Training in progress, step 10

adapter_config.json

@@ -23,10 +23,10 @@
     "q_proj",
     "o_proj",
     "v_proj",
-    "gate_proj",
     "k_proj",
-    "down_proj",
-    "up_proj"
+    "gate_proj",
+    "up_proj",
+    "down_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:124ef4fdc953020354a6a08a184b25002db40ec1092366cacb4e27ba354d142f
+oid sha256:e0b99f4fc183c1047f46deeee9c5f3b85db51cf743b0df0a7a30b75120ffbae7
 size 35668592

state.json

@@ -1,46 +1,15 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.48,
+  "epoch": 0.016,
   "eval_steps": 20,
-  "global_step": 60,
+  "global_step": 10,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.16,
-      "grad_norm": 0.04197464883327484,
-      "learning_rate": 0.0002631578947368421,
-      "loss": 1.9166,
-      "step": 20
-    },
-    {
-      "epoch": 0.16,
-      "eval_loss": 1.8241591453552246,
-      "eval_runtime": 297.4682,
-      "eval_samples_per_second": 3.362,
-      "eval_steps_per_second": 0.42,
-      "step": 20
-    },
-    {
-      "epoch": 0.32,
-      "grad_norm": 0.10275597870349884,
-      "learning_rate": 0.0004970326409495549,
-      "loss": 1.6713,
-      "step": 40
-    },
-    {
-      "epoch": 0.32,
-      "eval_loss": 1.538568139076233,
-      "eval_runtime": 304.6057,
-      "eval_samples_per_second": 3.283,
-      "eval_steps_per_second": 0.41,
-      "step": 40
-    }
-  ],
+  "log_history": [],
   "logging_steps": 20,
-  "max_steps": 375,
+  "max_steps": 1875,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 10,
@@ -50,13 +19,13 @@
         "should_epoch_stop": false,
         "should_evaluate": false,
         "should_log": false,
-        "should_save": true,
+        "should_save": false,
         "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 3880094771036160.0,
+  "total_flos": 0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d5d622493f1ded4a4395a8a495be5ddb356e13976d1325eca2493ea80f1c746a
+oid sha256:4898537838146fbf12c74d87a475327b6a5606d9348748214e9ab9ceaa968a02
 size 5432