Training in progress, step 156, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f881f7af17e8e692fe8e56b0bb3efdf3d1b82897cd46e96e09ffe05b97b0277
+oid sha256:075fad7910ad5434f099ab9697f7f34d98149f8051c6b928bc1fcda4d81a34cf
 size 550593856

last-checkpoint/global_step156/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:548a32937f39dcb1efd25459a1c324dc13150384367f773d2dd8448c92667292
+size 243590464

last-checkpoint/global_step156/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:933a00a04cf11d99eeabc52feb9b90fe01dd428bf5e1c284aea7a2d9d4b5d75c
+size 243590464

last-checkpoint/global_step156/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de149628d16885054f37021368e6dc7250b24a86d676fca4c7b73c052bc2a4ee
+size 243590464

last-checkpoint/global_step156/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16c2a1903e9427ca8cfc0c202e8aa9f0e5165e722912c3d6957fd3379ca85ba8
+size 243590464

last-checkpoint/global_step156/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3da911f1de45691c9ecb869622806d3dde56d75e3d47818f3033435c1bd25e28
+size 243590464

last-checkpoint/global_step156/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff36ec5a25eee06afd325568cd992f9a94c2c521b70320d9e272376429ac2663
+size 243590464

last-checkpoint/global_step156/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7c9560be272f0a903f912fef8dfcbbd8ffb32a2467848964432c04fd6982526
+size 243590464

last-checkpoint/global_step156/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ef76f92739b8fdf88dd6d6d990885b8e2c7713385d5a8b4f4be3b03f98a6178
+size 243590464

last-checkpoint/global_step156/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:826faad08e897171bea38c0247ede4f75e080d152d7b926bcc2aae1f38dc241d
+size 211435686

last-checkpoint/global_step156/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf376209455a315841c1ccc3f778555f40052195e31400dd3a4308d98bfdb581
+size 211435686

last-checkpoint/global_step156/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c34502a6ce26d4b43c5bdf522735779070cde32a1c44a278e866c690ba32dc24
+size 211435686

last-checkpoint/global_step156/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f178c64dcb2c4a64925325012066e8e57e05dad95b5c1a9c5c4b3a3de678c7c3
+size 211435686

last-checkpoint/global_step156/zero_pp_rank_4_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e898235613f32df36167407e9fcdd27b2e954facaa00f5b5374a5ef019d0126
+size 211435686

last-checkpoint/global_step156/zero_pp_rank_5_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:605de77a7d38dc8ad8ad07b662756b8d65a0c49c8d2071b9519b8375c19700f6
+size 211435686

last-checkpoint/global_step156/zero_pp_rank_6_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45bdcb91d93dc78080afd1d40dacac300abdc5046169b948fd228f2d50d8668b
+size 211435686

last-checkpoint/global_step156/zero_pp_rank_7_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01a71440de4b435aa80d6384b5767ae57296096e58d5939720ff88e2dbf3e97f
+size 211435686

last-checkpoint/latest

@@ -1 +1 @@
-global_step117
+global_step156

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aac43c4e4aa943b1080df7e7983872659b6a36d1bde144f8865b626eeee7f434
+oid sha256:71b89abcc0ce5c407fb9d4a92dbfe3b45e6c73df2570fd239cc5b08cb8a1fa73
 size 15920

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f7c947232f2bdf46ca9ace1c9ebaf1e33912f79599e506f7f7c8ff48a11c7b15
+oid sha256:7fed8f9051879059e4d4a6c0268a1c07bd3d17a63ff9695dbecba7f759b48265
 size 15920

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8cfa2ab32c5e9f8e4c49775769b133f4238f016c4e48e8b12b51160d4afbf783
+oid sha256:f5cbe3d657365bc422108101740b07139f39ec40e3d2fa73b70f4ab763f46ca0
 size 15920

