Training in progress, step 4000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62332de7696193cf95c4b58f4eb71109e913e5eb2307c4eb5d6049395383233d
+oid sha256:01efc0f310407dcb32af731fb94896e9a721a18f814a62b7ea7358ca8d8781e1
 size 363365712

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b3a7e528d79a9b026ad94e82546d4dd18bf55086ba1a8f27ea25d5ccc064e43
+oid sha256:cf61a46a074c6294f4114d14b385a465bfb53023df57b680112a43319077ad00
 size 184804245

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c90bdde377afa379e25942974e4de5693c679ed32584e381f0995f3b231e8b71
+oid sha256:66f08369bbc62bd74f8b034fc5cc0e38a502d3cef15e435846e5ec2ac283a034
 size 14917

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:833953e669cbb5eb96748291f5173e6f08a3adcb454e0306bb8141ba9e5d2b71
+oid sha256:ba010787a40885fa5288c3b46d8290d7967c71d7c8ba1316b622f7a19a33388b
 size 14917

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ac1c46a2776d12775d23d0f587efc112188137ce2140da35bc15d301c9f620e
+oid sha256:c0033c7745b46bdca3ecab5787678834ca68f7f7e1288869dceeb38812abc253
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfeb39df0143fba24e4d36209079730a02d1b954efa1d4d7185a519ce59f98de
+oid sha256:6ce54f98ea197bcd426469f897eba362d3e592c66e0cecfe82155d3d55c2df9f
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.2765957446808511,
+  "epoch": 1.702127659574468,
   "eval_steps": 500,
-  "global_step": 3000,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -608,6 +608,206 @@
       "mean_token_accuracy": 0.701934365928173,
       "num_tokens": 59164486.0,
       "step": 3000
+    },
+    {
+      "entropy": 1.577785237133503,
+      "epoch": 1.297872340425532,
+      "grad_norm": 0.7289667725563049,
+      "learning_rate": 6.951e-05,
+      "loss": 1.5484317016601563,
+      "mean_token_accuracy": 0.7029052957892418,
+      "num_tokens": 60154400.0,
+      "step": 3050
+    },
+    {
+      "entropy": 1.6015404342114925,
+      "epoch": 1.3191489361702127,
+      "grad_norm": 0.7111889719963074,
+      "learning_rate": 6.901e-05,
+      "loss": 1.567227325439453,
+      "mean_token_accuracy": 0.6997377443313598,
+      "num_tokens": 61135725.0,
+      "step": 3100
+    },
+    {
+      "entropy": 1.5653481243550778,
+      "epoch": 1.3404255319148937,
+      "grad_norm": 0.7378498911857605,
+      "learning_rate": 6.851e-05,
+      "loss": 1.5351031494140626,
+      "mean_token_accuracy": 0.7058432798087597,
+      "num_tokens": 62114114.0,
+      "step": 3150
+    },
+    {
+      "entropy": 1.5650252343714237,
+      "epoch": 1.3617021276595744,
+      "grad_norm": 0.8165695667266846,
+      "learning_rate": 6.801000000000001e-05,
+      "loss": 1.5345574951171874,
+      "mean_token_accuracy": 0.7055312724411488,
+      "num_tokens": 63102105.0,
+      "step": 3200
+    },
+    {
+      "entropy": 1.565962330698967,
+      "epoch": 1.3829787234042552,
+      "grad_norm": 0.7475385665893555,
+      "learning_rate": 6.751000000000001e-05,
+      "loss": 1.532633056640625,
+      "mean_token_accuracy": 0.7054442183673382,
+      "num_tokens": 64092884.0,
+      "step": 3250
+    },
+    {
+      "entropy": 1.5736293095350264,
+      "epoch": 1.4042553191489362,
+      "grad_norm": 0.7410607933998108,
+      "learning_rate": 6.701000000000001e-05,
+      "loss": 1.5408993530273438,
+      "mean_token_accuracy": 0.7043556262552738,
+      "num_tokens": 65077460.0,
+      "step": 3300
+    },
+    {
+      "entropy": 1.5616717764735222,
+      "epoch": 1.425531914893617,
+      "grad_norm": 0.7586492896080017,
+      "learning_rate": 6.651000000000001e-05,
+      "loss": 1.52689453125,
+      "mean_token_accuracy": 0.7062550820410252,
+      "num_tokens": 66073937.0,
+      "step": 3350
+    },
+    {
+      "entropy": 1.557564342021942,
+      "epoch": 1.4468085106382977,
+      "grad_norm": 0.7810468673706055,
+      "learning_rate": 6.601e-05,
+      "loss": 1.5238162231445314,
