Training in progress, step 25, checkpoint

last-checkpoint/adapter_config.json

@@ -20,12 +20,12 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
     "o_proj",
-    "k_proj",
-    "gate_proj",
-    "up_proj",
     "v_proj",
+    "up_proj",
+    "gate_proj",
+    "k_proj",
+    "down_proj",
     "q_proj"
   ],
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d0f66c6a47a25f385173d726904c546f03df4d695f3eb12ed84839a91227b0de
+oid sha256:b5a1958c57fa8a3e075e39de100457523fe983665ef747cd214c2060014226c0
 size 200068512

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7406bfe3f0274b2db7be65887c5911aeba64eced0bd24782e7d038f2c70e82c1
+oid sha256:b7bab43b300f4949a336e8d5f11c091fb566f7cad43eb09ab5f37c35196c0d97
 size 400361770

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0aa299d6b618ba4f73d7d6c4d99f3591b3ed1a9520f2ccb1ca96d7f9f3c7936d
+oid sha256:e87cdd914cacce0bf0f2d3377e1ea0cf41243706dad7f4209963b8c2a7ac5a75
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e69e2b49ea642509f0c688c16fb190b7cf27dac0a18903a5e2d1467d0343d8b8
+oid sha256:040b95482c646935022d148ebc91f462fbf2195cfa3365adbbe3bd6ca1f35a74
 size 1064

last-checkpoint/trainer_state.json

@@ -1,16 +1,16 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.04081632653061224,
+  "epoch": 0.02040816326530612,
   "eval_steps": 5,
-  "global_step": 50,
+  "global_step": 25,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.0008163265306122449,
-      "grad_norm": 0.7370944619178772,
+      "grad_norm": 0.7544206976890564,
       "learning_rate": 2e-05,
       "loss": 0.5845,
       "step": 1
@@ -18,433 +18,218 @@
     {
       "epoch": 0.0008163265306122449,
       "eval_loss": 0.6241262555122375,
-      "eval_runtime": 303.1464,
-      "eval_samples_per_second": 3.404,
-      "eval_steps_per_second": 1.702,
+      "eval_runtime": 303.7147,
+      "eval_samples_per_second": 3.398,
+      "eval_steps_per_second": 1.699,
       "step": 1
     },
     {
       "epoch": 0.0016326530612244899,
-      "grad_norm": 0.5706467628479004,
+      "grad_norm": 0.5853733420372009,
       "learning_rate": 4e-05,
       "loss": 0.5365,
       "step": 2
     },
     {
       "epoch": 0.0024489795918367346,
-      "grad_norm": 0.633495569229126,
+      "grad_norm": 0.6549996733665466,
       "learning_rate": 6e-05,
-      "loss": 0.6398,
+      "loss": 0.6425,
       "step": 3
     },
     {
       "epoch": 0.0032653061224489797,
-      "grad_norm": 0.6009469032287598,
+      "grad_norm": 0.614408552646637,
       "learning_rate": 8e-05,
-      "loss": 0.5877,
+      "loss": 0.5795,
       "step": 4
     },
     {
       "epoch": 0.004081632653061225,
-      "grad_norm": 0.5381947159767151,
+      "grad_norm": 0.5475334525108337,
       "learning_rate": 0.0001,
-      "loss": 0.5462,
+      "loss": 0.5454,
       "step": 5
     },
     {
       "epoch": 0.004081632653061225,
-      "eval_loss": 0.5509580969810486,
-      "eval_runtime": 305.7326,
-      "eval_samples_per_second": 3.375,
-      "eval_steps_per_second": 1.688,
+      "eval_loss": 0.5514864325523376,
+      "eval_runtime": 306.4705,
+      "eval_samples_per_second": 3.367,
+      "eval_steps_per_second": 1.684,
       "step": 5
     },
     {
       "epoch": 0.004897959183673469,
-      "grad_norm": 0.5985301733016968,
+      "grad_norm": 0.6220462918281555,
       "learning_rate": 0.00012,
-      "loss": 0.5205,
+      "loss": 0.5158,
       "step": 6
     },
     {
       "epoch": 0.005714285714285714,
-      "grad_norm": 0.5854198336601257,
+      "grad_norm": 0.6376144289970398,
       "learning_rate": 0.00014,
-      "loss": 0.4624,
+      "loss": 0.4644,
       "step": 7
     },
     {
       "epoch": 0.006530612244897959,
-      "grad_norm": 0.7701842784881592,
+      "grad_norm": 0.7968347668647766,
       "learning_rate": 0.00016,
-      "loss": 0.4551,
+      "loss": 0.4495,