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c73c8b459dde07f85f2cffa5de5fb0bf136e4a80afc50d2475bc570d14a65b2e
+oid sha256:dafe72f7606d70daf9651f02eb33505031fd74c64c42caa3b416885081a3a279
 size 15920

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9533672e6284f7cd4c7c4acb5ad621a319e75dbe8a3e90685943824a71eea36
+oid sha256:3c62d508b5b72fe3ae01dd0978107829aac79e266e3887fdb5b71e9ab552a058
 size 15920

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a9be2d2ec0fbf037c14fb0f5f0467f7c960e90414356feec416f8eeb1684671
+oid sha256:2af92289a5626dcb7fa831a68cf4de5926abfbbf164dcbd6c3968dc0685c46d8
 size 15920

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c6a4539a42d09819ea8a238c4dc3e947e079d2a1390a7babff9377b3718e2a09
+oid sha256:3d97e7ac49adbd28762ed99d933cc4505e066155bf934c98c26916b8462bde41
 size 15920

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e097a64585a03993feb0a88a81b38258f3325c6b6c17bdedb7335173ed053ac4
+oid sha256:3ca057b2b9856c6b9c98bc92f3b0e1d16858df77c847fd5298f787624eabf383
 size 15920

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:924db454eec3725afade5f0aae6c3ada154b140ee15da5b50e9bc51987b43065
+oid sha256:aa19ded294b26c77ff9357b30aeec2441cdd5bbf5a173650ff5ac1680ce0a913
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.3046875,
+  "epoch": 0.40625,
   "eval_steps": 39,
-  "global_step": 117,
+  "global_step": 156,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -858,6 +858,287 @@
       "eval_samples_per_second": 1.223,
       "eval_steps_per_second": 0.153,
       "step": 117
+    },
+    {
+      "epoch": 0.3072916666666667,
+      "grad_norm": 0.14857253164474957,
+      "learning_rate": 8.484051962083579e-05,
+      "loss": 2.3431,
+      "step": 118
+    },
+    {
+      "epoch": 0.3098958333333333,
+      "grad_norm": 0.14191988741229736,
+      "learning_rate": 8.454870245003141e-05,
+      "loss": 2.3861,
+      "step": 119
+    },
+    {
+      "epoch": 0.3125,
+      "grad_norm": 0.1516024307266096,
+      "learning_rate": 8.425468421509349e-05,
+      "loss": 2.411,
+      "step": 120
+    },
+    {
+      "epoch": 0.3151041666666667,
+      "grad_norm": 0.15002982823803926,
+      "learning_rate": 8.395848681725416e-05,
+      "loss": 2.49,
+      "step": 121
+    },
+    {
+      "epoch": 0.3177083333333333,
+      "grad_norm": 0.14904411125984457,
+      "learning_rate": 8.366013232007002e-05,
+      "loss": 2.2998,
+      "step": 122
+    },
+    {
+      "epoch": 0.3203125,
+      "grad_norm": 0.15513959182528284,
+      "learning_rate": 8.335964294777862e-05,
+      "loss": 2.4748,
+      "step": 123
+    },
+    {
+      "epoch": 0.3229166666666667,
+      "grad_norm": 0.16028597088352084,
+      "learning_rate": 8.305704108364301e-05,
+      "loss": 2.5415,
+      "step": 124
+    },
+    {
+      "epoch": 0.3255208333333333,
+      "grad_norm": 0.1382845005964039,
+      "learning_rate": 8.275234926828446e-05,
+      "loss": 2.3863,
+      "step": 125
+    },
+    {
+      "epoch": 0.328125,
+      "grad_norm": 0.13146327268638525,
+      "learning_rate": 8.244559019800328e-05,
+      "loss": 2.2718,
+      "step": 126
+    },
+    {
+      "epoch": 0.3307291666666667,
+      "grad_norm": 0.14918540441469405,
+      "learning_rate": 8.213678672308841e-05,
+      "loss": 2.29,
+      "step": 127
