Training in progress, epoch 4, checkpoint

checkpoint-14396/README.md

@@ -199,4 +199,4 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
 [More Information Needed]
 ### Framework versions
 
-- PEFT 0.15.2
+- PEFT 0.14.0

checkpoint-14396/adapter_config.json

@@ -3,7 +3,6 @@
   "auto_mapping": null,
   "base_model_name_or_path": "deepseek-ai/deepseek-coder-1.3b-base",
   "bias": "none",
-  "corda_config": null,
   "eva_config": null,
   "exclude_modules": null,
   "fan_in_fan_out": false,
@@ -24,11 +23,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "v_proj"
+    "v_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
-  "trainable_token_indices": null,
   "use_dora": false,
   "use_rslora": false
 }

checkpoint-14396/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4552003d4aed2d171aa12f13b56bd4e3d1951140354c130a7fb7a976463fcbfd
+oid sha256:152f2bb5b2c3b4788f9cefdce5400dcd3946a55af39556c3ca1c4bf4503405e9
 size 268636736

checkpoint-14396/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:acd872dd1a8406eacac2032410f7f8612d7ae59d4b6975f580f2b9a5dc7d0367
-size 12664203
+oid sha256:f93ee4c24f5cd936e262090acef3a06daf13d03862f8de2e35d1bb01a73d2505
+size 12663802

checkpoint-14396/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd76de073554bab3bf63798d356c83633c951117c52b2cf2f8c3b3df2e5fc056
-size 16389
+oid sha256:94bae8dd697775680906ac882bbf81037c8e3544d815f4eaf6a76d6b00185681
+size 15984

checkpoint-14396/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bc4fb5bc79880cb4a7fe50e20a14b4f2f222f8494b761f2e3593c473e32c4284
-size 16389
+oid sha256:ece8d5367228c825aeea47316a0f12ddc1816ee7751e15259af27deec7a9c2ec
+size 15984

checkpoint-14396/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6592a1a62f251522f65c6fe02fc9c35301d0d572d89e215bb10e178bcf02bd25
-size 16389
+oid sha256:6c23b1f15395bdbd944711bfaf443482928daa5c26e3afda09c62d27178b8ec1
+size 15984

checkpoint-14396/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c4ec9a9627bcfaf87c9421a31dda169e0b8540242f246786eeef3d43de2d0a4b
-size 16389
+oid sha256:f1a92583d6564824e05f416e7e8d1ebde7085ce28d371da6ffeda8caedf9ea35
+size 15984

checkpoint-14396/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4360d47b9701f2986ae6c729c0c6f4b6d156d3dfa4b45b4901555c0402265fc4
-size 16389
+oid sha256:6bc29c3bea4d3701298bcb2c6c9b76ca6072fe61c77e46a8e97be7b4a8bc334b
+size 15984

checkpoint-14396/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5181ad7110d04115d21787efc2d289d0e3a40efbf2103bacf37524485df12fe
-size 16389
+oid sha256:5e2ad477cc000f37aadc901fc70b5137d99d418a9be8902fd8cee43d18c08817
+size 15984

checkpoint-14396/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3182ad80745fdc9a5aaa3649d4caddf06da689f4dc561247ca4910c5f480fc7
-size 16389
+oid sha256:c8eff4df6699f83f3c04b19747d954c1c9115ae27f8de210dbb6df92e444317a
+size 15984

checkpoint-14396/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f144a013c3292cdfe94f8d0f8c10b74c5f0a8a0f2951d181cac8d83eb0c5acab
-size 16389
+oid sha256:adb69ab24f0f80a4316db2fa5b4c4b03b4e76b19e6d6c6d186489fc4ce860e1a
+size 15984

checkpoint-14396/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2bd52e0e7759a636c9051712edaf53f879b6f7024df637a06eafa124d9a5f794
-size 1465
+oid sha256:308dee51aa198e8fb7d189ea73c50c27f265697af57fe64c9f15c781303073d4
+size 1064

