Training in progress, step 5, checkpoint

last-checkpoint/README.md

@@ -1,5 +1,5 @@
 ---
-base_model: unsloth/Llama-3.2-1B-Instruct
+base_model: unsloth/Llama-3.2-3B-Instruct
 library_name: peft
 ---
 

last-checkpoint/adapter_config.json

@@ -1,7 +1,7 @@
 {
   "alpha_pattern": {},
   "auto_mapping": null,
-  "base_model_name_or_path": "unsloth/Llama-3.2-1B-Instruct",
+  "base_model_name_or_path": "unsloth/Llama-3.2-3B-Instruct",
   "bias": "none",
   "fan_in_fan_out": null,
   "inference_mode": true,
@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "gate_proj",
-    "k_proj",
     "up_proj",
-    "o_proj",
-    "q_proj",
     "v_proj",
-    "down_proj"
+    "k_proj",
+    "q_proj",
+    "gate_proj",
+    "down_proj",
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55355fd5952cf80856902cdedbaa5578c5033d8b5a54622199ec88d09ed34369
-size 45118424
+oid sha256:5cdcca758e4f51b86ca61d6dd0e1f205a4136be14bada11320e201462d69597d
+size 97307544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d59b40a37084391786c286fe6df292100bc7dbdccf9eb87e041e091ee2290fda
-size 23159290
+oid sha256:d3ce688998528313f8a65f43827d8582ea44a8e07d8a071787455a2ff2b245bd
+size 49846260

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32fae2f082bc2b8a4f93f074fb88d1c6113a4081d21b0babdad2ac6534f83080
+oid sha256:61f388398ea994e40704f6497243bd60adc9aa3f265cb5542e356410fda11e77
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb578e75c11a81e85dda67a691f96ba4793a02960f1409fd3e1511aac873491a
+oid sha256:cda2d7da3ce8a95be6df505b84eace6ccd5aa18ffc6d1bcc9a79572045c8d78f
 size 1064

last-checkpoint/trainer_state.json

@@ -1,114 +1,63 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.042105263157894736,
+  "epoch": 0.021052631578947368,
   "eval_steps": 3,
-  "global_step": 10,
+  "global_step": 5,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.004210526315789474,
-      "grad_norm": 1.1962615251541138,
+      "grad_norm": 0.42628446221351624,
       "learning_rate": 2e-05,
-      "loss": 1.3218,
+      "loss": 0.1517,
       "step": 1
     },
     {
       "epoch": 0.004210526315789474,
-      "eval_loss": 1.2625290155410767,
-      "eval_runtime": 14.4612,
-      "eval_samples_per_second": 6.915,
-      "eval_steps_per_second": 3.458,
+      "eval_loss": 0.24416916072368622,
+      "eval_runtime": 29.4222,
+      "eval_samples_per_second": 3.399,
+      "eval_steps_per_second": 1.699,
       "step": 1
     },
     {
       "epoch": 0.008421052631578947,
-      "grad_norm": 1.1140599250793457,
+      "grad_norm": 0.433437317609787,
       "learning_rate": 4e-05,
-      "loss": 1.3312,
+      "loss": 0.1501,
       "step": 2
     },
     {
       "epoch": 0.01263157894736842,
-      "grad_norm": 1.1057511568069458,
+      "grad_norm": 0.3588545322418213,
       "learning_rate": 6e-05,
-      "loss": 1.3071,
+      "loss": 0.1181,
       "step": 3
     },
     {
       "epoch": 0.01263157894736842,
-      "eval_loss": 1.2579314708709717,
-      "eval_runtime": 14.457,
-      "eval_samples_per_second": 6.917,
-      "eval_steps_per_second": 3.459,
+      "eval_loss": 0.2361646592617035,
+      "eval_runtime": 29.4356,
+      "eval_samples_per_second": 3.397,
+      "eval_steps_per_second": 1.699,
       "step": 3
     },
     {
       "epoch": 0.016842105263157894,
-      "grad_norm": 2.108193874359131,
+      "grad_norm": 1.395027756690979,
       "learning_rate": 8e-05,
-      "loss": 1.04,
+      "loss": 0.4504,
       "step": 4
     },
     {
       "epoch": 0.021052631578947368,
-      "grad_norm": 1.9758830070495605,
+      "grad_norm": 1.5977227687835693,
       "learning_rate": 0.0001,
-      "loss": 1.2265,
+      "loss": 0.4201,
       "step": 5
-    },
-    {
-      "epoch": 0.02526315789473684,
-      "grad_norm": 1.707103967666626,
-      "learning_rate": 0.00012,
-      "loss": 1.4942,
-      "step": 6
-    },
-    {
-      "epoch": 0.02526315789473684,
-      "eval_loss": 1.2140295505523682,
-      "eval_runtime": 14.4388,
-      "eval_samples_per_second": 6.926,
-      "eval_steps_per_second": 3.463,
-      "step": 6
-    },
-    {
-      "epoch": 0.029473684210526315,
-      "grad_norm": 1.1118305921554565,
-      "learning_rate": 0.00014,
-      "loss": 1.2998,
-      "step": 7
-    },
-    {
-      "epoch": 0.03368421052631579,
-      "grad_norm": 1.2043912410736084,
-      "learning_rate": 0.00016,
-      "loss": 1.6177,
-      "step": 8
-    },
-    {
-      "epoch": 0.037894736842105266,
-      "grad_norm": 1.3674756288528442,
-      "learning_rate": 0.00018,
-      "loss": 1.277,
-      "step": 9
-    },
-    {
-      "epoch": 0.037894736842105266,
-      "eval_loss": 1.216686487197876,
-      "eval_runtime": 14.4543,
-      "eval_samples_per_second": 6.918,
-      "eval_steps_per_second": 3.459,
-      "step": 9
-    },
-    {
-      "epoch": 0.042105263157894736,
-      "grad_norm": 1.0281931161880493,
-      "learning_rate": 0.0002,
-      "loss": 1.4828,
-      "step": 10
     }
   ],
   "logging_steps": 1,
@@ -123,12 +72,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 1935445139128320.0,
+  "total_flos": 2794842929233920.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f3fba08eeeb586587427cc2d5f17808d9e4f299ce6f2dc52ed4d0f7fb28b25a
+oid sha256:0796ff0b1a86f9f71fa065e52907f37a5901a78d76587d76fddb76319a6b4ce3
 size 6648