starlineventures/chatgpt

adapter_config.json

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

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:81570fbc870f2786147ab5e78079b66b9a69b8958fea72686db4b7e8a477d14d
+oid sha256:537817d8a957663d34acfd9678d3fcdb85a1092df7428631b9b5efebce235b80
 size 94422368

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "total_flos": 0.0,
-    "train_loss": 0.45867320310835746,
-    "train_runtime": 113.026,
-    "train_samples_per_second": 5.388,
-    "train_steps_per_second": 1.805
+    "train_loss": 0.4417308241713281,
+    "train_runtime": 112.6516,
+    "train_samples_per_second": 5.406,
+    "train_steps_per_second": 1.811
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "total_flos": 0.0,
-    "train_loss": 0.45867320310835746,
-    "train_runtime": 113.026,
-    "train_samples_per_second": 5.388,
-    "train_steps_per_second": 1.805
+    "train_loss": 0.4417308241713281,
+    "train_runtime": 112.6516,
+    "train_samples_per_second": 5.406,
+    "train_steps_per_second": 1.811
 }

trainer_state.json

@@ -10,152 +10,152 @@
   "log_history": [
     {
       "epoch": 0.14705882352941177,
-      "grad_norm": 4.9361724853515625,
+      "grad_norm": 5.014659881591797,
       "learning_rate": 9.509803921568627e-05,
-      "loss": 0.7989,
+      "loss": 0.8047,
       "step": 10
     },
     {
       "epoch": 0.29411764705882354,
-      "grad_norm": 5.818482875823975,
+      "grad_norm": 3.9647512435913086,
       "learning_rate": 9.019607843137255e-05,
-      "loss": 0.5605,
+      "loss": 0.572,
       "step": 20
     },
     {
       "epoch": 0.4411764705882353,
-      "grad_norm": 5.857405185699463,
+      "grad_norm": 3.050414800643921,
       "learning_rate": 8.529411764705883e-05,
-      "loss": 0.6352,
+      "loss": 0.5933,
       "step": 30
     },
     {
       "epoch": 0.5882352941176471,
-      "grad_norm": 1.7787846326828003,
+      "grad_norm": 1.585654377937317,
       "learning_rate": 8.039215686274511e-05,
-      "loss": 0.4679,
+      "loss": 0.4563,
       "step": 40
     },
     {
       "epoch": 0.7352941176470589,
-      "grad_norm": 1.7786710262298584,
+      "grad_norm": 1.262968897819519,
       "learning_rate": 7.549019607843137e-05,
-      "loss": 0.4386,
+      "loss": 0.4342,
       "step": 50
     },
     {
       "epoch": 0.8823529411764706,
-      "grad_norm": 2.569049119949341,
+      "grad_norm": 0.7713326215744019,
       "learning_rate": 7.058823529411765e-05,
-      "loss": 0.5042,
+      "loss": 0.4967,
       "step": 60
     },
     {
       "epoch": 1.0294117647058822,
-      "grad_norm": 1.8246957063674927,
+      "grad_norm": 1.4584881067276,
       "learning_rate": 6.568627450980392e-05,
-      "loss": 0.5466,
+      "loss": 0.5211,
       "step": 70
     },
     {
       "epoch": 1.1764705882352942,
-      "grad_norm": 2.425893783569336,
+      "grad_norm": 2.0033822059631348,
       "learning_rate": 6.078431372549019e-05,
-      "loss": 0.413,
+      "loss": 0.3788,
       "step": 80
     },
     {
       "epoch": 1.3235294117647058,
-      "grad_norm": 1.5483649969100952,
+      "grad_norm": 2.14152455329895,
       "learning_rate": 5.588235294117647e-05,
-      "loss": 0.4693,
+      "loss": 0.4405,
       "step": 90
     },
     {
       "epoch": 1.4705882352941178,
-      "grad_norm": 1.3050321340560913,
+      "grad_norm": 2.8985249996185303,
       "learning_rate": 5.0980392156862745e-05,
-      "loss": 0.4126,
+      "loss": 0.3855,
       "step": 100
     },
     {
       "epoch": 1.6176470588235294,
-      "grad_norm": 1.3402087688446045,
+      "grad_norm": 2.2722222805023193,
       "learning_rate": 4.607843137254902e-05,
-      "loss": 0.3931,
+      "loss": 0.3668,
       "step": 110
     },
     {
       "epoch": 1.7647058823529411,
-      "grad_norm": 1.9078748226165771,
+      "grad_norm": 1.7401117086410522,
       "learning_rate": 4.11764705882353e-05,
-      "loss": 0.4831,
+      "loss": 0.4555,
       "step": 120
     },
     {
       "epoch": 1.9117647058823528,
-      "grad_norm": 2.0981125831604004,
+      "grad_norm": 2.3890228271484375,
       "learning_rate": 3.627450980392157e-05,
-      "loss": 0.4249,
+      "loss": 0.4121,
       "step": 130
     },
     {
       "epoch": 2.0588235294117645,
-      "grad_norm": 2.552691698074341,
+      "grad_norm": 2.051992654800415,
       "learning_rate": 3.137254901960784e-05,
-      "loss": 0.3972,
+      "loss": 0.3758,
       "step": 140
     },
     {
       "epoch": 2.2058823529411766,
-      "grad_norm": 1.179951548576355,
+      "grad_norm": 2.088096857070923,
       "learning_rate": 2.647058823529412e-05,
-      "loss": 0.3928,
+      "loss": 0.3738,
       "step": 150
     },
     {
       "epoch": 2.3529411764705883,
-      "grad_norm": 1.4807623624801636,
+      "grad_norm": 1.384030818939209,
       "learning_rate": 2.1568627450980395e-05,
-      "loss": 0.4141,
+      "loss": 0.393,
       "step": 160
     },
     {
       "epoch": 2.5,
-      "grad_norm": 2.2445242404937744,
+      "grad_norm": 1.6448745727539062,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.3197,
+      "loss": 0.2952,
       "step": 170
     },
     {
       "epoch": 2.6470588235294117,
-      "grad_norm": 1.3446571826934814,
+      "grad_norm": 1.227141261100769,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 0.3921,
+      "loss": 0.3821,
       "step": 180
     },
     {
       "epoch": 2.7941176470588234,
-      "grad_norm": 2.043297290802002,
+      "grad_norm": 2.20298433303833,
       "learning_rate": 6.862745098039216e-06,
-      "loss": 0.3877,
+      "loss": 0.3722,
       "step": 190
     },
     {
       "epoch": 2.9411764705882355,
-      "grad_norm": 0.9528805613517761,
+      "grad_norm": 1.9124171733856201,
       "learning_rate": 1.96078431372549e-06,
-      "loss": 0.3592,
+      "loss": 0.36,
       "step": 200
     },
     {
       "epoch": 3.0,
       "step": 204,
       "total_flos": 0.0,
-      "train_loss": 0.45867320310835746,
-      "train_runtime": 113.026,
-      "train_samples_per_second": 5.388,
-      "train_steps_per_second": 1.805
+      "train_loss": 0.4417308241713281,
+      "train_runtime": 112.6516,
+      "train_samples_per_second": 5.406,
+      "train_steps_per_second": 1.811
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a342ca2204ee91e163d377f9114d821ab032d9f89163f6ca3e4864769ccd118
+oid sha256:7a4e8f2d3250958257fd416d205e9505bf49a86acfd66fa7e8bdd8b8358c882d
 size 5432