Training in progress, step 2850, checkpoint

last-checkpoint/adapter_config.json

@@ -29,13 +29,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "up_proj",
+    "o_proj",
+    "k_proj",
     "down_proj",
-    "q_proj",
-    "gate_proj",
     "v_proj",
-    "k_proj",
-    "o_proj",
-    "up_proj"
+    "q_proj",
+    "gate_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f729200728ba44a9ee48ef9f7a746ee4013df5b4441295deb59e7e533620a072
+oid sha256:bd9069c95306b127720930a77fad3324cb0d7eda4a839db90d5605d41852db31
 size 108113968

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:971c9d1c5b74e903eb9d3f94fe8af4db0d58ee1f9669348434fb9761d8371e7b
+oid sha256:22c04e45f62499567b0d0bf6dc6fe8b59e32b7921c27ac5943ee035968aa590d
 size 57081771

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b81ad22ca863c3bf9e4ae89b44de94db92351a68d9ae0bd607ab062743a9371
+oid sha256:5ed08c1881bd90abbd6a4f09b7d8dcdcd76153d3d15f906bdc29b7fedf5425cb
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.8096875895672112,
+  "epoch": 1.6334193178561192,
   "eval_steps": 300,
-  "global_step": 2825,
+  "global_step": 2850,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1138,12 +1138,22 @@
       "mean_token_accuracy": 0.9419060254096985,
       "num_tokens": 250323.0,
       "step": 2825
+    },
+    {
+      "entropy": 0.15991646617650987,
+      "epoch": 1.6334193178561192,
+      "grad_norm": 0.19363631308078766,
+      "learning_rate": 0.0002,
+      "loss": 0.15416499137878417,
+      "mean_token_accuracy": 0.9481097102165222,
+      "num_tokens": 82145.0,
+      "step": 2850
     }
   ],
   "logging_steps": 25,
   "max_steps": 3489,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 1,
+  "num_train_epochs": 2,
   "save_steps": 25,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -1157,7 +1167,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.7417012377326387e+17,
+  "total_flos": 2.7903573973558886e+17,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b6b8cd0e3f39da2d16e5b5e0bab780af893d21cb8f98534cd79dde7079e4d98
+oid sha256:96134eafdc3eaeac8127a27f3b984cecea62d6c6dc170b55ad5610e26cab65b1
 size 5649