Upload folder using huggingface_hub

checkpoint-step-400/adapter_config.json

@@ -34,13 +34,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "up_proj",
-    "down_proj",
     "gate_proj",
-    "q_proj",
     "k_proj",
+    "v_proj",
+    "down_proj",
     "o_proj",
-    "v_proj"
+    "up_proj",
+    "q_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

checkpoint-step-400/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:922b6d5903270b4140fe04520bf87d296e0f830c4926c135ea7c408bea27a85d
+oid sha256:3cc9b2482c9b9d897229fa4d79799dd0c56f5758f5dc3b876924b902d28a4155
 size 239536272

checkpoint-step-400/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c68746d085882a73d44f039e63800dc17292b2318427c454274fc423c6ab75f2
+oid sha256:d986517448d7aecb6fa58d482e334cbe73e194802d364ff25fffe361592ac4ed
 size 122203333

checkpoint-step-400/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:232aaf53450a9a142256e42b5eeb426498e782a3d459e18ba17af0155f722be4
+oid sha256:bc7d2691dc1aea11c26d9c01835daff5aa07675dcf61d839b13fd431caad08da
 size 1465

checkpoint-step-400/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 400,
-  "best_metric": 0.9693769812583923,
+  "best_metric": 0.9676186442375183,
   "best_model_checkpoint": "./outputs/checkpoint-400",
   "epoch": 0.7518796992481203,
   "eval_steps": 100,
@@ -18,90 +18,90 @@
     },
     {
       "epoch": 0.18796992481203006,
-      "grad_norm": 0.8309985995292664,
-      "learning_rate": 9.907020872865275e-05,
-      "loss": 0.775905990600586,
+      "grad_norm": 0.8308438062667847,
+      "learning_rate": 9.683053040103493e-05,
+      "loss": 0.7757094573974609,
       "step": 100
     },
     {
       "epoch": 0.18796992481203006,
-      "eval_loss": 0.9841089248657227,
-      "eval_runtime": 71.1499,
-      "eval_samples_per_second": 6.648,
-      "eval_steps_per_second": 1.673,
+      "eval_loss": 0.9840108752250671,
+      "eval_runtime": 81.3507,
+      "eval_samples_per_second": 5.814,
+      "eval_steps_per_second": 1.463,
       "step": 100
     },
     {
       "epoch": 0.2819548872180451,
-      "grad_norm": 1.0482571125030518,
-      "learning_rate": 9.81214421252372e-05,
-      "loss": 0.7683392333984375,
+      "grad_norm": 1.0474584102630615,
+      "learning_rate": 9.359637774902976e-05,
+      "loss": 0.7674188995361328,
       "step": 150
     },
     {
       "epoch": 0.37593984962406013,
-      "grad_norm": 0.8791717290878296,
-      "learning_rate": 9.717267552182164e-05,
-      "loss": 0.7416677856445313,
+      "grad_norm": 0.8780233860015869,
+      "learning_rate": 9.036222509702458e-05,
+      "loss": 0.7395731353759766,
       "step": 200
     },
     {
       "epoch": 0.37593984962406013,
-      "eval_loss": 0.9950897097587585,
-      "eval_runtime": 70.4425,
-      "eval_samples_per_second": 6.715,
-      "eval_steps_per_second": 1.689,
+      "eval_loss": 0.9940316081047058,
+      "eval_runtime": 80.2928,
+      "eval_samples_per_second": 5.891,
+      "eval_steps_per_second": 1.482,
       "step": 200
     },
     {
       "epoch": 0.4699248120300752,
-      "grad_norm": 0.912263035774231,
-      "learning_rate": 9.622390891840608e-05,
-      "loss": 0.7662993621826172,
+      "grad_norm": 0.9176077246665955,
+      "learning_rate": 8.712807244501941e-05,
+      "loss": 0.7627955627441406,
       "step": 250
     },
     {
       "epoch": 0.5639097744360902,
-      "grad_norm": 0.8556099534034729,
-      "learning_rate": 9.527514231499052e-05,
-      "loss": 0.720741195678711,
+      "grad_norm": 0.8625491857528687,
+      "learning_rate": 8.389391979301423e-05,
+      "loss": 0.7164928436279296,
       "step": 300
     },
     {
       "epoch": 0.5639097744360902,
-      "eval_loss": 0.9841504693031311,
-      "eval_runtime": 70.3252,
-      "eval_samples_per_second": 6.726,
-      "eval_steps_per_second": 1.692,
+      "eval_loss": 0.9823606014251709,
+      "eval_runtime": 80.8571,
+      "eval_samples_per_second": 5.85,
+      "eval_steps_per_second": 1.472,
       "step": 300
     },
     {
       "epoch": 0.6578947368421053,
-      "grad_norm": 0.8402419686317444,
-      "learning_rate": 9.432637571157496e-05,
-      "loss": 0.7455893707275391,
+      "grad_norm": 0.8384292125701904,
+      "learning_rate": 8.065976714100906e-05,
+      "loss": 0.73987548828125,
       "step": 350
     },
     {
       "epoch": 0.7518796992481203,
-      "grad_norm": 0.8201877474784851,
-      "learning_rate": 9.337760910815939e-05,
-      "loss": 0.7341905975341797,
+      "grad_norm": 0.8384448885917664,
+      "learning_rate": 7.742561448900388e-05,
+      "loss": 0.7277918243408203,
       "step": 400
     },
     {
       "epoch": 0.7518796992481203,
-      "eval_loss": 0.9693769812583923,
-      "eval_runtime": 70.2206,
-      "eval_samples_per_second": 6.736,
-      "eval_steps_per_second": 1.695,
+      "eval_loss": 0.9676186442375183,
+      "eval_runtime": 80.4443,
+      "eval_samples_per_second": 5.88,
+      "eval_steps_per_second": 1.479,
       "step": 400
     }
   ],
   "logging_steps": 50,
-  "max_steps": 5320,
+  "max_steps": 1596,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 3,
   "save_steps": 100,
   "stateful_callbacks": {
     "TrainerControl": {

checkpoint-step-400/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bc05806d2c86681ef6c541c75942126f28b3a31aec0d903899ed9bf1ea303bac
+oid sha256:3c88f5ec73ab38f2f0a5bb321f3c58aaa6f716ae0e803b69ff493f9e0c43f221
 size 5713