Upload DeepCoder14B_DSPy LoRA adapter checkpoint

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:229486ce1d4253a54d0cd47c5005c0eacd850b87a95cd3216a73e019d10945c6
+oid sha256:5dc489c7ab4ab92650304b6f75c652bde544d2c53f17b9cc633f1cf12789c71d
 size 6799256256

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5062a362186c064ec8701c9f44214104908f82eb3da4e29d6a4dcd1e277acdcb
+oid sha256:18dfdd7067e9a4837aa29ad3b34d650c432e5374748c5580051b51610e4e9175
 size 6519149878

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7642d4d8677c028eb706576fdd39c921ffb1007ec52044603d519b96b28ba821
+oid sha256:3c59efed336e3af92ec9b9410099a59c2927ff1388b85211c0161b5fff7836a2
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49f276f737c8546acf52b47aa0a82341e11a9aa193441ce56e73d93c1135b634
+oid sha256:54b8ad803d85e44a0c1ade1fcda24008ebe595e19cb0f6fef0c012f10d4f5f35
 size 1064

trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 275,
   "best_metric": 0.9421520233154297,
   "best_model_checkpoint": "./dspy-8bit-lora-deepcoder-safe/checkpoint-275",
-  "epoch": 1.9415929203539823,
+  "epoch": 2.9982300884955753,
   "eval_steps": 25,
-  "global_step": 275,
+  "global_step": 425,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -217,6 +217,120 @@
       "eval_samples_per_second": 2.059,
       "eval_steps_per_second": 2.059,
       "step": 275
+    },
+    {
+      "epoch": 2.113274336283186,
+      "grad_norm": 1.1352859735488892,
+      "learning_rate": 7.791411042944787e-05,
+      "loss": 0.5105,
+      "mean_token_accuracy": 0.8593325974400511,
+      "num_tokens": 680687.0,
+      "step": 300
+    },
+    {
+      "epoch": 2.113274336283186,
+      "eval_loss": 0.9843020439147949,
+      "eval_mean_token_accuracy": 0.7742766647040844,
+      "eval_num_tokens": 680687.0,
+      "eval_runtime": 98.2979,
+      "eval_samples_per_second": 2.035,
+      "eval_steps_per_second": 2.035,
+      "step": 300
+    },
+    {
+      "epoch": 2.2902654867256635,
+      "grad_norm": 0.8646391034126282,
+      "learning_rate": 6.257668711656443e-05,
+      "loss": 0.4312,
+      "mean_token_accuracy": 0.878243299126625,
+      "num_tokens": 733979.0,
+      "step": 325
+    },
+    {
+      "epoch": 2.2902654867256635,
+      "eval_loss": 0.993569552898407,
+      "eval_mean_token_accuracy": 0.775199833214283,
+      "eval_num_tokens": 733979.0,
+      "eval_runtime": 96.9581,
+      "eval_samples_per_second": 2.063,
+      "eval_steps_per_second": 2.063,
+      "step": 325
+    },
+    {
+      "epoch": 2.4672566371681417,
+      "grad_norm": 1.0304445028305054,
+      "learning_rate": 4.723926380368098e-05,
+      "loss": 0.3952,
+      "mean_token_accuracy": 0.8835302656888961,
+      "num_tokens": 789649.0,
+      "step": 350
+    },
+    {
+      "epoch": 2.4672566371681417,
+      "eval_loss": 0.9702684283256531,
+      "eval_mean_token_accuracy": 0.77936582878232,
+      "eval_num_tokens": 789649.0,
+      "eval_runtime": 98.4219,
+      "eval_samples_per_second": 2.032,
+      "eval_steps_per_second": 2.032,
+      "step": 350
+    },
+    {
+      "epoch": 2.6442477876106194,
+      "grad_norm": 0.8441785573959351,
+      "learning_rate": 3.1901840490797544e-05,
+      "loss": 0.3966,
+      "mean_token_accuracy": 0.8844721549749375,
+      "num_tokens": 845295.0,
+      "step": 375
+    },
+    {
+      "epoch": 2.6442477876106194,
+      "eval_loss": 0.9620086550712585,
+      "eval_mean_token_accuracy": 0.7812778241932392,
+      "eval_num_tokens": 845295.0,
+      "eval_runtime": 97.2727,
+      "eval_samples_per_second": 2.056,
+      "eval_steps_per_second": 2.056,
+      "step": 375
+    },
+    {
+      "epoch": 2.8212389380530976,
+      "grad_norm": 0.8250017762184143,
+      "learning_rate": 1.656441717791411e-05,
+      "loss": 0.3743,
+      "mean_token_accuracy": 0.8831119546294213,
+      "num_tokens": 908151.0,
+      "step": 400
+    },
+    {
+      "epoch": 2.8212389380530976,
+      "eval_loss": 0.9527075290679932,
+      "eval_mean_token_accuracy": 0.7827157293260097,
+      "eval_num_tokens": 908151.0,
+      "eval_runtime": 98.1465,
+      "eval_samples_per_second": 2.038,
+      "eval_steps_per_second": 2.038,
+      "step": 400
+    },
+    {
+      "epoch": 2.9982300884955753,
+      "grad_norm": 0.8464552164077759,
+      "learning_rate": 1.226993865030675e-06,
+      "loss": 0.4002,
+      "mean_token_accuracy": 0.883520859181881,
+      "num_tokens": 967441.0,
+      "step": 425
+    },
+    {
+      "epoch": 2.9982300884955753,
+      "eval_loss": 0.9509771466255188,
+      "eval_mean_token_accuracy": 0.7838174617290496,
+      "eval_num_tokens": 967441.0,
+      "eval_runtime": 98.6966,
+      "eval_samples_per_second": 2.026,
+      "eval_steps_per_second": 2.026,
+      "step": 425
     }
   ],
   "logging_steps": 25,
@@ -236,7 +350,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5.329544702974157e+16,
+  "total_flos": 8.204357166595891e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null