checkpoint-14396/trainer_state.json

@@ -1,5 +1,4 @@
 {
-  "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 4.0,
@@ -11,350 +10,350 @@
   "log_history": [
     {
       "epoch": 0.13892747985551543,
-      "grad_norm": 0.6735790967941284,
-      "learning_rate": 0.0007907937390015746,
-      "loss": 0.3742,
+      "grad_norm": 0.9573676586151123,
+      "learning_rate": 0.0007907566916736131,
+      "loss": 0.377,
       "step": 500
     },
     {
       "epoch": 0.2000555709919422,
-      "eval_loss": 0.2844131588935852,
-      "eval_runtime": 5.091,
-      "eval_samples_per_second": 98.212,
-      "eval_steps_per_second": 6.286,
+      "eval_loss": 0.28939542174339294,
+      "eval_runtime": 4.6518,
+      "eval_samples_per_second": 107.484,
+      "eval_steps_per_second": 6.879,
       "step": 720
     },
     {
       "epoch": 0.27785495971103086,
-      "grad_norm": 0.6214504241943359,
-      "learning_rate": 0.0007815319070112068,
-      "loss": 0.2859,
+      "grad_norm": 0.5571001768112183,
+      "learning_rate": 0.0007815133833472261,
+      "loss": 0.2824,
       "step": 1000
     },
     {
       "epoch": 0.4001111419838844,
-      "eval_loss": 0.2708466649055481,
-      "eval_runtime": 4.8349,
-      "eval_samples_per_second": 103.416,
-      "eval_steps_per_second": 6.619,
+      "eval_loss": 0.2604603171348572,
+      "eval_runtime": 4.3083,
+      "eval_samples_per_second": 116.054,
+      "eval_steps_per_second": 7.427,
       "step": 1440
     },
     {
       "epoch": 0.41678243956654626,
-      "grad_norm": 0.5057302117347717,
-      "learning_rate": 0.0007722700750208392,
-      "loss": 0.2679,
+      "grad_norm": 0.6729081273078918,
+      "learning_rate": 0.0007722515513568584,
+      "loss": 0.2627,
       "step": 1500
     },
     {
       "epoch": 0.5557099194220617,
-      "grad_norm": 0.623817503452301,
-      "learning_rate": 0.0007630082430304714,
-      "loss": 0.2527,
+      "grad_norm": 0.5200651288032532,
+      "learning_rate": 0.0007629897193664907,
+      "loss": 0.2488,
       "step": 2000
     },
     {
       "epoch": 0.6001667129758266,
-      "eval_loss": 0.24877774715423584,
-      "eval_runtime": 5.9656,
-      "eval_samples_per_second": 83.813,
-      "eval_steps_per_second": 5.364,
+      "eval_loss": 0.24864767491817474,
+      "eval_runtime": 3.8713,
+      "eval_samples_per_second": 129.157,
+      "eval_steps_per_second": 8.266,
       "step": 2160
     },
     {
       "epoch": 0.6946373992775771,
-      "grad_norm": 1.3911620378494263,
-      "learning_rate": 0.0007537649347040845,
-      "loss": 0.2441,
+      "grad_norm": 0.6645380258560181,
+      "learning_rate": 0.0007537464110401038,
+      "loss": 0.2424,
       "step": 2500
     },
     {
       "epoch": 0.8002222839677688,
-      "eval_loss": 0.2391277253627777,
-      "eval_runtime": 5.9872,
-      "eval_samples_per_second": 83.512,
-      "eval_steps_per_second": 5.345,
+      "eval_loss": 0.2434273064136505,
+      "eval_runtime": 4.2012,
+      "eval_samples_per_second": 119.013,
+      "eval_steps_per_second": 7.617,
       "step": 2880
     },
     {
       "epoch": 0.8335648791330925,
-      "grad_norm": 0.7922923564910889,
-      "learning_rate": 0.0007445216263776975,
-      "loss": 0.235,
+      "grad_norm": 0.8886778354644775,
+      "learning_rate": 0.000744484579049736,
+      "loss": 0.2332,
       "step": 3000
     },
     {
       "epoch": 0.972492358988608,
-      "grad_norm": 0.8539218902587891,
-      "learning_rate": 0.0007352597943873298,
-      "loss": 0.2331,
+      "grad_norm": 0.5990277528762817,
+      "learning_rate": 0.0007352227470593684,
+      "loss": 0.2307,
       "step": 3500
     },
     {
       "epoch": 1.000277854959711,
-      "eval_loss": 0.23887279629707336,
-      "eval_runtime": 5.6004,
-      "eval_samples_per_second": 89.279,
-      "eval_steps_per_second": 5.714,
+      "eval_loss": 0.23320984840393066,
