Training in progress, step 5200

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b404c5888690bba9b58b3507f7eac805a1ab29333af2757222a053dc146812a2
+oid sha256:3a536e08820601dbe832e7ccdf2b20258bad72ee53bc15086cac2fc1dbfb9c21
 size 14764744

last-checkpoint/adapter_config.json

@@ -31,10 +31,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "o_proj",
+    "q_proj",
     "v_proj",
-    "q_proj"
+    "o_proj",
+    "k_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:b404c5888690bba9b58b3507f7eac805a1ab29333af2757222a053dc146812a2
+oid sha256:52217ccaabbb2c9a520ff6c890c15d98aa47a4c28fad51432c095d328a2a29be
 size 14764744

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b70332f33a8ca8b6397c23b7a822c65a312fe7275d2f0aafc90652f066bf3c8
+oid sha256:a6af5e95ef3e872afad1bdf1fdf3ea5f5155675c6586239ad660ccd39c37347e
 size 29614667

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84fc7edbd29147d5ce877470f0df0e22cc30c1e49a8b878f910eae3920c253ad
+oid sha256:a82aeb7d859423fae1de432598feb0f08060628848416302157961cd81d904dc
 size 14917

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e4eab5120346413affb45f49bfe9a21539ddac698022feee9d9f093a97d3f2b
+oid sha256:51933555f7d7492da069489fcbf900fea3e97204e288093912746efcc0733e6e
 size 14917

