      "content": "<|prompter|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -244,7 +244,7 @@
     },
     {
       "id": 50278,
-      "content": "ANSWER:",
+      "content": "<|assistant|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -253,61 +253,7 @@
     },
     {
       "id": 50279,
-      "content": "<|im_start|>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 50280,
-      "content": "<|im_end|>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 50281,
-      "content": "<|end_of_turn|>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 50282,
-      "content": "[PAD]",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 50283,
-      "content": "<|start_of_turn|>",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 50284,
-      "content": "### Human:",
-      "single_word": false,
-      "lstrip": false,
-      "rstrip": false,
-      "normalized": false,
-      "special": true
-    },
-    {
-      "id": 50285,
-      "content": "### Assistant:",
+      "content": "<|end|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,

trainer_state.json

@@ -1,319 +1,403 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.16250253910217347,
+  "epoch": 1.9883495145631067,
   "eval_steps": 500,
-  "global_step": 100,
+  "global_step": 128,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.0,
+      "epoch": 0.03,
       "learning_rate": 2e-05,
-      "loss": 1.7252,
+      "loss": 2.5792,
       "step": 2
     },
     {
-      "epoch": 0.01,
+      "epoch": 0.06,
       "learning_rate": 2e-05,
-      "loss": 1.8477,
+      "loss": 2.816,
       "step": 4
     },
     {
-      "epoch": 0.01,
+      "epoch": 0.09,
       "learning_rate": 2e-05,
-      "loss": 1.6071,
+      "loss": 2.7111,
       "step": 6
     },
     {
-      "epoch": 0.01,
+      "epoch": 0.12,
       "learning_rate": 2e-05,
-      "loss": 2.0372,
+      "loss": 2.8995,
       "step": 8
     },
     {
-      "epoch": 0.02,
+      "epoch": 0.16,
       "learning_rate": 2e-05,
-      "loss": 1.9284,
+      "loss": 2.9567,
       "step": 10
     },
     {
-      "epoch": 0.02,
+      "epoch": 0.19,
       "learning_rate": 2e-05,
-      "loss": 2.0431,
+      "loss": 3.1137,
       "step": 12
     },
     {
-      "epoch": 0.02,
+      "epoch": 0.22,
       "learning_rate": 2e-05,
-      "loss": 2.1371,
+      "loss": 3.0994,
       "step": 14
     },
     {
-      "epoch": 0.03,
+      "epoch": 0.25,
       "learning_rate": 2e-05,
-      "loss": 1.9433,
+      "loss": 3.9126,
       "step": 16
     },
     {
-      "epoch": 0.03,
+      "epoch": 0.28,
       "learning_rate": 2e-05,
-      "loss": 1.987,
+      "loss": 2.5394,
       "step": 18
     },
     {
-      "epoch": 0.03,
+      "epoch": 0.31,
       "learning_rate": 2e-05,
-      "loss": 2.1397,
+      "loss": 2.6201,
       "step": 20
     },
     {
-      "epoch": 0.04,
+      "epoch": 0.34,
       "learning_rate": 2e-05,
-      "loss": 2.0042,
+      "loss": 2.6232,
       "step": 22
     },
     {
-      "epoch": 0.04,
+      "epoch": 0.37,
       "learning_rate": 2e-05,
-      "loss": 2.0193,
+      "loss": 2.9296,
       "step": 24
     },
     {
-      "epoch": 0.04,
+      "epoch": 0.4,
       "learning_rate": 2e-05,
-      "loss": 2.1728,
+      "loss": 2.9391,
       "step": 26
     },
     {
-      "epoch": 0.05,
+      "epoch": 0.43,
       "learning_rate": 2e-05,
-      "loss": 2.4132,
+      "loss": 3.1474,
       "step": 28
     },
     {
-      "epoch": 0.05,
+      "epoch": 0.47,
       "learning_rate": 2e-05,
-      "loss": 2.2364,
+      "loss": 3.1631,
       "step": 30
     },
     {
