Upload folder using huggingface_hub

README.md

@@ -201,4 +201,4 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
 
 ### Framework versions
 
-- PEFT 0.8.2.dev0
+- PEFT 0.8.2

adapter_config.json

@@ -20,12 +20,12 @@
   "revision": null,
   "target_modules": [
     "up_proj",
-    "gate_proj",
+    "down_proj",
     "v_proj",
-    "q_proj",
-    "k_proj",
+    "gate_proj",
     "o_proj",
-    "down_proj"
+    "k_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a111c948bdfc4e0385c1cefccdad91a853056d6403f1cc698b68117ddc5fe324
+oid sha256:8f64bed87df2db7786b5d61c82db610e2a7202aae62276222b6ac573914b823f
 size 276869160

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1730dc203bcae94b33bc3dcbf9a722804120e24cbe9c1029ec1450ca0880bec3
+oid sha256:15b2812c80d28fe92fd38c445c8df351ef14d2298f1058aba8d344b1b9bdf179
 size 553854714

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d839c288abf2b3689142bcb7f27a1ac0e84df0da29746c6b101fea61b5c09bf9
+oid sha256:5f8c50f43adcf0bbe157bc7fb4e1945303f60cacb4cde045807899a4e4b00f1b
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:125d43444f1813249da3606d71ed977ab54afa10359eef2e59ffd6fef0ad576c
+oid sha256:d7b626b01aeed2512c1536f9c6e3c1741564cc725165563957c4329b855d633e
 size 1064