+      "eval_runtime": 4.3777,
+      "eval_samples_per_second": 114.216,
+      "eval_steps_per_second": 7.31,
       "step": 3600
     },
     {
       "epoch": 1.1114198388441234,
-      "grad_norm": 0.6812190413475037,
-      "learning_rate": 0.0007259979623969622,
-      "loss": 0.2201,
+      "grad_norm": 0.9014330506324768,
+      "learning_rate": 0.0007259609150690007,
+      "loss": 0.2171,
       "step": 4000
     },
     {
       "epoch": 1.2003334259516532,
-      "eval_loss": 0.2331564724445343,
-      "eval_runtime": 5.2457,
-      "eval_samples_per_second": 95.317,
-      "eval_steps_per_second": 6.1,
+      "eval_loss": 0.23165521025657654,
+      "eval_runtime": 4.5533,
+      "eval_samples_per_second": 109.81,
+      "eval_steps_per_second": 7.028,
       "step": 4320
     },
     {
       "epoch": 1.2503473186996388,
-      "grad_norm": 0.6926279067993164,
-      "learning_rate": 0.0007167361304065945,
-      "loss": 0.2199,
+      "grad_norm": 0.9070354104042053,
+      "learning_rate": 0.000716699083078633,
+      "loss": 0.2174,
       "step": 4500
     },
     {
       "epoch": 1.3892747985551543,
-      "grad_norm": 0.857128918170929,
-      "learning_rate": 0.0007074742984162268,
-      "loss": 0.2157,
+      "grad_norm": 0.9023842215538025,
+      "learning_rate": 0.000707455774752246,
+      "loss": 0.215,
       "step": 5000
     },
     {
       "epoch": 1.4003889969435954,
-      "eval_loss": 0.22209732234477997,
-      "eval_runtime": 4.9003,
-      "eval_samples_per_second": 102.035,
-      "eval_steps_per_second": 6.53,
+      "eval_loss": 0.21996016800403595,
+      "eval_runtime": 4.2649,
+      "eval_samples_per_second": 117.235,
+      "eval_steps_per_second": 7.503,
       "step": 5040
     },
     {
       "epoch": 1.5282022784106695,
-      "grad_norm": 0.8350584506988525,
-      "learning_rate": 0.000698212466425859,
-      "loss": 0.2122,
+      "grad_norm": 0.8724926710128784,
+      "learning_rate": 0.0006981939427618783,
+      "loss": 0.2101,
       "step": 5500
     },
     {
       "epoch": 1.6004445679355377,
-      "eval_loss": 0.22165100276470184,
-      "eval_runtime": 4.8879,
-      "eval_samples_per_second": 102.294,
-      "eval_steps_per_second": 6.547,
+      "eval_loss": 0.21771642565727234,
+      "eval_runtime": 4.6518,
+      "eval_samples_per_second": 107.486,
+      "eval_steps_per_second": 6.879,
       "step": 5760
     },
     {
       "epoch": 1.667129758266185,
-      "grad_norm": 0.6862344145774841,
-      "learning_rate": 0.0006889691580994721,
-      "loss": 0.2158,
+      "grad_norm": 1.1229606866836548,
+      "learning_rate": 0.0006889321107715106,
+      "loss": 0.209,
       "step": 6000
     },
     {
       "epoch": 1.8005001389274797,
-      "eval_loss": 0.21265414357185364,
-      "eval_runtime": 5.1036,
-      "eval_samples_per_second": 97.97,
-      "eval_steps_per_second": 6.27,
+      "eval_loss": 0.2162657231092453,
+      "eval_runtime": 3.9728,
+      "eval_samples_per_second": 125.857,
+      "eval_steps_per_second": 8.055,
       "step": 6480
     },
     {
       "epoch": 1.8060572381217006,
-      "grad_norm": 0.64631587266922,
-      "learning_rate": 0.0006797073261091044,
-      "loss": 0.2069,
+      "grad_norm": 0.562541127204895,
+      "learning_rate": 0.000679670278781143,
+      "loss": 0.2053,
       "step": 6500
     },
     {
       "epoch": 1.9449847179772157,
-      "grad_norm": 0.6139649748802185,
-      "learning_rate": 0.0006704454941187368,
-      "loss": 0.2063,
+      "grad_norm": 0.9058884382247925,
+      "learning_rate": 0.0006704084467907753,
+      "loss": 0.2047,
       "step": 7000
     },
     {
       "epoch": 2.000555709919422,
-      "eval_loss": 0.2054276168346405,
-      "eval_runtime": 5.6047,
-      "eval_samples_per_second": 89.211,
-      "eval_steps_per_second": 5.71,
+      "eval_loss": 0.21751119196414948,