-      "epoch": 0.05,
+      "epoch": 0.5,
       "learning_rate": 2e-05,
-      "loss": 2.3455,
+      "loss": 4.1919,
       "step": 32
     },
     {
-      "epoch": 0.06,
+      "epoch": 0.53,
       "learning_rate": 2e-05,
-      "loss": 2.3668,
+      "loss": 2.5069,
       "step": 34
     },
     {
-      "epoch": 0.06,
+      "epoch": 0.56,
       "learning_rate": 2e-05,
-      "loss": 2.3953,
+      "loss": 2.771,
       "step": 36
     },
     {
-      "epoch": 0.06,
+      "epoch": 0.59,
       "learning_rate": 2e-05,
-      "loss": 2.5691,
+      "loss": 2.8112,
       "step": 38
     },
     {
-      "epoch": 0.07,
+      "epoch": 0.62,
       "learning_rate": 2e-05,
-      "loss": 2.685,
+      "loss": 2.8988,
       "step": 40
     },
     {
-      "epoch": 0.07,
+      "epoch": 0.65,
       "learning_rate": 2e-05,
-      "loss": 2.7927,
+      "loss": 2.994,
       "step": 42
     },
     {
-      "epoch": 0.07,
+      "epoch": 0.68,
       "learning_rate": 2e-05,
-      "loss": 2.9057,
+      "loss": 2.94,
       "step": 44
     },
     {
-      "epoch": 0.07,
+      "epoch": 0.71,
       "learning_rate": 2e-05,
-      "loss": 3.2075,
+      "loss": 3.0248,
       "step": 46
     },
     {
-      "epoch": 0.08,
+      "epoch": 0.75,
       "learning_rate": 2e-05,
-      "loss": 3.5339,
+      "loss": 3.7811,
       "step": 48
     },
     {
-      "epoch": 0.08,
+      "epoch": 0.78,
       "learning_rate": 2e-05,
-      "loss": 4.2786,
+      "loss": 2.5554,
       "step": 50
     },
     {
-      "epoch": 0.08,
+      "epoch": 0.81,
       "learning_rate": 2e-05,
-      "loss": 1.7911,
+      "loss": 2.656,
       "step": 52
     },
     {
-      "epoch": 0.09,
+      "epoch": 0.84,
       "learning_rate": 2e-05,
-      "loss": 1.6751,
+      "loss": 2.7889,
       "step": 54
     },
     {
-      "epoch": 0.09,
+      "epoch": 0.87,
       "learning_rate": 2e-05,
-      "loss": 1.7581,
+      "loss": 2.7613,
       "step": 56
     },
     {
-      "epoch": 0.09,
+      "epoch": 0.9,
       "learning_rate": 2e-05,
-      "loss": 1.783,
+      "loss": 2.8959,
       "step": 58
     },
     {
-      "epoch": 0.1,
+      "epoch": 0.93,
       "learning_rate": 2e-05,
-      "loss": 1.8982,
+      "loss": 2.914,
       "step": 60
     },
     {
-      "epoch": 0.1,
+      "epoch": 0.96,
       "learning_rate": 2e-05,
-      "loss": 1.926,
+      "loss": 3.1729,
       "step": 62
     },
     {
-      "epoch": 0.1,
+      "epoch": 0.99,
       "learning_rate": 2e-05,
-      "loss": 1.8847,
+      "loss": 3.6644,
       "step": 64
     },
     {
-      "epoch": 0.11,
+      "epoch": 1.03,
       "learning_rate": 2e-05,
-      "loss": 2.0012,
+      "loss": 2.5554,
       "step": 66
     },
     {
-      "epoch": 0.11,
+      "epoch": 1.06,
       "learning_rate": 2e-05,
-      "loss": 2.0488,
+      "loss": 2.5432,
       "step": 68
     },
     {
-      "epoch": 0.11,
+      "epoch": 1.09,
       "learning_rate": 2e-05,
-      "loss": 2.042,
+      "loss": 2.6235,
       "step": 70
     },
     {
-      "epoch": 0.12,
+      "epoch": 1.12,
       "learning_rate": 2e-05,
-      "loss": 2.0934,
+      "loss": 2.7702,
       "step": 72
     },
     {
-      "epoch": 0.12,
+      "epoch": 1.15,
       "learning_rate": 2e-05,
-      "loss": 1.9756,
+      "loss": 2.943,
       "step": 74
     },
     {
-      "epoch": 0.12,
+      "epoch": 1.18,
       "learning_rate": 2e-05,
-      "loss": 1.8503,
+      "loss": 2.8071,
       "step": 76
     },
     {
-      "epoch": 0.13,
+      "epoch": 1.21,
       "learning_rate": 2e-05,
-      "loss": 2.1736,