       "step": 8
     },
     {
       "epoch": 0.007346938775510204,
-      "grad_norm": 0.656694769859314,
+      "grad_norm": 0.6648410558700562,
       "learning_rate": 0.00018,
-      "loss": 0.4233,
+      "loss": 0.4173,
       "step": 9
     },
     {
       "epoch": 0.00816326530612245,
-      "grad_norm": 0.8429622054100037,
+      "grad_norm": 0.8778969645500183,
       "learning_rate": 0.0002,
-      "loss": 0.487,
+      "loss": 0.4878,
       "step": 10
     },
     {
       "epoch": 0.00816326530612245,
-      "eval_loss": 0.4371676445007324,
-      "eval_runtime": 305.7536,
-      "eval_samples_per_second": 3.375,
-      "eval_steps_per_second": 1.688,
+      "eval_loss": 0.43802592158317566,
+      "eval_runtime": 306.6738,
+      "eval_samples_per_second": 3.365,
+      "eval_steps_per_second": 1.683,
       "step": 10
     },
     {
       "epoch": 0.008979591836734694,
-      "grad_norm": 0.6876599192619324,
+      "grad_norm": 0.7390214800834656,
       "learning_rate": 0.0001996917333733128,
-      "loss": 0.4084,
+      "loss": 0.4016,
       "step": 11
     },
     {
       "epoch": 0.009795918367346938,
-      "grad_norm": 0.5299339890480042,
+      "grad_norm": 0.5307807922363281,
       "learning_rate": 0.00019876883405951377,
-      "loss": 0.4388,
+      "loss": 0.439,
       "step": 12
     },
     {
       "epoch": 0.010612244897959184,
-      "grad_norm": 0.5224944353103638,
+      "grad_norm": 0.515744686126709,
       "learning_rate": 0.00019723699203976766,
-      "loss": 0.3741,
+      "loss": 0.37,
       "step": 13
     },
     {
       "epoch": 0.011428571428571429,
-      "grad_norm": 0.5660629868507385,
+      "grad_norm": 0.5767286419868469,
       "learning_rate": 0.00019510565162951537,
-      "loss": 0.3746,
+      "loss": 0.3815,
       "step": 14
     },
     {
       "epoch": 0.012244897959183673,
-      "grad_norm": 0.4487418532371521,
+      "grad_norm": 0.4507094919681549,
       "learning_rate": 0.0001923879532511287,
-      "loss": 0.3684,
+      "loss": 0.3675,
       "step": 15
     },
     {
       "epoch": 0.012244897959183673,
-      "eval_loss": 0.3932255804538727,
-      "eval_runtime": 306.0955,
-      "eval_samples_per_second": 3.371,
-      "eval_steps_per_second": 1.686,
+      "eval_loss": 0.3925292193889618,
+      "eval_runtime": 306.4671,
+      "eval_samples_per_second": 3.367,
+      "eval_steps_per_second": 1.684,
       "step": 15
     },
     {
       "epoch": 0.013061224489795919,
-      "grad_norm": 0.4456219971179962,
+      "grad_norm": 0.45942962169647217,
       "learning_rate": 0.0001891006524188368,
-      "loss": 0.4175,
+      "loss": 0.4134,
       "step": 16
     },
     {
       "epoch": 0.013877551020408163,
-      "grad_norm": 0.46617552638053894,
+      "grad_norm": 0.4870969355106354,
       "learning_rate": 0.00018526401643540922,
-      "loss": 0.3896,
+      "loss": 0.389,
       "step": 17
     },
     {
       "epoch": 0.014693877551020407,
-      "grad_norm": 0.39636608958244324,
+      "grad_norm": 0.4066920876502991,
       "learning_rate": 0.00018090169943749476,
-      "loss": 0.3915,
+      "loss": 0.4022,
       "step": 18
     },
     {
       "epoch": 0.015510204081632653,
-      "grad_norm": 0.42811569571495056,
+      "grad_norm": 0.44875219464302063,
       "learning_rate": 0.0001760405965600031,
-      "loss": 0.3479,
+      "loss": 0.3507,
       "step": 19
     },
     {
       "epoch": 0.0163265306122449,
-      "grad_norm": 0.4594975411891937,
+      "grad_norm": 0.49346503615379333,
       "learning_rate": 0.00017071067811865476,
-      "loss": 0.3663,
+      "loss": 0.3737,
       "step": 20
     },
     {
       "epoch": 0.0163265306122449,
-      "eval_loss": 0.3825583755970001,
-      "eval_runtime": 305.9194,
-      "eval_samples_per_second": 3.373,
-      "eval_steps_per_second": 1.687,