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.953846153846154,
+  "epoch": 0.9846153846153847,
   "eval_steps": 500,
-  "global_step": 144,
+  "global_step": 48,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -11,442 +11,154 @@
     {
       "epoch": 0.04,
       "learning_rate": 0.0,
-      "loss": 1.9175,
+      "loss": 1.905,
       "step": 2
     },
     {
       "epoch": 0.08,
       "learning_rate": 0.0,
-      "loss": 2.1462,
+      "loss": 2.1079,
       "step": 4
     },
     {
       "epoch": 0.12,
       "learning_rate": 0.0,
-      "loss": 2.2927,
+      "loss": 2.3465,
       "step": 6
     },
     {
       "epoch": 0.16,
       "learning_rate": 0.0,
-      "loss": 2.2939,
+      "loss": 2.3622,
       "step": 8
     },
     {
       "epoch": 0.21,
-      "learning_rate": 8.000000000000001e-06,
-      "loss": 2.3964,
+      "learning_rate": 0.0,
+      "loss": 2.448,
       "step": 10
     },
     {
       "epoch": 0.25,
-      "learning_rate": 1.6000000000000003e-05,
-      "loss": 2.5863,
+      "learning_rate": 0.0,
+      "loss": 2.6364,
       "step": 12
     },
     {
       "epoch": 0.29,
-      "learning_rate": 1.999744599547812e-05,
-      "loss": 1.8329,
+      "learning_rate": 2e-05,
+      "loss": 1.9455,
       "step": 14
     },
     {
       "epoch": 0.33,
-      "learning_rate": 1.99770217861636e-05,
-      "loss": 2.1592,
+      "learning_rate": 1.9906859460363307e-05,
+      "loss": 2.179,
       "step": 16
     },
     {
       "epoch": 0.37,
-      "learning_rate": 1.9936215093023884e-05,
-      "loss": 2.2509,
+      "learning_rate": 1.9629172873477995e-05,
+      "loss": 2.2703,
       "step": 18
     },
     {
       "epoch": 0.41,
-      "learning_rate": 1.9875109281794828e-05,
-      "loss": 2.1665,
+      "learning_rate": 1.917211301505453e-05,
+      "loss": 2.2115,
       "step": 20
     },
     {
       "epoch": 0.45,
-      "learning_rate": 1.9793829188147406e-05,
-      "loss": 2.3913,
+      "learning_rate": 1.8544194045464888e-05,
+      "loss": 2.3446,
       "step": 22
     },
     {
       "epoch": 0.49,
-      "learning_rate": 1.9692540862655587e-05,
-      "loss": 2.5814,
+      "learning_rate": 1.77571129070442e-05,
+      "loss": 2.4233,
       "step": 24
     },
     {
       "epoch": 0.53,
-      "learning_rate": 1.9571451231564523e-05,
-      "loss": 1.8951,
+      "learning_rate": 1.6825531432186545e-05,
+      "loss": 1.9121,
       "step": 26
     },
     {
       "epoch": 0.57,
-      "learning_rate": 1.9430807674052092e-05,
-      "loss": 2.1537,
+      "learning_rate": 1.5766803221148676e-05,
+      "loss": 2.2741,
       "step": 28
     },
     {
       "epoch": 0.62,
-      "learning_rate": 1.9270897516847406e-05,
-      "loss": 2.1855,
+      "learning_rate": 1.4600650377311523e-05,
+      "loss": 2.1354,
       "step": 30
     },
     {
       "epoch": 0.66,
-      "learning_rate": 1.9092047447238775e-05,
-      "loss": 2.1352,
+      "learning_rate": 1.3348796121709862e-05,
+      "loss": 2.1284,
       "step": 32
     },
     {
       "epoch": 0.7,
-      "learning_rate": 1.8894622845670282e-05,
-      "loss": 2.1292,
+      "learning_rate": 1.2034560130526341e-05,
+      "loss": 2.0004,
       "step": 34
     },
     {
       "epoch": 0.74,
-      "learning_rate": 1.86790270392905e-05,
-      "loss": 2.3216,
+      "learning_rate": 1.0682424133646712e-05,
+      "loss": 2.1918,
       "step": 36
     },
     {
       "epoch": 0.78,
-      "learning_rate": 1.8445700477978207e-05,
-      "loss": 1.8196,
+      "learning_rate": 9.317575866353293e-06,
+      "loss": 1.9007,
       "step": 38
     },
     {
       "epoch": 0.82,
-      "learning_rate": 1.8195119834528535e-05,
-      "loss": 2.123,
+      "learning_rate": 7.965439869473664e-06,
+      "loss": 2.1758,
       "step": 40
     },
     {
       "epoch": 0.86,
-      "learning_rate": 1.792779703083777e-05,
-      "loss": 2.17,
+      "learning_rate": 6.651203878290139e-06,
+      "loss": 2.2133,
       "step": 42
     },
     {
       "epoch": 0.9,
-      "learning_rate": 1.764427819207624e-05,
-      "loss": 2.1112,
+      "learning_rate": 5.399349622688479e-06,
+      "loss": 2.1743,
       "step": 44
     },
     {
       "epoch": 0.94,
-      "learning_rate": 1.734514253098589e-05,
-      "loss": 2.101,
+      "learning_rate": 4.2331967788513295e-06,
+      "loss": 2.2372,
       "step": 46
     },
     {
       "epoch": 0.98,
-      "learning_rate": 1.7031001164581828e-05,
-      "loss": 2.2077,
+      "learning_rate": 3.174468567813461e-06,
+      "loss": 2.4227,
       "step": 48
-    },
-    {
-      "epoch": 1.03,
-      "learning_rate": 1.670249586567531e-05,
-      "loss": 1.9385,
-      "step": 50
-    },
-    {
-      "epoch": 1.07,
-      "learning_rate": 1.636029775176862e-05,
-      "loss": 2.0721,
-      "step": 52
-    },
-    {
-      "epoch": 1.11,
-      "learning_rate": 1.6005105914000508e-05,
-      "loss": 2.1402,
-      "step": 54
-    },
-    {
-      "epoch": 1.15,
-      "learning_rate": 1.5637645988943008e-05,
-      "loss": 2.3016,
-      "step": 56
-    },
-    {
-      "epoch": 1.19,
-      "learning_rate": 1.5258668676167548e-05,
-      "loss": 2.1148,
-      "step": 58
-    },
-    {
-      "epoch": 1.23,
-      "learning_rate": 1.48689482046087e-05,
-      "loss": 2.1329,
-      "step": 60
-    },
-    {
-      "epoch": 1.27,
-      "learning_rate": 1.4469280750858854e-05,
-      "loss": 2.0325,
-      "step": 62
-    },
-    {
-      "epoch": 1.31,
