starlineventures/chatgpt

adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "up_proj",
-    "q_proj",
+    "down_proj",
     "o_proj",
+    "q_proj",
     "v_proj",
+    "k_proj",
     "gate_proj",
-    "down_proj"
+    "up_proj"
   ],
   "task_type": null,
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fff7b051d12ef77bb0c869afde32d845678781cf3b5494203b66572011f998b2
+oid sha256:86ecc4ebb035abab1d5b06ebbb895750d78a6b579cadc196c3312310b4fc1bb3
 size 35668592

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "total_flos": 0.0,
-    "train_loss": 0.4158199636959562,
-    "train_runtime": 99.4711,
-    "train_samples_per_second": 6.122,
-    "train_steps_per_second": 2.051
+    "train_loss": 0.41664891704624774,
+    "train_runtime": 99.7082,
+    "train_samples_per_second": 6.108,
+    "train_steps_per_second": 2.046
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "total_flos": 0.0,
-    "train_loss": 0.4158199636959562,
-    "train_runtime": 99.4711,
-    "train_samples_per_second": 6.122,
-    "train_steps_per_second": 2.051
+    "train_loss": 0.41664891704624774,
+    "train_runtime": 99.7082,
+    "train_samples_per_second": 6.108,
+    "train_steps_per_second": 2.046
 }