+      "eval_loss": 0.38162362575531006,
+      "eval_runtime": 306.1047,
+      "eval_samples_per_second": 3.371,
+      "eval_steps_per_second": 1.686,
       "step": 20
     },
     {
       "epoch": 0.017142857142857144,
-      "grad_norm": 0.42077338695526123,
+      "grad_norm": 0.42135682702064514,
       "learning_rate": 0.00016494480483301836,
-      "loss": 0.3459,
+      "loss": 0.3401,
       "step": 21
     },
     {
       "epoch": 0.017959183673469388,
-      "grad_norm": 0.448321670293808,
+      "grad_norm": 0.45447424054145813,
       "learning_rate": 0.00015877852522924732,
-      "loss": 0.3624,
+      "loss": 0.3611,
       "step": 22
     },
     {
       "epoch": 0.018775510204081632,
-      "grad_norm": 0.40496572852134705,
+      "grad_norm": 0.4188750982284546,
       "learning_rate": 0.0001522498564715949,
-      "loss": 0.2865,
+      "loss": 0.2922,
       "step": 23
     },
     {
       "epoch": 0.019591836734693877,
-      "grad_norm": 0.4883009195327759,
+      "grad_norm": 0.5015071630477905,
       "learning_rate": 0.00014539904997395468,
-      "loss": 0.3692,
+      "loss": 0.3682,
       "step": 24
     },
     {
       "epoch": 0.02040816326530612,
-      "grad_norm": 0.47144022583961487,
+      "grad_norm": 0.4825016260147095,
       "learning_rate": 0.000138268343236509,
-      "loss": 0.3643,
+      "loss": 0.367,
       "step": 25
     },
     {
       "epoch": 0.02040816326530612,
-      "eval_loss": 0.3697856366634369,
-      "eval_runtime": 306.009,
-      "eval_samples_per_second": 3.372,
-      "eval_steps_per_second": 1.686,
+      "eval_loss": 0.3707207441329956,
+      "eval_runtime": 306.4625,
+      "eval_samples_per_second": 3.367,
+      "eval_steps_per_second": 1.684,
       "step": 25
-    },
-    {
-      "epoch": 0.02122448979591837,
-      "grad_norm": 0.4498879015445709,
-      "learning_rate": 0.00013090169943749476,
-      "loss": 0.3263,
-      "step": 26
-    },
-    {
-      "epoch": 0.022040816326530613,
-      "grad_norm": 0.46243250370025635,
-      "learning_rate": 0.00012334453638559057,
-      "loss": 0.4412,
-      "step": 27
-    },
-    {
-      "epoch": 0.022857142857142857,
-      "grad_norm": 0.48129433393478394,
-      "learning_rate": 0.0001156434465040231,
-      "loss": 0.3595,
-      "step": 28
-    },
-    {
-      "epoch": 0.0236734693877551,
-      "grad_norm": 0.41295281052589417,
-      "learning_rate": 0.0001078459095727845,
-      "loss": 0.3386,
-      "step": 29
-    },
-    {
-      "epoch": 0.024489795918367346,
-      "grad_norm": 0.42476311326026917,
-      "learning_rate": 0.0001,
-      "loss": 0.342,
-      "step": 30
-    },
-    {
-      "epoch": 0.024489795918367346,
-      "eval_loss": 0.3653227388858795,
-      "eval_runtime": 305.8828,
-      "eval_samples_per_second": 3.374,
-      "eval_steps_per_second": 1.687,
-      "step": 30
-    },
-    {
-      "epoch": 0.025306122448979593,
-      "grad_norm": 0.40272751450538635,
-      "learning_rate": 9.215409042721552e-05,
-      "loss": 0.4082,
-      "step": 31
-    },
-    {
-      "epoch": 0.026122448979591838,
-      "grad_norm": 0.38805681467056274,
-      "learning_rate": 8.435655349597689e-05,
-      "loss": 0.2459,
-      "step": 32
-    },
-    {
-      "epoch": 0.026938775510204082,
-      "grad_norm": 0.4278540015220642,
-      "learning_rate": 7.66554636144095e-05,
-      "loss": 0.3213,
-      "step": 33
-    },
-    {
-      "epoch": 0.027755102040816326,
-      "grad_norm": 0.574360728263855,
-      "learning_rate": 6.909830056250527e-05,
-      "loss": 0.42,
-      "step": 34
-    },
-    {
-      "epoch": 0.02857142857142857,
-      "grad_norm": 0.4692784547805786,
-      "learning_rate": 6.173165676349103e-05,
-      "loss": 0.3466,
-      "step": 35
-    },
-    {
-      "epoch": 0.02857142857142857,
-      "eval_loss": 0.3594253957271576,