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc3584409b6a491e8aa7c44256ce66d6e741edd01d505aad2290535ea1f122a5
+oid sha256:7d223d319f4b8402bb6d092355111dac11a4bb47d6e23f14a8f4c5bc3c5b9c8f
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:888d7613ab3941533567ef2056d0b1a29e7ddd4c4f700ebef4104eb2e754870d
+oid sha256:dd465d73b04f6c59912627375bc7905b039595a268513d7eb113da343a1d81e2
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.8612151746113766,
+  "epoch": 0.8267665676269216,
   "eval_steps": 100,
-  "global_step": 5000,
+  "global_step": 4800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5336,228 +5336,6 @@
       "eval_samples_per_second": 5.161,
       "eval_steps_per_second": 1.291,
       "step": 4800
-    },
-    {
-      "entropy": 0.857286661863327,
-      "epoch": 0.8284889979761444,
-      "grad_norm": 0.4545460641384125,
-      "learning_rate": 3.434378229417844e-06,
-      "loss": 0.8539424896240234,
-      "mean_token_accuracy": 0.8060515314340592,
-      "num_tokens": 136995.0,
-      "step": 4810
-    },
-    {
-      "entropy": 0.8996703952550889,
-      "epoch": 0.8302114283253671,
-      "grad_norm": 0.43648988008499146,
-      "learning_rate": 3.39993110575267e-06,
-      "loss": 0.8872912406921387,
-      "mean_token_accuracy": 0.8003117755055428,
-      "num_tokens": 269371.0,
-      "step": 4820
-    },
-    {
-      "entropy": 0.897844985127449,
-      "epoch": 0.8319338586745898,
-      "grad_norm": 0.405977338552475,
-      "learning_rate": 3.365483982087496e-06,
-      "loss": 0.8877438545227051,
-      "mean_token_accuracy": 0.8004682302474976,
-      "num_tokens": 407343.0,
-      "step": 4830
-    },
-    {
-      "entropy": 0.8919987097382546,
-      "epoch": 0.8336562890238126,
-      "grad_norm": 0.43606001138687134,
-      "learning_rate": 3.3310368584223218e-06,
-      "loss": 0.8915199279785156,
-      "mean_token_accuracy": 0.8018717259168625,
-      "num_tokens": 543192.0,
-      "step": 4840
-    },
-    {
-      "entropy": 0.8664562180638313,
-      "epoch": 0.8353787193730353,
-      "grad_norm": 0.4047588109970093,
-      "learning_rate": 3.296589734757148e-06,
-      "loss": 0.8543147087097168,
-      "mean_token_accuracy": 0.8059070542454719,
-      "num_tokens": 678711.0,
-      "step": 4850
-    },
-    {
-      "entropy": 0.9019272044301033,
-      "epoch": 0.8371011497222581,
-      "grad_norm": 0.40798190236091614,
-      "learning_rate": 3.262142611091974e-06,
-      "loss": 0.901366138458252,
-      "mean_token_accuracy": 0.7967334732413291,
-      "num_tokens": 815193.0,
-      "step": 4860
-    },
-    {
-      "entropy": 0.8930811554193496,
-      "epoch": 0.8388235800714808,
-      "grad_norm": 0.44522613286972046,
-      "learning_rate": 3.2276954874268003e-06,
-      "loss": 0.8780466079711914,
-      "mean_token_accuracy": 0.7989543616771698,
-      "num_tokens": 951512.0,
-      "step": 4870
-    },
-    {
-      "entropy": 0.897969713807106,
-      "epoch": 0.8405460104207036,
-      "grad_norm": 0.4132952094078064,
-      "learning_rate": 3.193248363761626e-06,
-      "loss": 0.89752197265625,
-      "mean_token_accuracy": 0.7986764028668404,
-      "num_tokens": 1091226.0,
-      "step": 4880
-    },
-    {
-      "entropy": 0.8763776004314423,
-      "epoch": 0.8422684407699264,
-      "grad_norm": 0.464271605014801,
-      "learning_rate": 3.1588012400964525e-06,
-      "loss": 0.866519546508789,
-      "mean_token_accuracy": 0.8051836535334587,
-      "num_tokens": 1230839.0,
-      "step": 4890
-    },
-    {
-      "entropy": 0.8817414686083793,
-      "epoch": 0.8439908711191492,
-      "grad_norm": 0.40158164501190186,
-      "learning_rate": 3.1243541164312784e-06,
-      "loss": 0.8720556259155273,
-      "mean_token_accuracy": 0.8009096115827561,
-      "num_tokens": 1371321.0,
-      "step": 4900
-    },
-    {
-      "epoch": 0.8439908711191492,
-      "eval_entropy": 0.8905009640159186,
-      "eval_loss": 0.886862576007843,
-      "eval_mean_token_accuracy": 0.8004815642430033,
-      "eval_num_tokens": 1371321.0,
-      "eval_runtime": 2214.5743,
-      "eval_samples_per_second": 4.66,
-      "eval_steps_per_second": 1.165,
-      "step": 4900
-    },
-    {
-      "entropy": 0.8792809963226318,
-      "epoch": 0.8457133014683719,
-      "grad_norm": 0.45575663447380066,
-      "learning_rate": 3.0899069927661042e-06,
-      "loss": 0.8713504791259765,
-      "mean_token_accuracy": 0.8019704148173332,
-      "num_tokens": 1508027.0,
-      "step": 4910
-    },
-    {
-      "entropy": 0.852638853341341,
-      "epoch": 0.8474357318175946,
-      "grad_norm": 0.5255561470985413,
-      "learning_rate": 3.05545986910093e-06,
-      "loss": 0.8388608932495117,
-      "mean_token_accuracy": 0.808800981938839,
-      "num_tokens": 1644684.0,
-      "step": 4920
-    },
-    {
-      "entropy": 0.852977229654789,
-      "epoch": 0.8491581621668174,
-      "grad_norm": 0.43369248509407043,
-      "learning_rate": 3.0210127454357564e-06,
-      "loss": 0.8463091850280762,
-      "mean_token_accuracy": 0.8071771740913392,
-      "num_tokens": 1782176.0,
-      "step": 4930
-    },
-    {
-      "entropy": 0.8610281966626644,
-      "epoch": 0.8508805925160401,
-      "grad_norm": 0.44929254055023193,
-      "learning_rate": 2.9865656217705823e-06,
-      "loss": 0.8570366859436035,
-      "mean_token_accuracy": 0.8059208214282989,
-      "num_tokens": 1920144.0,
-      "step": 4940
-    },
-    {
-      "entropy": 0.8712201908230781,
-      "epoch": 0.8526030228652629,
-      "grad_norm": 0.4670601487159729,
-      "learning_rate": 2.9521184981054086e-06,
-      "loss": 0.8666061401367188,
-      "mean_token_accuracy": 0.8054374232888222,
-      "num_tokens": 2059336.0,
-      "step": 4950
-    },
-    {
-      "entropy": 0.8756440788507461,
-      "epoch": 0.8543254532144856,
-      "grad_norm": 0.41233476996421814,
-      "learning_rate": 2.9176713744402345e-06,
-      "loss": 0.862638282775879,
-      "mean_token_accuracy": 0.8034965068101882,
-      "num_tokens": 2201332.0,
-      "step": 4960
-    },
-    {
-      "entropy": 0.9280423223972321,
-      "epoch": 0.8560478835637084,
-      "grad_norm": 0.37263938784599304,
-      "learning_rate": 2.883224250775061e-06,
-      "loss": 0.923769760131836,
-      "mean_token_accuracy": 0.7912818253040313,
-      "num_tokens": 2339368.0,
-      "step": 4970
-    },
-    {
-      "entropy": 0.8748306304216384,
-      "epoch": 0.8577703139129311,
-      "grad_norm": 0.47100237011909485,
-      "learning_rate": 2.8487771271098863e-06,
-      "loss": 0.8737285614013672,
-      "mean_token_accuracy": 0.8047029227018356,
-      "num_tokens": 2476367.0,
-      "step": 4980
-    },
-    {
-      "entropy": 0.9129098683595658,
-      "epoch": 0.8594927442621539,
-      "grad_norm": 0.528985321521759,
-      "learning_rate": 2.8143300034447126e-06,
-      "loss": 0.9065395355224609,
-      "mean_token_accuracy": 0.7945742771029473,
-      "num_tokens": 2613684.0,
-      "step": 4990
-    },
-    {
-      "entropy": 0.8374424375593662,
-      "epoch": 0.8612151746113766,
-      "grad_norm": 0.5343524217605591,
-      "learning_rate": 2.7798828797795385e-06,
-      "loss": 0.8323655128479004,
-      "mean_token_accuracy": 0.8117164567112922,
-      "num_tokens": 2748511.0,
-      "step": 5000
-    },
-    {
-      "epoch": 0.8612151746113766,
-      "eval_entropy": 0.8905347423907262,
-      "eval_loss": 0.886751651763916,
-      "eval_mean_token_accuracy": 0.8005248703313678,
-      "eval_num_tokens": 2748511.0,
-      "eval_runtime": 2213.3653,
-      "eval_samples_per_second": 4.663,
-      "eval_steps_per_second": 1.166,
-      "step": 5000
     }
   ],
   "logging_steps": 10,
@@ -5577,7 +5355,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8.214454786955674e+17,
+  "total_flos": 7.887012992552796e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null