trainer_state.json

@@ -10,152 +10,152 @@
   "log_history": [
     {
       "epoch": 0.14705882352941177,
-      "grad_norm": 1.1240489482879639,
+      "grad_norm": 1.1769816875457764,
       "learning_rate": 9.509803921568627e-05,
-      "loss": 3.006,
+      "loss": 3.0083,
       "step": 10
     },
     {
       "epoch": 0.29411764705882354,
-      "grad_norm": 0.9972021579742432,
+      "grad_norm": 0.9931313395500183,
       "learning_rate": 9.019607843137255e-05,
-      "loss": 0.3771,
+      "loss": 0.3765,
       "step": 20
     },
     {
       "epoch": 0.4411764705882353,
-      "grad_norm": 1.274720549583435,
+      "grad_norm": 1.2492034435272217,
       "learning_rate": 8.529411764705883e-05,
-      "loss": 0.4497,
+      "loss": 0.45,
       "step": 30
     },
     {
       "epoch": 0.5882352941176471,
-      "grad_norm": 0.8991940021514893,
+      "grad_norm": 0.9271335005760193,
       "learning_rate": 8.039215686274511e-05,
-      "loss": 0.3536,
+      "loss": 0.354,
       "step": 40
     },
     {
       "epoch": 0.7352941176470589,
-      "grad_norm": 0.9119272828102112,
+      "grad_norm": 0.9514768719673157,
       "learning_rate": 7.549019607843137e-05,
-      "loss": 0.3353,
+      "loss": 0.3358,
       "step": 50
     },
     {
       "epoch": 0.8823529411764706,
-      "grad_norm": 0.6662648320198059,
+      "grad_norm": 0.6885753870010376,
       "learning_rate": 7.058823529411765e-05,
-      "loss": 0.4034,
+      "loss": 0.403,
       "step": 60
     },
     {
       "epoch": 1.0294117647058822,
-      "grad_norm": 0.7458919882774353,
+      "grad_norm": 0.7592061161994934,
       "learning_rate": 6.568627450980392e-05,
-      "loss": 0.4318,
+      "loss": 0.431,
       "step": 70
     },
     {
       "epoch": 1.1764705882352942,
-      "grad_norm": 0.937239944934845,
+      "grad_norm": 0.9498554468154907,
       "learning_rate": 6.078431372549019e-05,
-      "loss": 0.2594,
+      "loss": 0.2607,
       "step": 80
     },
     {
       "epoch": 1.3235294117647058,
-      "grad_norm": 0.8029321432113647,
+      "grad_norm": 0.8069510459899902,
       "learning_rate": 5.588235294117647e-05,
-      "loss": 0.2861,
+      "loss": 0.2868,
       "step": 90
     },
     {
       "epoch": 1.4705882352941178,
-      "grad_norm": 0.7472988367080688,
+      "grad_norm": 0.7954164147377014,
       "learning_rate": 5.0980392156862745e-05,
-      "loss": 0.2578,
+      "loss": 0.2608,
       "step": 100
     },
     {
       "epoch": 1.6176470588235294,
-      "grad_norm": 0.6276091933250427,
+      "grad_norm": 0.6208708882331848,
       "learning_rate": 4.607843137254902e-05,
-      "loss": 0.2554,
+      "loss": 0.2575,
       "step": 110
     },
     {
       "epoch": 1.7647058823529411,
-      "grad_norm": 0.551144540309906,
+      "grad_norm": 0.5807853937149048,
       "learning_rate": 4.11764705882353e-05,
-      "loss": 0.3111,
+      "loss": 0.3115,
       "step": 120
     },
     {
       "epoch": 1.9117647058823528,
-      "grad_norm": 1.0049179792404175,
+      "grad_norm": 0.9814097881317139,
       "learning_rate": 3.627450980392157e-05,
-      "loss": 0.2822,
+      "loss": 0.2838,
       "step": 130
     },
     {
       "epoch": 2.0588235294117645,
-      "grad_norm": 0.6848428845405579,
+      "grad_norm": 0.7195965051651001,
       "learning_rate": 3.137254901960784e-05,
-      "loss": 0.2257,
+      "loss": 0.2277,
       "step": 140
     },
     {
       "epoch": 2.2058823529411766,
-      "grad_norm": 0.7736006379127502,
+      "grad_norm": 0.7863514423370361,
       "learning_rate": 2.647058823529412e-05,
-      "loss": 0.2035,
+      "loss": 0.1986,
       "step": 150
     },
     {
       "epoch": 2.3529411764705883,
-      "grad_norm": 0.640957236289978,
+      "grad_norm": 0.6758102178573608,
       "learning_rate": 2.1568627450980395e-05,
-      "loss": 0.2181,
+      "loss": 0.2185,
       "step": 160
     },
     {
       "epoch": 2.5,
-      "grad_norm": 1.5639573335647583,
+      "grad_norm": 1.3710891008377075,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.1645,
+      "loss": 0.1649,
       "step": 170
     },
     {
       "epoch": 2.6470588235294117,
-      "grad_norm": 0.870328962802887,
+      "grad_norm": 0.8887768387794495,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 0.1894,
+      "loss": 0.1913,
       "step": 180
     },
     {
       "epoch": 2.7941176470588234,
-      "grad_norm": 1.0381242036819458,
+      "grad_norm": 1.0456489324569702,
       "learning_rate": 6.862745098039216e-06,
-      "loss": 0.2089,
+      "loss": 0.2106,
       "step": 190
     },
     {
       "epoch": 2.9411764705882355,
-      "grad_norm": 1.1926450729370117,
+      "grad_norm": 1.153358817100525,
       "learning_rate": 1.96078431372549e-06,
-      "loss": 0.1855,
+      "loss": 0.188,
       "step": 200
     },
     {
       "epoch": 3.0,
       "step": 204,
       "total_flos": 0.0,
-      "train_loss": 0.4158199636959562,
-      "train_runtime": 99.4711,
-      "train_samples_per_second": 6.122,
-      "train_steps_per_second": 2.051
+      "train_loss": 0.41664891704624774,
+      "train_runtime": 99.7082,
+      "train_samples_per_second": 6.108,
+      "train_steps_per_second": 2.046
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2952133924800416bdb80493b0f16baf261b3a81b8d0a0bcd5bfedc4aed9112f
+oid sha256:1daa919d0ce200e377fefdad454b4385952ae4a5fdfc46d20940a70fdf517c8b
 size 5432