-      "eval_runtime": 305.9204,
-      "eval_samples_per_second": 3.373,
-      "eval_steps_per_second": 1.687,
-      "step": 35
-    },
-    {
-      "epoch": 0.029387755102040815,
-      "grad_norm": 0.40394076704978943,
-      "learning_rate": 5.4600950026045326e-05,
-      "loss": 0.3218,
-      "step": 36
-    },
-    {
-      "epoch": 0.030204081632653063,
-      "grad_norm": 0.40415486693382263,
-      "learning_rate": 4.7750143528405126e-05,
-      "loss": 0.3449,
-      "step": 37
-    },
-    {
-      "epoch": 0.031020408163265307,
-      "grad_norm": 0.5078446865081787,
-      "learning_rate": 4.12214747707527e-05,
-      "loss": 0.3804,
-      "step": 38
-    },
-    {
-      "epoch": 0.03183673469387755,
-      "grad_norm": 0.4428996443748474,
-      "learning_rate": 3.5055195166981645e-05,
-      "loss": 0.4278,
-      "step": 39
-    },
-    {
-      "epoch": 0.0326530612244898,
-      "grad_norm": 0.5538510084152222,
-      "learning_rate": 2.9289321881345254e-05,
-      "loss": 0.53,
-      "step": 40
-    },
-    {
-      "epoch": 0.0326530612244898,
-      "eval_loss": 0.3550981879234314,
-      "eval_runtime": 305.9407,
-      "eval_samples_per_second": 3.373,
-      "eval_steps_per_second": 1.687,
-      "step": 40
-    },
-    {
-      "epoch": 0.03346938775510204,
-      "grad_norm": 0.5039154887199402,
-      "learning_rate": 2.3959403439996907e-05,
-      "loss": 0.3739,
-      "step": 41
-    },
-    {
-      "epoch": 0.03428571428571429,
-      "grad_norm": 0.4173552095890045,
-      "learning_rate": 1.9098300562505266e-05,
-      "loss": 0.3184,
-      "step": 42
-    },
-    {
-      "epoch": 0.03510204081632653,
-      "grad_norm": 0.3404304087162018,
-      "learning_rate": 1.4735983564590783e-05,
-      "loss": 0.3103,
-      "step": 43
-    },
-    {
-      "epoch": 0.035918367346938776,
-      "grad_norm": 0.4408942759037018,
-      "learning_rate": 1.0899347581163221e-05,
-      "loss": 0.3103,
-      "step": 44
-    },
-    {
-      "epoch": 0.036734693877551024,
-      "grad_norm": 0.4228384792804718,
-      "learning_rate": 7.612046748871327e-06,
-      "loss": 0.328,
-      "step": 45
-    },
-    {
-      "epoch": 0.036734693877551024,
-      "eval_loss": 0.3556883931159973,
-      "eval_runtime": 306.0718,
-      "eval_samples_per_second": 3.372,
-      "eval_steps_per_second": 1.686,
-      "step": 45
-    },
-    {
-      "epoch": 0.037551020408163265,
-      "grad_norm": 0.36199188232421875,
-      "learning_rate": 4.8943483704846475e-06,
-      "loss": 0.2883,
-      "step": 46
-    },
-    {
-      "epoch": 0.03836734693877551,
-      "grad_norm": 0.4111153483390808,
-      "learning_rate": 2.7630079602323442e-06,
-      "loss": 0.2967,
-      "step": 47
-    },
-    {
-      "epoch": 0.03918367346938775,
-      "grad_norm": 0.3962281346321106,
-      "learning_rate": 1.231165940486234e-06,
-      "loss": 0.3148,
-      "step": 48
-    },
-    {
-      "epoch": 0.04,
-      "grad_norm": 0.41017216444015503,
-      "learning_rate": 3.0826662668720364e-07,
-      "loss": 0.3113,
-      "step": 49
-    },
-    {
-      "epoch": 0.04081632653061224,
-      "grad_norm": 0.4169033467769623,
-      "learning_rate": 0.0,
-      "loss": 0.3195,
-      "step": 50
-    },
-    {
-      "epoch": 0.04081632653061224,
-      "eval_loss": 0.35500943660736084,
-      "eval_runtime": 306.0822,
-      "eval_samples_per_second": 3.372,
-      "eval_steps_per_second": 1.686,
-      "step": 50
     }
   ],
   "logging_steps": 1,
@@ -459,12 +244,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 7.707288333503693e+16,
+  "total_flos": 3.83447180771328e+16,
   "train_batch_size": 2,
   "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:ab4f2c6f49088c35f2e82a3466d0efce6faeea0c8f475325fa63f3479bc8ed25
+oid sha256:fe1e904a16c4b011350bab7b53595b43a35f758347a5d7fc1ee6fa65abe253d8
 size 6776