ckpt checkpoint-376 (base_pilot_nometa_sft)

checkpoint-376/global_step376/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae54f19b86b32db4905ba22cc3f9c4659b71edbe4ac7cf77ace971b34d74657f
+oid sha256:56c30ec4f4d4acca8672fecfd266e5ed1afa257e5cc35d062bfeb47a49e5a018
 size 24565700179

checkpoint-376/global_step376/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f24479a0964caa93b97427d0ec920817ed1f4a789e2217689562c9034ee8c681
+oid sha256:e85996c2ef17729eeed2b0eaf4b8f69bf2d1ed4e3e79474674dc6c6b03edfb75
 size 24565700179

checkpoint-376/global_step376/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7fa4356e497f7df5ad356cd21a30f1d896f1eda38ccf9f4b9f0b57847512e01d
+oid sha256:7ec11d852e38e4817025c99fed8ea845cdca32aa1b1177e7fab8d6ceb96af9ec
 size 24565700179

checkpoint-376/global_step376/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a62ee1b636a58d539e9800a91ab261b5ef1e735ccc653371b48e7771a5525ef
+oid sha256:5f76c1dcc5f2c2692d205ac1c6b6d2cfa37abedda03b5e6d7d039842f70d9cb4
 size 24565700179

checkpoint-376/model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72384e97fa226024d3b54ea96b36ff1cdf5082a444a3493bf5518bc8f8998402
+oid sha256:0c6deb9056c8380da7f2c15655bdf635ac459e937a250a4a4cd949e6939c419b
 size 4900086816

checkpoint-376/model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55591546087093e24e5b27142b18a1b8f66b86a2f92855f6e18d2b642dee72c4
+oid sha256:5e302ab860cd735acfd5b05f44e3c2e97b4fbfdf44599c83a09a0aed203f8c59
 size 4915960368

checkpoint-376/model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a97258be80b251ce4c0bf98582fd56736d4e69606466dd78897951f604546a9
+oid sha256:d48a8a14cf1024ff6d523a49eae0112ec70ad4780a3a1e4a9fbe909cdfdd2420
 size 4983068496

checkpoint-376/model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:34292e54d4faf03043670ddf229da015d3a307d37a17b5511e28d950e4710ec9
+oid sha256:f0f3f275213a7d5c1e83e4b8e0201426ebb9d58cce75c27e241cc5722cf09592
 size 1578059384

