Upload 15 files

phi2-qlora-finetuned/adapter_config.json

@@ -14,19 +14,22 @@
   "loftq_config": {},
   "lora_alpha": 16,
   "lora_bias": false,
-  "lora_dropout": 0.1,
+  "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 64,
+  "r": 16,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "dense",
-    "dense_h_to_4h",
-    "query_key_value",
-    "dense_4h_to_h"
+    "k_proj",
+    "q_proj",
+    "up_proj",
+    "down_proj",
+    "gate_proj",
+    "v_proj",
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

phi2-qlora-finetuned/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd1bf3d294393a980846405d4f01275417a9a4050b44d287df1e53dc20171d26
-size 41951600
+oid sha256:665ed085a1c6782b6f007a8a0d370c75d0888e3672e2dcd344c7f36751aafe53
+size 31483040

phi2-qlora-finetuned/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7e65a6c9ff45185d6c3341bcd6b755b6434d2812a6dfb72451b364a5f9b08a8
-size 83926074
+oid sha256:417c1cd07d456d14577d780ae5bcb4ce553d325313e89cc6ac7f81d51d7891dc
+size 63028090

phi2-qlora-finetuned/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3efacd67464c6f3a31ca4e9c2c7111263cd783b56d7a453ce6846f3ef038256
+oid sha256:7e1e762d3d91354a1077502c44abb720ff6aaa5d1c35eed7a138a66648653703
 size 14244

phi2-qlora-finetuned/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:20196fdcd6e066aadc548dcc3c9ff5244ea31938c846cb03623649ad99fe654f
+oid sha256:578927bb1c801cdd14af41be9b1907db16bd1c7b35d8fc1fc2779e79adb5109e
 size 988

phi2-qlora-finetuned/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95bc4bac3452830d3646edef4935599d16c47da1770ba8eb4f20060808418f56
+oid sha256:34bebae8dbb4044169aedad702183ea2a5f7688635af093bea630f24dc71f1ed
 size 1064

phi2-qlora-finetuned/trainer_state.json

@@ -1,24 +1,112 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.020366598778004074,
+  "epoch": 0.24439918533604887,
   "eval_steps": 500,
-  "global_step": 10,
+  "global_step": 120,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.020366598778004074,
-      "grad_norm": 0.05782133340835571,
-      "learning_rate": 0.0002,
-      "loss": 1.7741,
-      "mean_token_accuracy": 0.607810240983963,
+      "grad_norm": 0.1955823004245758,
+      "learning_rate": 0.00019868265225415265,
+      "loss": 1.6792,
+      "mean_token_accuracy": 0.6286436378955841,
       "step": 10
+    },
+    {
+      "epoch": 0.04073319755600815,
+      "grad_norm": 0.43605607748031616,
+      "learning_rate": 0.00019075754196709572,
+      "loss": 1.5504,
+      "mean_token_accuracy": 0.6437817469239235,
+      "step": 20
+    },
+    {
+      "epoch": 0.06109979633401222,
+      "grad_norm": 0.5884078145027161,
+      "learning_rate": 0.00017621620551276366,
+      "loss": 1.317,
+      "mean_token_accuracy": 0.6953216314315795,
+      "step": 30
+    },
+    {
+      "epoch": 0.0814663951120163,
+      "grad_norm": 0.36727696657180786,
+      "learning_rate": 0.00015611870653623825,
+      "loss": 1.054,
+      "mean_token_accuracy": 0.7708626106381417,
+      "step": 40
+    },
+    {
+      "epoch": 0.10183299389002037,
+      "grad_norm": 0.2848515510559082,
+      "learning_rate": 0.000131930153013598,
+      "loss": 0.6444,
+      "mean_token_accuracy": 0.867109614610672,
+      "step": 50
+    },
+    {
+      "epoch": 0.12219959266802444,
+      "grad_norm": 0.2064630389213562,
+      "learning_rate": 0.00010541389085854176,
+      "loss": 0.9981,
+      "mean_token_accuracy": 0.756032009422779,
+      "step": 60
+    },
+    {
+      "epoch": 0.1425661914460285,
+      "grad_norm": 0.18142254650592804,
+      "learning_rate": 7.85029559788976e-05,
+      "loss": 0.9005,
+      "mean_token_accuracy": 0.775178787112236,
+      "step": 70
+    },
+    {
+      "epoch": 0.1629327902240326,
+      "grad_norm": 0.21299591660499573,
+      "learning_rate": 5.3159155930021e-05,
+      "loss": 0.8883,
+      "mean_token_accuracy": 0.7880317449569703,
+      "step": 80
+    },
+    {
+      "epoch": 0.18329938900203666,
+      "grad_norm": 0.21353664994239807,
+      "learning_rate": 3.123005411465766e-05,
+      "loss": 0.8977,
+      "mean_token_accuracy": 0.801851412653923,
+      "step": 90
+    },
+    {
+      "epoch": 0.20366598778004075,
+      "grad_norm": 0.2567966878414154,
+      "learning_rate": 1.4314282383241096e-05,
+      "loss": 0.5785,
+      "mean_token_accuracy": 0.8794851988554001,
+      "step": 100
+    },
+    {
+      "epoch": 0.2240325865580448,
+      "grad_norm": 0.23000894486904144,
+      "learning_rate": 3.6450007480777093e-06,
+      "loss": 1.0155,
+      "mean_token_accuracy": 0.7514406576752662,
+      "step": 110
+    },
+    {
+      "epoch": 0.24439918533604887,
+      "grad_norm": 0.19669267535209656,
+      "learning_rate": 0.0,
+      "loss": 0.938,
+      "mean_token_accuracy": 0.7690818622708321,
+      "step": 120
     }
   ],
   "logging_steps": 10,
-  "max_steps": 100,
+  "max_steps": 120,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 10,
@@ -29,12 +117,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 1750345677004800.0,
+  "total_flos": 1.339699180744704e+16,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

phi2-qlora-finetuned/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1068cda2d19e4aa77d1a75e6fbdc97a36147805511b2655adb34a475792f849b
+oid sha256:064fa4c2e54fa8a50b7b4d9697c1ac99d08654a2dc35cecd028c114e5f16ce98
 size 5560