+      "mean_token_accuracy": 0.7069078533351422,
+      "num_tokens": 67061606.0,
+      "step": 3400
+    },
+    {
+      "entropy": 1.5599790331721306,
+      "epoch": 1.4680851063829787,
+      "grad_norm": 0.7768591046333313,
+      "learning_rate": 6.551e-05,
+      "loss": 1.527623748779297,
+      "mean_token_accuracy": 0.7075633393228054,
+      "num_tokens": 68050209.0,
+      "step": 3450
+    },
+    {
+      "entropy": 1.5402766379714012,
+      "epoch": 1.4893617021276595,
+      "grad_norm": 0.7596350908279419,
+      "learning_rate": 6.501e-05,
+      "loss": 1.5053843688964843,
+      "mean_token_accuracy": 0.709761653393507,
+      "num_tokens": 69043340.0,
+      "step": 3500
+    },
+    {
+      "entropy": 1.5391169971227645,
+      "epoch": 1.5106382978723403,
+      "grad_norm": 0.7082200050354004,
+      "learning_rate": 6.451e-05,
+      "loss": 1.5009793090820311,
+      "mean_token_accuracy": 0.7094894219934941,
+      "num_tokens": 70028141.0,
+      "step": 3550
+    },
+    {
+      "entropy": 1.5485185173153877,
+      "epoch": 1.5319148936170213,
+      "grad_norm": 0.7340750098228455,
+      "learning_rate": 6.401e-05,
+      "loss": 1.5126968383789063,
+      "mean_token_accuracy": 0.7095904792845249,
+      "num_tokens": 71021512.0,
+      "step": 3600
+    },
+    {
+      "entropy": 1.5227306772768499,
+      "epoch": 1.5531914893617023,
+      "grad_norm": 0.7728723287582397,
+      "learning_rate": 6.351000000000001e-05,
+      "loss": 1.4854464721679688,
+      "mean_token_accuracy": 0.7113837064802646,
+      "num_tokens": 72010002.0,
+      "step": 3650
+    },
+    {
+      "entropy": 1.5513358226418494,
+      "epoch": 1.574468085106383,
+      "grad_norm": 0.7605530023574829,
+      "learning_rate": 6.301000000000001e-05,
+      "loss": 1.5147122192382811,
+      "mean_token_accuracy": 0.7083025954663753,
+      "num_tokens": 72997531.0,
+      "step": 3700
+    },
+    {
+      "entropy": 1.5391620790958405,
+      "epoch": 1.5957446808510638,
+      "grad_norm": 0.7805915474891663,
+      "learning_rate": 6.251e-05,
+      "loss": 1.5053895568847657,
+      "mean_token_accuracy": 0.7098572531342506,
+      "num_tokens": 73981617.0,
+      "step": 3750
+    },
+    {
+      "entropy": 1.5519435784220696,
+      "epoch": 1.6170212765957448,
+      "grad_norm": 0.7576701045036316,
+      "learning_rate": 6.201e-05,
+      "loss": 1.5180355834960937,
+      "mean_token_accuracy": 0.7073439501225949,
+      "num_tokens": 74975613.0,
+      "step": 3800
+    },
+    {
+      "entropy": 1.5172939459979533,
+      "epoch": 1.6382978723404256,
+      "grad_norm": 0.7765750288963318,
+      "learning_rate": 6.151e-05,
+      "loss": 1.4821212768554688,
+      "mean_token_accuracy": 0.7136698444187641,
+      "num_tokens": 75966971.0,
+      "step": 3850
+    },
+    {
+      "entropy": 1.5298295491933822,
+      "epoch": 1.6595744680851063,
+      "grad_norm": 0.7504763007164001,
+      "learning_rate": 6.101e-05,
+      "loss": 1.492011260986328,
+      "mean_token_accuracy": 0.7108374397456646,
+      "num_tokens": 76950920.0,
+      "step": 3900
+    },
+    {
+      "entropy": 1.5399506105482579,
+      "epoch": 1.6808510638297873,
+      "grad_norm": 0.7663020491600037,
+      "learning_rate": 6.051e-05,
+      "loss": 1.5039883422851563,
+      "mean_token_accuracy": 0.7101991818845272,
+      "num_tokens": 77939066.0,
+      "step": 3950
+    },
+    {
+      "entropy": 1.5202897056937217,
+      "epoch": 1.702127659574468,
+      "grad_norm": 0.8080466985702515,
+      "learning_rate": 6.0009999999999996e-05,
+      "loss": 1.485948028564453,
+      "mean_token_accuracy": 0.7131342147290707,
+      "num_tokens": 78916374.0,
+      "step": 4000
     }
   ],
   "logging_steps": 50,
@@ -627,7 +827,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.496366043450573e+17,
+  "total_flos": 5.994971121907139e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null