checkpoint-376/trainer_state.json

@@ -11,261 +11,261 @@
   "log_history": [
     {
       "epoch": 0.026595744680851064,
-      "grad_norm": 7.6010222657865505,
+      "grad_norm": 7.330494644548605,
       "learning_rate": 4.7368421052631574e-07,
-      "loss": 1.4162,
+      "loss": 1.0029,
       "step": 10
     },
     {
       "epoch": 0.05319148936170213,
-      "grad_norm": 4.139263320336168,
+      "grad_norm": 4.589645623628691,
       "learning_rate": 1e-06,
-      "loss": 1.3893,
+      "loss": 0.9945,
       "step": 20
     },
     {
       "epoch": 0.0797872340425532,
-      "grad_norm": 5.442129187478661,
+      "grad_norm": 4.777196563499304,
       "learning_rate": 1.5263157894736842e-06,
-      "loss": 1.3626,
+      "loss": 0.8627,
       "step": 30
     },
     {
       "epoch": 0.10638297872340426,
-      "grad_norm": 4.029110964909961,
+      "grad_norm": 3.622116815775469,
       "learning_rate": 1.999956805025437e-06,
-      "loss": 1.194,
+      "loss": 0.8094,
       "step": 40
     },
     {
       "epoch": 0.13297872340425532,
-      "grad_norm": 5.3017185408683725,
+      "grad_norm": 4.702145513968783,
       "learning_rate": 1.994777921806989e-06,
-      "loss": 1.0244,
+      "loss": 0.6803,
       "step": 50
     },
     {
       "epoch": 0.1595744680851064,
-      "grad_norm": 2.9169682766028417,
+      "grad_norm": 3.3150741263240806,
       "learning_rate": 1.9810112804959865e-06,
-      "loss": 0.8202,
+      "loss": 0.5826,
       "step": 60
     },
     {
       "epoch": 0.18617021276595744,
-      "grad_norm": 3.887506889882244,
+      "grad_norm": 4.25655201170495,
       "learning_rate": 1.9587757262967055e-06,
-      "loss": 0.7944,
+      "loss": 0.6075,
       "step": 70
     },
     {
       "epoch": 0.2127659574468085,
-      "grad_norm": 3.255363906878193,
+      "grad_norm": 4.289723518654417,
       "learning_rate": 1.9282632151787463e-06,
-      "loss": 0.7347,
+      "loss": 0.6156,
       "step": 80
     },
     {
       "epoch": 0.2393617021276596,
-      "grad_norm": 3.0007717020527704,
+      "grad_norm": 3.343959574597276,
       "learning_rate": 1.889737156751965e-06,
-      "loss": 0.7116,
+      "loss": 0.6416,
       "step": 90
     },
     {
       "epoch": 0.26595744680851063,
-      "grad_norm": 3.5944235473139727,
+      "grad_norm": 3.8097499500081664,
       "learning_rate": 1.843530140293603e-06,
-      "loss": 0.6396,
+      "loss": 0.5418,
       "step": 100
     },
     {
       "epoch": 0.2925531914893617,
-      "grad_norm": 2.976134151030474,
+      "grad_norm": 3.7316543679274674,
       "learning_rate": 1.7900410635584497e-06,
-      "loss": 0.6337,
+      "loss": 0.5479,
       "step": 110
     },
     {
       "epoch": 0.3191489361702128,
-      "grad_norm": 2.9656478687330203,
+      "grad_norm": 4.777510893759255,
       "learning_rate": 1.7297316891585675e-06,
-      "loss": 0.6472,
+      "loss": 0.5815,
       "step": 120
     },
     {
       "epoch": 0.34574468085106386,
-      "grad_norm": 2.200746754139919,
+      "grad_norm": 2.5742962688465525,
       "learning_rate": 1.6631226582407952e-06,
-      "loss": 0.5883,
+      "loss": 0.5131,
       "step": 130
     },
     {
       "epoch": 0.3723404255319149,
-      "grad_norm": 2.2326965867407993,
+      "grad_norm": 2.9390390289937462,
       "learning_rate": 1.5907889958753131e-06,
-      "loss": 0.5891,
+      "loss": 0.5089,
       "step": 140
     },
     {
       "epoch": 0.39893617021276595,
-      "grad_norm": 2.9059518651351044,
+      "grad_norm": 3.4276391098454533,
       "learning_rate": 1.5133551469565311e-06,
-      "loss": 0.6321,
+      "loss": 0.5515,
       "step": 150
     },
     {
       "epoch": 0.425531914893617,
-      "grad_norm": 2.532254194193504,
+      "grad_norm": 2.861723720021689,
       "learning_rate": 1.431489585470564e-06,
-      "loss": 0.5925,
+      "loss": 0.4847,
       "step": 160
     },
     {
       "epoch": 0.4521276595744681,
-      "grad_norm": 1.8350243322634734,
+      "grad_norm": 2.411450013188873,
       "learning_rate": 1.3458990436666312e-06,
-      "loss": 0.6705,
+      "loss": 0.6332,
       "step": 170
     },
     {
       "epoch": 0.4787234042553192,
-      "grad_norm": 3.0786294557126457,
+      "grad_norm": 3.508782658191763,
       "learning_rate": 1.2573224109510109e-06,
-      "loss": 0.5636,
+      "loss": 0.4837,
       "step": 180
     },
     {
       "epoch": 0.5053191489361702,
-      "grad_norm": 2.652406257824746,
+      "grad_norm": 2.850639066073856,
       "learning_rate": 1.1665243551734219e-06,
-      "loss": 0.5365,
+      "loss": 0.4551,
       "step": 190
     },
     {
       "epoch": 0.5319148936170213,
