Training in progress, step 500

adapter_config.json

@@ -24,21 +24,21 @@
   "revision": null,
   "target_modules": [
     "lm_heads.3",
-    "out_proj",
-    "embed_tokens.1",
-    "lm_heads.0",
-    "v_proj",
     "embed_tokens.0",
-    "embed_tokens.3",
-    "q_proj",
-    "fc1",
     "embed_tokens.2",
-    "enc_to_dec_proj",
-    "lm_heads.2",
-    "lm_heads.1",
+    "v_proj",
+    "q_proj",
     "fc2",
+    "lm_heads.1",
+    "lm_heads.2",
+    "fc1",
+    "embed_tokens.1",
+    "lm_heads.0",
+    "embed_tokens.3",
     "k_proj",
-    "audio_enc_to_dec_proj"
+    "enc_to_dec_proj",
+    "audio_enc_to_dec_proj",
+    "out_proj"
   ],
   "task_type": null,
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19d442511f113fbf128dd4b4125ab1561bcd367e0abece36f861aa3a753a6f90
+oid sha256:4e9aee3c102dd92705019df0ee99d4604e7b6f060a6d8411db9f457bd070868c
 size 87103456

trainer_state.json

@@ -1,39 +1,123 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 4.0,
+  "epoch": 3.6721311475409837,
   "eval_steps": 500,
-  "global_step": 4,
+  "global_step": 28,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 2.0,
-      "grad_norm": 0.28847259283065796,
-      "learning_rate": 0.0001,
-      "loss": 2.4057,
+      "epoch": 0.26229508196721313,
+      "grad_norm": 1.36014723777771,
+      "learning_rate": 0.00018571428571428572,
+      "loss": 9.4034,
       "step": 2
     },
     {
-      "epoch": 4.0,
-      "grad_norm": 0.39801156520843506,
-      "learning_rate": 0.0,
-      "loss": 2.3519,
+      "epoch": 0.5245901639344263,
+      "grad_norm": 2.117264986038208,
+      "learning_rate": 0.00017142857142857143,
+      "loss": 8.9051,
       "step": 4
     },
     {
-      "epoch": 4.0,
-      "step": 4,
-      "total_flos": 4471506350256.0,
-      "train_loss": 2.3788044452667236,
-      "train_runtime": 6.9877,
-      "train_samples_per_second": 2.29,
-      "train_steps_per_second": 0.572
+      "epoch": 0.7868852459016393,
+      "grad_norm": 2.277207851409912,
+      "learning_rate": 0.00015714285714285716,
+      "loss": 8.3058,
+      "step": 6
+    },
+    {
+      "epoch": 1.0491803278688525,
+      "grad_norm": 1.9988380670547485,
+      "learning_rate": 0.00014285714285714287,
+      "loss": 7.8175,
+      "step": 8
+    },
+    {
+      "epoch": 1.3114754098360657,
+      "grad_norm": 1.7698421478271484,
+      "learning_rate": 0.00012857142857142858,
+      "loss": 7.2821,
+      "step": 10
+    },
+    {
+      "epoch": 1.5737704918032787,
+      "grad_norm": 2.32663893699646,
+      "learning_rate": 0.00011428571428571428,
+      "loss": 7.4096,
+      "step": 12
+    },
+    {
+      "epoch": 1.8360655737704918,
+      "grad_norm": 1.4435824155807495,
+      "learning_rate": 0.0001,
+      "loss": 7.1488,
+      "step": 14
+    },
+    {
+      "epoch": 2.098360655737705,
+      "grad_norm": 1.5567710399627686,
+      "learning_rate": 8.571428571428571e-05,
+      "loss": 6.7253,
+      "step": 16
+    },
+    {
+      "epoch": 2.360655737704918,
+      "grad_norm": 1.3461309671401978,
+      "learning_rate": 7.142857142857143e-05,
+      "loss": 6.9232,
+      "step": 18
+    },
+    {
+      "epoch": 2.6229508196721314,
+      "grad_norm": 1.2633932828903198,
+      "learning_rate": 5.714285714285714e-05,
+      "loss": 6.7861,
+      "step": 20
+    },
+    {
+      "epoch": 2.8852459016393444,
+      "grad_norm": 0.7925029993057251,
+      "learning_rate": 4.2857142857142856e-05,
+      "loss": 6.8389,
+      "step": 22
+    },
+    {
+      "epoch": 3.1475409836065573,
+      "grad_norm": 1.3408855199813843,
+      "learning_rate": 2.857142857142857e-05,
+      "loss": 6.5778,
+      "step": 24
+    },
+    {
+      "epoch": 3.4098360655737707,
+      "grad_norm": 1.250908613204956,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 6.8636,
+      "step": 26
+    },
+    {
+      "epoch": 3.6721311475409837,
+      "grad_norm": 1.3394016027450562,
+      "learning_rate": 0.0,
+      "loss": 6.9516,
+      "step": 28
+    },
+    {
+      "epoch": 3.6721311475409837,
+      "step": 28,
+      "total_flos": 134754941373624.0,
+      "train_loss": 7.424190691539219,
+      "train_runtime": 121.6682,
+      "train_samples_per_second": 4.011,
+      "train_steps_per_second": 0.23
     }
   ],
   "logging_steps": 2,
-  "max_steps": 4,
+  "max_steps": 28,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
   "save_steps": 500,
@@ -49,7 +133,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4471506350256.0,
+  "total_flos": 134754941373624.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f143200a1c3e49ed254281489474e15594df2e4750b346b8897649dc3e2ae490
+oid sha256:bf98e72b2ffaac26af2a5f6b503a28a6a832efb516b8f0746c342f7fb6e02eda
 size 5304