Training in progress, step 4000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e844c140f0b364ef240d9d283f3474471aa887e9a4e7f8afe6ca0d5062fa1d8
+oid sha256:eec6ae3c232115c0c2fe4bcf646b1c03138328548b645022df7b55df2d6ff03b
 size 847599616

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6e57c8be8d618705f188b94fc5169164f579e23fc7d687c76672146f866dfc5
+oid sha256:2c8dc5c30771f73079678db63765b219aafca9ea4b8e33ef2987d0d846fa62d3
 size 1386414411

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cd45c472c02f69686aade1ec7cee7512c8dba6e6bde5e0bd023fd4f8f2f3548
+oid sha256:7864fb6e928e45d4215a1ae96eab6e34cbc82f54e4a38e3cf2e1449f98239105
 size 14917

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09714a2ef70dd987c21ebd404103aea88be622e980ab90fe03c4cb22a4d6d0de
+oid sha256:bb8ff3d7c5e846455471242a9719dd9767ec3729631e3efca61eaecbd0ad4ee7
 size 14917

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0ed1549b52af39b27c28b24276ca81b4d37dad6af883ad09a3c1268b5bcb1b0
+oid sha256:561e0049aa6e5e793844e867537a9c947ef9ba1ad03fe6ba44d328d1dffc9702
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.13333333333333333,
+  "epoch": 0.26666666666666666,
   "eval_steps": 1000,
-  "global_step": 2000,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -166,6 +166,164 @@
       "eval_samples_per_second": 7.646,
       "eval_steps_per_second": 0.382,
       "step": 2000
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 1.456148386001587,
+      "learning_rate": 0.00039122008179569447,
+      "loss": 10.4031689453125,
+      "step": 2100
+    },
+    {
+      "epoch": 0.14666666666666667,
+      "grad_norm": 1.4301263093948364,
+      "learning_rate": 0.00038988163405964595,
+      "loss": 10.173363037109375,
+      "step": 2200
+    },
+    {
+      "epoch": 0.15333333333333332,
+      "grad_norm": 1.0371617078781128,
+      "learning_rate": 0.00038845090041619265,
+      "loss": 10.063040771484374,
+      "step": 2300
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 1.0449143648147583,
+      "learning_rate": 0.0003869285762277543,
+      "loss": 10.145305786132813,
+      "step": 2400
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 1.491047739982605,
+      "learning_rate": 0.00038531540137141165,
+      "loss": 10.095411376953125,
+      "step": 2500
+    },
+    {
+      "epoch": 0.17333333333333334,
+      "grad_norm": 1.216232419013977,
+      "learning_rate": 0.0003836121598793126,
+      "loss": 10.10360107421875,
+      "step": 2600
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 1.0485482215881348,
+      "learning_rate": 0.0003818196795576189,
+      "loss": 9.9664892578125,
+      "step": 2700
+    },
+    {
+      "epoch": 0.18666666666666668,
+      "grad_norm": 1.1029548645019531,
+      "learning_rate": 0.00037993883158417654,
+      "loss": 9.9088525390625,
+      "step": 2800
+    },
+    {
+      "epoch": 0.19333333333333333,
+      "grad_norm": 1.0070501565933228,
+      "learning_rate": 0.00037797053008510834,
+      "loss": 9.888450927734375,
+      "step": 2900
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 1.1356112957000732,
+      "learning_rate": 0.00037591573169053167,
+      "loss": 9.8390869140625,
+      "step": 3000
+    },
+    {
+      "epoch": 0.2,
+      "eval_accuracy": 0.2266908023483366,
+      "eval_loss": 9.943662643432617,
+      "eval_runtime": 65.3824,
+      "eval_samples_per_second": 7.647,
+      "eval_steps_per_second": 0.382,
+      "step": 3000
+    },
+    {
+      "epoch": 0.20666666666666667,
+      "grad_norm": 1.178817629814148,
+      "learning_rate": 0.0003737754350696192,
+      "loss": 9.806249389648437,
+      "step": 3100
+    },
+    {
+      "epoch": 0.21333333333333335,
+      "grad_norm": 1.1421138048171997,
+      "learning_rate": 0.00037155068044522735,
+      "loss": 9.715457763671875,
+      "step": 3200
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 1.6934456825256348,
+      "learning_rate": 0.00036924254908832937,
+      "loss": 9.71255859375,
+      "step": 3300
+    },
+    {
+      "epoch": 0.22666666666666666,
+      "grad_norm": 1.1598275899887085,
+      "learning_rate": 0.00036685216279249815,
+      "loss": 9.716998291015624,
+      "step": 3400
+    },
+    {
+      "epoch": 0.23333333333333334,
+      "grad_norm": 1.1768417358398438,
+      "learning_rate": 0.000364380683328694,
+      "loss": 9.557822875976562,
+      "step": 3500
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 1.4840753078460693,
+      "learning_rate": 0.0003618293118806232,
+      "loss": 9.56420654296875,
+      "step": 3600
+    },
+    {
+      "epoch": 0.24666666666666667,
+      "grad_norm": 1.2799503803253174,
+      "learning_rate": 0.00035919928846094094,
+      "loss": 9.548779296875,
+      "step": 3700
+    },
+    {
+      "epoch": 0.25333333333333335,
+      "grad_norm": 1.1078740358352661,
+      "learning_rate": 0.00035649189130858256,
+      "loss": 9.44154541015625,
+      "step": 3800
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 1.173609972000122,
+      "learning_rate": 0.00035370843626751663,
+      "loss": 9.414360961914063,
+      "step": 3900
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "grad_norm": 1.2394672632217407,
+      "learning_rate": 0.00035085027614722075,
+      "loss": 9.300681762695312,
+      "step": 4000
+    },
+    {
+      "epoch": 0.26666666666666666,
+      "eval_accuracy": 0.2545146771037182,
+      "eval_loss": 9.379138946533203,
+      "eval_runtime": 65.4591,
+      "eval_samples_per_second": 7.638,
+      "eval_steps_per_second": 0.382,
+      "step": 4000
     }
   ],
   "logging_steps": 100,