+      "loss": 3.1157,
       "step": 78
     },
     {
-      "epoch": 0.13,
+      "epoch": 1.24,
       "learning_rate": 2e-05,
-      "loss": 2.089,
+      "loss": 3.4279,
       "step": 80
     },
     {
-      "epoch": 0.13,
+      "epoch": 1.27,
       "learning_rate": 2e-05,
-      "loss": 2.3295,
+      "loss": 2.8172,
       "step": 82
     },
     {
-      "epoch": 0.14,
+      "epoch": 1.3,
       "learning_rate": 2e-05,
-      "loss": 2.2796,
+      "loss": 2.4719,
       "step": 84
     },
     {
-      "epoch": 0.14,
+      "epoch": 1.34,
       "learning_rate": 2e-05,
-      "loss": 2.2786,
+      "loss": 2.6784,
       "step": 86
     },
     {
-      "epoch": 0.14,
+      "epoch": 1.37,
       "learning_rate": 2e-05,
-      "loss": 2.2979,
+      "loss": 2.7028,
       "step": 88
     },
     {
-      "epoch": 0.15,
+      "epoch": 1.4,
       "learning_rate": 2e-05,
-      "loss": 2.5102,
+      "loss": 2.8362,
       "step": 90
     },
     {
-      "epoch": 0.15,
+      "epoch": 1.43,
       "learning_rate": 2e-05,
-      "loss": 2.6685,
+      "loss": 2.8774,
       "step": 92
     },
     {
-      "epoch": 0.15,
+      "epoch": 1.46,
       "learning_rate": 2e-05,
-      "loss": 2.7812,
+      "loss": 2.9223,
       "step": 94
     },
     {
-      "epoch": 0.16,
+      "epoch": 1.49,
       "learning_rate": 2e-05,
-      "loss": 2.943,
+      "loss": 3.4649,
       "step": 96
     },
     {
-      "epoch": 0.16,
+      "epoch": 1.52,
       "learning_rate": 2e-05,
-      "loss": 2.9432,
+      "loss": 2.9887,
       "step": 98
     },
     {
-      "epoch": 0.16,
+      "epoch": 1.55,
       "learning_rate": 2e-05,
-      "loss": 3.6691,
+      "loss": 2.4352,
       "step": 100
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 2e-05,
+      "loss": 2.6306,
+      "step": 102
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 2e-05,
+      "loss": 2.7199,
+      "step": 104
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 2e-05,
+      "loss": 2.8404,
+      "step": 106
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 2e-05,
+      "loss": 2.9678,
+      "step": 108
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2e-05,
+      "loss": 2.987,
+      "step": 110
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2e-05,
+      "loss": 3.3557,
+      "step": 112
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2e-05,
+      "loss": 2.9119,
+      "step": 114
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2e-05,
+      "loss": 2.6182,
+      "step": 116
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 2e-05,
+      "loss": 2.7202,
+      "step": 118
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 2e-05,
+      "loss": 2.8067,
+      "step": 120
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2e-05,
+      "loss": 2.7643,
+      "step": 122
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 2e-05,
+      "loss": 2.8363,
+      "step": 124
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 2e-05,
+      "loss": 3.0004,
+      "step": 126
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 2e-05,
+      "loss": 3.1952,
+      "step": 128
     }
   ],
   "logging_steps": 2,
-  "max_steps": 1845,
-  "num_train_epochs": 3,
-  "save_steps": 50,
-  "total_flos": 2179598319624192.0,
+  "max_steps": 640,
+  "num_train_epochs": 10,
+  "save_steps": 64,
+  "total_flos": 1466230489497600.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:174c8bffd086ebc6d1aa8c350cefa932511681b2eff2b9dd236c0f93f68017d2
+oid sha256:56858e9a674ce140574860fbcec77a85eb683295eeceb03666fa0328f632f049
 size 4027