+      "eval_runtime": 3.8866,
+      "eval_samples_per_second": 128.646,
+      "eval_steps_per_second": 8.233,
       "step": 7200
     },
     {
       "epoch": 2.0839121978327313,
-      "grad_norm": 0.5134121179580688,
-      "learning_rate": 0.000661183662128369,
-      "loss": 0.1961,
+      "grad_norm": 0.5113864541053772,
+      "learning_rate": 0.0006611466148004076,
+      "loss": 0.1972,
       "step": 7500
     },
     {
       "epoch": 2.2006112809113643,
-      "eval_loss": 0.20968343317508698,
-      "eval_runtime": 4.9427,
-      "eval_samples_per_second": 101.16,
-      "eval_steps_per_second": 6.474,
+      "eval_loss": 0.21081140637397766,
+      "eval_runtime": 3.9021,
+      "eval_samples_per_second": 128.136,
+      "eval_steps_per_second": 8.201,
       "step": 7920
     },
     {
       "epoch": 2.222839677688247,
-      "grad_norm": 0.667497456073761,
-      "learning_rate": 0.0006519218301380013,
-      "loss": 0.198,
+      "grad_norm": 0.6523454785346985,
+      "learning_rate": 0.0006518847828100398,
+      "loss": 0.1968,
       "step": 8000
     },
     {
       "epoch": 2.361767157543762,
-      "grad_norm": 0.7320435643196106,
-      "learning_rate": 0.0006426599981476336,
-      "loss": 0.1942,
+      "grad_norm": 0.48720309138298035,
+      "learning_rate": 0.0006426414744836529,
+      "loss": 0.1934,
       "step": 8500
     },
     {
       "epoch": 2.4006668519033063,
-      "eval_loss": 0.201798677444458,
-      "eval_runtime": 5.0966,
-      "eval_samples_per_second": 98.105,
-      "eval_steps_per_second": 6.279,
+      "eval_loss": 0.20452378690242767,
+      "eval_runtime": 4.661,
+      "eval_samples_per_second": 107.272,
+      "eval_steps_per_second": 6.865,
       "step": 8640
     },
     {
       "epoch": 2.5006946373992776,
-      "grad_norm": 0.5457369685173035,
-      "learning_rate": 0.000633398166157266,
-      "loss": 0.1904,
+      "grad_norm": 0.6927788257598877,
+      "learning_rate": 0.0006333796424932852,
+      "loss": 0.1906,
       "step": 9000
     },
     {
       "epoch": 2.600722422895249,
-      "eval_loss": 0.19970637559890747,
-      "eval_runtime": 4.9095,
-      "eval_samples_per_second": 101.843,
-      "eval_steps_per_second": 6.518,
+      "eval_loss": 0.1991214007139206,
+      "eval_runtime": 4.7026,
+      "eval_samples_per_second": 106.324,
+      "eval_steps_per_second": 6.805,
       "step": 9360
     },
     {
       "epoch": 2.639622117254793,
-      "grad_norm": Infinity,
-      "learning_rate": 0.0006241363341668982,
-      "loss": 0.1928,
+      "grad_norm": 0.6956183910369873,
+      "learning_rate": 0.0006241178105029175,
+      "loss": 0.1893,
       "step": 9500
     },
     {
       "epoch": 2.7785495971103087,
-      "grad_norm": 0.5789324045181274,
-      "learning_rate": 0.0006148930258405113,
-      "loss": 0.1914,
+      "grad_norm": 0.40230458974838257,
+      "learning_rate": 0.0006148559785125498,
+      "loss": 0.1889,
       "step": 10000
     },
     {
       "epoch": 2.800777993887191,
-      "eval_loss": 0.20011071860790253,
-      "eval_runtime": 4.9182,
-      "eval_samples_per_second": 101.663,
-      "eval_steps_per_second": 6.506,
+      "eval_loss": 0.20080548524856567,
+      "eval_runtime": 4.5769,
+      "eval_samples_per_second": 109.245,
+      "eval_steps_per_second": 6.992,
       "step": 10080
     },
     {
       "epoch": 2.917477076965824,
-      "grad_norm": 0.5479323863983154,
-      "learning_rate": 0.0006056311938501436,
-      "loss": 0.193,
+      "grad_norm": 1.1422420740127563,
+      "learning_rate": 0.0006055941465221821,
+      "loss": 0.1911,
       "step": 10500
     },
     {
       "epoch": 3.000833564879133,
-      "eval_loss": 0.19800344109535217,
-      "eval_runtime": 5.4208,
-      "eval_samples_per_second": 92.236,
-      "eval_steps_per_second": 5.903,
+      "eval_loss": 0.2006545066833496,