-      "grad_norm": 3.5966072765074117,
+      "grad_norm": 4.853039782410914,
       "learning_rate": 1.0742887213722372e-06,
-      "loss": 0.5462,
+      "loss": 0.4612,
       "step": 200
     },
     {
       "epoch": 0.5585106382978723,
-      "grad_norm": 2.9565756077686127,
+      "grad_norm": 3.217718418518325,
       "learning_rate": 9.81411764966112e-07,
-      "loss": 0.6126,
+      "loss": 0.5484,
       "step": 210
     },
     {
       "epoch": 0.5851063829787234,
-      "grad_norm": 2.3656522512638687,
+      "grad_norm": 2.8907729359597347,
       "learning_rate": 8.886952778087909e-07,
-      "loss": 0.5133,
+      "loss": 0.4264,
       "step": 220
     },
     {
       "epoch": 0.6117021276595744,
-      "grad_norm": 2.1988671277513983,
+      "grad_norm": 3.2095424449531156,
       "learning_rate": 7.969396664487533e-07,
-      "loss": 0.5517,
+      "loss": 0.4699,
       "step": 230
     },
     {
       "epoch": 0.6382978723404256,
-      "grad_norm": 2.295342808209699,
+      "grad_norm": 3.4175232947819536,
       "learning_rate": 7.069370423479869e-07,
-      "loss": 0.5678,
+      "loss": 0.4994,
       "step": 240
     },
     {
       "epoch": 0.6648936170212766,
-      "grad_norm": 3.1209844389100647,
+      "grad_norm": 3.4355980304379496,
       "learning_rate": 6.194643837109014e-07,
-      "loss": 0.5932,
+      "loss": 0.4898,
       "step": 250
     },
     {
       "epoch": 0.6914893617021277,
-      "grad_norm": 2.390560790393853,
+      "grad_norm": 3.272456667918856,
       "learning_rate": 5.352768279562314e-07,
-      "loss": 0.479,
+      "loss": 0.4025,
       "step": 260
     },
     {
       "epoch": 0.7180851063829787,
-      "grad_norm": 2.5929836635045698,
+      "grad_norm": 2.9174558359705083,
       "learning_rate": 4.5510115273693475e-07,
-      "loss": 0.5222,
+      "loss": 0.4427,
       "step": 270
     },
     {
       "epoch": 0.7446808510638298,
-      "grad_norm": 2.472545460893312,
+      "grad_norm": 2.911990774521814,
       "learning_rate": 3.796295017853928e-07,
-      "loss": 0.5291,
+      "loss": 0.4657,
       "step": 280
     },
     {
       "epoch": 0.7712765957446809,
-      "grad_norm": 3.396077889684821,
+      "grad_norm": 4.011942260659963,
       "learning_rate": 3.0951340974769033e-07,
-      "loss": 0.522,
+      "loss": 0.4296,
       "step": 290
     },
     {
       "epoch": 0.7978723404255319,
-      "grad_norm": 5.596285164590074,
+      "grad_norm": 2.4209418515895,
       "learning_rate": 2.4535817758961643e-07,
-      "loss": 0.559,
+      "loss": 0.4792,
       "step": 300
     },
     {
       "epoch": 0.824468085106383,
-      "grad_norm": 2.875148228447102,
+      "grad_norm": 2.9059308358072706,
       "learning_rate": 1.8771764713060356e-07,
-      "loss": 0.5465,
+      "loss": 0.4676,
       "step": 310
     },
     {
       "epoch": 0.851063829787234,
-      "grad_norm": 2.676944211948832,
+      "grad_norm": 3.536456067389615,
       "learning_rate": 1.3708941981621813e-07,
-      "loss": 0.5321,
+      "loss": 0.4544,
       "step": 320
     },
     {
       "epoch": 0.8776595744680851,
-      "grad_norm": 2.6773448738870935,
+      "grad_norm": 3.745026666881224,
       "learning_rate": 9.391056100476735e-08,
-      "loss": 0.5303,
+      "loss": 0.4928,
       "step": 330
     },
     {
       "epoch": 0.9042553191489362,
-      "grad_norm": 2.281962816726945,
+      "grad_norm": 2.6728901514289394,
       "learning_rate": 5.855382685223009e-08,
-      "loss": 0.5314,
+      "loss": 0.4729,
       "step": 340
     },
     {
       "epoch": 0.9308510638297872,
-      "grad_norm": 2.731706908879832,
+      "grad_norm": 3.6834641292564916,
       "learning_rate": 3.132444636820575e-08,
-      "loss": 0.56,
+      "loss": 0.4815,
       "step": 350
     },
     {
       "epoch": 0.9574468085106383,
-      "grad_norm": 2.274431531910594,
+      "grad_norm": 3.1783192399032814,
       "learning_rate": 1.2457486422878138e-08,
-      "loss": 0.4837,
+      "loss": 0.419,
       "step": 360
     },
     {
       "epoch": 0.9840425531914894,
-      "grad_norm": 3.299439785574286,
+      "grad_norm": 4.067207854555839,
       "learning_rate": 2.115822452463223e-09,
-      "loss": 0.5721,
+      "loss": 0.5107,
       "step": 370
     }
   ],
@@ -286,7 +286,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5653755265024.0,
+  "total_flos": 3922571436032.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null

checkpoint-376/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71de3f3391491b3a823751366118d4158fd29fe1233536c7afe35108153fab07
-size 7057
+oid sha256:6a0e37f82719c07a27dc65b3a04bece7517301cce05685a623492bca132c559d
+size 7121