-      "learning_rate": 1.4060482812625055e-05,
-      "loss": 2.0027,
-      "step": 64
-    },
-    {
-      "epoch": 1.35,
-      "learning_rate": 1.3643389540670963e-05,
-      "loss": 2.067,
-      "step": 66
-    },
-    {
-      "epoch": 1.39,
-      "learning_rate": 1.3218853032651719e-05,
-      "loss": 1.9725,
-      "step": 68
-    },
-    {
-      "epoch": 1.44,
-      "learning_rate": 1.2787740592327232e-05,
-      "loss": 2.1139,
-      "step": 70
-    },
-    {
-      "epoch": 1.48,
-      "learning_rate": 1.2350932957710322e-05,
-      "loss": 2.1837,
-      "step": 72
-    },
-    {
-      "epoch": 1.52,
-      "learning_rate": 1.1909322501769407e-05,
-      "loss": 1.956,
-      "step": 74
-    },
-    {
-      "epoch": 1.56,
-      "learning_rate": 1.1463811409361667e-05,
-      "loss": 1.8643,
-      "step": 76
-    },
-    {
-      "epoch": 1.6,
-      "learning_rate": 1.1015309834121083e-05,
-      "loss": 2.1062,
-      "step": 78
-    },
-    {
-      "epoch": 1.64,
-      "learning_rate": 1.05647340390667e-05,
-      "loss": 2.1883,
-      "step": 80
-    },
-    {
-      "epoch": 1.68,
-      "learning_rate": 1.01130045247298e-05,
-      "loss": 1.999,
-      "step": 82
-    },
-    {
-      "epoch": 1.72,
-      "learning_rate": 9.661044148624038e-06,
-      "loss": 2.0649,
-      "step": 84
-    },
-    {
-      "epoch": 1.76,
-      "learning_rate": 9.209776239900453e-06,
-      "loss": 1.7508,
-      "step": 86
-    },
-    {
-      "epoch": 1.81,
-      "learning_rate": 8.76012271303888e-06,
-      "loss": 2.0394,
-      "step": 88
-    },
-    {
-      "epoch": 1.85,
-      "learning_rate": 8.313002184429529e-06,
-      "loss": 2.0459,
-      "step": 90
-    },
-    {
-      "epoch": 1.89,
-      "learning_rate": 7.869328095692313e-06,
-      "loss": 2.1346,
-      "step": 92
-    },
-    {
-      "epoch": 1.93,
-      "learning_rate": 7.430006847567972e-06,
-      "loss": 2.1252,
-      "step": 94
-    },
-    {
-      "epoch": 1.97,
-      "learning_rate": 6.995935948193294e-06,
-      "loss": 2.0431,
-      "step": 96
-    },
-    {
-      "epoch": 2.01,
-      "learning_rate": 6.568002179543409e-06,
-      "loss": 2.1634,
-      "step": 98
-    },
-    {
-      "epoch": 2.05,
-      "learning_rate": 6.147079785787038e-06,
-      "loss": 1.8052,
-      "step": 100
-    },
-    {
-      "epoch": 2.09,
-      "learning_rate": 5.7340286872557515e-06,
-      "loss": 2.0465,
-      "step": 102
-    },
-    {
-      "epoch": 2.13,
-      "learning_rate": 5.329692723675994e-06,
-      "loss": 2.0533,
-      "step": 104
-    },
-    {
-      "epoch": 2.17,
-      "learning_rate": 4.934897930252887e-06,
-      "loss": 2.1753,
-      "step": 106
-    },
-    {
-      "epoch": 2.22,
-      "learning_rate": 4.550450850127626e-06,
-      "loss": 2.1203,
-      "step": 108
-    },
-    {
-      "epoch": 2.26,
-      "learning_rate": 4.177136886656067e-06,
-      "loss": 1.9635,
-      "step": 110
-    },
-    {
-      "epoch": 2.3,
-      "learning_rate": 3.815718698874672e-06,
-      "loss": 1.9614,
-      "step": 112
-    },
-    {
-      "epoch": 2.34,
-      "learning_rate": 3.466934643431795e-06,
-      "loss": 2.0104,
-      "step": 114
-    },
-    {
-      "epoch": 2.38,
-      "learning_rate": 3.1314972661673572e-06,
-      "loss": 2.0814,
-      "step": 116
-    },
-    {
-      "epoch": 2.42,
-      "learning_rate": 2.8100918464225304e-06,
-      "loss": 2.1685,
-      "step": 118
-    },
-    {
-      "epoch": 2.46,
-      "learning_rate": 2.5033749970533015e-06,
-      "loss": 1.914,
-      "step": 120
-    },
-    {
-      "epoch": 2.5,
-      "learning_rate": 2.211973323008041e-06,
-      "loss": 2.0804,
-      "step": 122
-    },
-    {
-      "epoch": 2.54,
-      "learning_rate": 1.936482141209486e-06,
-      "loss": 1.9071,
-      "step": 124
-    },
-    {
-      "epoch": 2.58,
-      "learning_rate": 1.6774642643563955e-06,
-      "loss": 2.1307,
-      "step": 126
-    },
-    {
-      "epoch": 2.63,
-      "learning_rate": 1.4354488511294418e-06,
-      "loss": 2.1478,
-      "step": 128
-    },
-    {
-      "epoch": 2.67,
-      "learning_rate": 1.2109303251503434e-06,
-      "loss": 1.9693,
-      "step": 130
-    },
-    {
-      "epoch": 2.71,
-      "learning_rate": 1.0043673649027519e-06,
-      "loss": 1.9913,
-      "step": 132
-    },
-    {
-      "epoch": 2.75,
-      "learning_rate": 8.161819666783888e-07,
-      "loss": 1.9936,
-      "step": 134
-    },
-    {
-      "epoch": 2.79,
-      "learning_rate": 6.467585824627886e-07,
-      "loss": 1.8671,
-      "step": 136
-    },
-    {
-      "epoch": 2.83,
-      "learning_rate": 4.964433345219354e-07,
-      "loss": 2.1326,
-      "step": 138
-    },
-    {
-      "epoch": 2.87,
-      "learning_rate": 3.6554330829429716e-07,
-      "loss": 2.1233,
-      "step": 140
-    },
-    {
-      "epoch": 2.91,
-      "learning_rate": 2.5432592503288e-07,
-      "loss": 2.0922,
-      "step": 142
-    },
-    {
-      "epoch": 2.95,
-      "learning_rate": 1.630183954789233e-07,
-      "loss": 2.0815,
-      "step": 144
     }
   ],
   "logging_steps": 2,
-  "max_steps": 144,
+  "max_steps": 48,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 1,
   "save_steps": 500,
-  "total_flos": 1.2751629349782528e+16,
+  "total_flos": 4314823005533184.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef422f322db7c57192c86b04e279604e2fec5f74b9f7720451b6bcd47fa7d026
+oid sha256:01617994a7d97f026776e87200feaef2d9ea5a15af1993988df5e127ab3a2e36
 size 4728