+      "eval_runtime": 4.7461,
+      "eval_samples_per_second": 105.351,
+      "eval_steps_per_second": 6.742,
       "step": 10800
     },
     {
       "epoch": 3.0564045568213394,
-      "grad_norm": 0.509381890296936,
-      "learning_rate": 0.0005963693618597759,
-      "loss": 0.1849,
+      "grad_norm": 0.8359866142272949,
+      "learning_rate": 0.0005963323145318144,
+      "loss": 0.1858,
       "step": 11000
     },
     {
       "epoch": 3.1953320366768545,
-      "grad_norm": 0.5641015768051147,
-      "learning_rate": 0.0005871075298694081,
-      "loss": 0.1803,
+      "grad_norm": 0.5189564824104309,
+      "learning_rate": 0.0005870704825414468,
+      "loss": 0.1789,
       "step": 11500
     },
     {
       "epoch": 3.2008891358710754,
-      "eval_loss": 0.19798952341079712,
-      "eval_runtime": 6.0408,
-      "eval_samples_per_second": 82.771,
-      "eval_steps_per_second": 5.297,
+      "eval_loss": 0.19731611013412476,
+      "eval_runtime": 4.0407,
+      "eval_samples_per_second": 123.742,
+      "eval_steps_per_second": 7.919,
       "step": 11520
     },
     {
       "epoch": 3.33425951653237,
-      "grad_norm": 0.5248855352401733,
-      "learning_rate": 0.0005778456978790405,
-      "loss": 0.1785,
+      "grad_norm": 0.578628659248352,
+      "learning_rate": 0.0005778271742150597,
+      "loss": 0.1799,
       "step": 12000
     },
     {
       "epoch": 3.4009447068630174,
-      "eval_loss": 0.1982451230287552,
-      "eval_runtime": 6.0297,
-      "eval_samples_per_second": 82.922,
-      "eval_steps_per_second": 5.307,
+      "eval_loss": 0.19467230141162872,
+      "eval_runtime": 4.0481,
+      "eval_samples_per_second": 123.514,
+      "eval_steps_per_second": 7.905,
       "step": 12240
     },
     {
       "epoch": 3.4731869963878856,
-      "grad_norm": 1.105663537979126,
-      "learning_rate": 0.0005685838658886728,
-      "loss": 0.1758,
+      "grad_norm": 0.8292349576950073,
+      "learning_rate": 0.0005685653422246921,
+      "loss": 0.1769,
       "step": 12500
     },
     {
       "epoch": 3.6010002778549595,
-      "eval_loss": 0.1906416267156601,
-      "eval_runtime": 5.9927,
-      "eval_samples_per_second": 83.434,
-      "eval_steps_per_second": 5.34,
+      "eval_loss": 0.19331130385398865,
+      "eval_runtime": 4.0814,
+      "eval_samples_per_second": 122.508,
+      "eval_steps_per_second": 7.841,
       "step": 12960
     },
     {
       "epoch": 3.612114476243401,
-      "grad_norm": 1.019622802734375,
-      "learning_rate": 0.0005593405575622859,
-      "loss": 0.1765,
+      "grad_norm": 0.6857467293739319,
+      "learning_rate": 0.0005593035102343244,
+      "loss": 0.1796,
       "step": 13000
     },
     {
       "epoch": 3.7510419560989163,
-      "grad_norm": 0.8380174040794373,
-      "learning_rate": 0.0005500787255719182,
-      "loss": 0.1756,
+      "grad_norm": 0.5829126238822937,
+      "learning_rate": 0.0005500416782439567,
+      "loss": 0.176,
       "step": 13500
     },
     {
       "epoch": 3.801055848846902,
-      "eval_loss": 0.18713021278381348,
-      "eval_runtime": 4.8131,
-      "eval_samples_per_second": 103.884,
-      "eval_steps_per_second": 6.649,
+      "eval_loss": 0.18971021473407745,
+      "eval_runtime": 4.5985,
+      "eval_samples_per_second": 108.73,
+      "eval_steps_per_second": 6.959,
       "step": 13680
     },
     {
       "epoch": 3.889969435954432,
-      "grad_norm": 0.858310878276825,
-      "learning_rate": 0.0005408168935815504,
-      "loss": 0.1773,
+      "grad_norm": 0.7102944254875183,
+      "learning_rate": 0.0005407798462535889,
+      "loss": 0.1753,
       "step": 14000
     }
   ],

checkpoint-14396/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43d31253d1d9828754b954b23155d823529e2ebc7f148e5c229e9bc7b696124d
-size 6033
+oid sha256:1ac71e3cdd10165b76004d5b1c4b8cd4a183b94ec338868fce8dc9960bf7c890
+size 5496