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.14572964949235084,
+      "learning_rate": 8.182596184611514e-05,
+      "loss": 2.3865,
+      "step": 128
+    },
+    {
+      "epoch": 0.3359375,
+      "grad_norm": 0.1467709610422986,
+      "learning_rate": 8.151313872023172e-05,
+      "loss": 2.3566,
+      "step": 129
+    },
+    {
+      "epoch": 0.3385416666666667,
+      "grad_norm": 0.1696428030171741,
+      "learning_rate": 8.119834064743469e-05,
+      "loss": 2.4145,
+      "step": 130
+    },
+    {
+      "epoch": 0.3411458333333333,
+      "grad_norm": 0.1567212274267596,
+      "learning_rate": 8.088159107683314e-05,
+      "loss": 2.3996,
+      "step": 131
+    },
+    {
+      "epoch": 0.34375,
+      "grad_norm": 0.16059171688564705,
+      "learning_rate": 8.056291360290201e-05,
+      "loss": 2.4796,
+      "step": 132
+    },
+    {
+      "epoch": 0.3463541666666667,
+      "grad_norm": 0.17528470844982455,
+      "learning_rate": 8.024233196372453e-05,
+      "loss": 2.3711,
+      "step": 133
+    },
+    {
+      "epoch": 0.3489583333333333,
+      "grad_norm": 0.15733889602776618,
+      "learning_rate": 7.9919870039224e-05,
+      "loss": 2.2677,
+      "step": 134
+    },
+    {
+      "epoch": 0.3515625,
+      "grad_norm": 0.14873870088112393,
+      "learning_rate": 7.959555184938495e-05,
+      "loss": 2.4515,
+      "step": 135
+    },
+    {
+      "epoch": 0.3541666666666667,
+      "grad_norm": 0.15417965735327782,
+      "learning_rate": 7.926940155246397e-05,
+      "loss": 2.4285,
+      "step": 136
+    },
+    {
+      "epoch": 0.3567708333333333,
+      "grad_norm": 0.1431617281765595,
+      "learning_rate": 7.894144344319014e-05,
+      "loss": 2.2096,
+      "step": 137
+    },
+    {
+      "epoch": 0.359375,
+      "grad_norm": 0.1582812354896834,
+      "learning_rate": 7.861170195095537e-05,
+      "loss": 2.5397,
+      "step": 138
+    },
+    {
+      "epoch": 0.3619791666666667,
+      "grad_norm": 0.17372786439002758,
+      "learning_rate": 7.828020163799455e-05,
+      "loss": 2.4293,
+      "step": 139
+    },
+    {
+      "epoch": 0.3645833333333333,
+      "grad_norm": 0.12985922414383083,
+      "learning_rate": 7.794696719755612e-05,
+      "loss": 2.3064,
+      "step": 140
+    },
+    {
+      "epoch": 0.3671875,
+      "grad_norm": 0.14444453543114258,
+      "learning_rate": 7.761202345206249e-05,
+      "loss": 2.4924,
+      "step": 141
+    },
+    {
+      "epoch": 0.3697916666666667,
+      "grad_norm": 0.14361378320872797,
+      "learning_rate": 7.727539535126118e-05,
+      "loss": 2.485,
+      "step": 142
+    },
+    {
+      "epoch": 0.3723958333333333,
+      "grad_norm": 0.14506765397834698,
+      "learning_rate": 7.69371079703662e-05,
+      "loss": 2.4081,
+      "step": 143
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 0.15082242594669004,
+      "learning_rate": 7.65971865081904e-05,
+      "loss": 2.4042,
+      "step": 144
+    },
+    {
+      "epoch": 0.3776041666666667,
+      "grad_norm": 0.14811116555183387,
+      "learning_rate": 7.625565628526818e-05,
+      "loss": 2.5335,
+      "step": 145
+    },
+    {
+      "epoch": 0.3802083333333333,
+      "grad_norm": 0.14726265807121236,
+      "learning_rate": 7.591254274196959e-05,
+      "loss": 2.2424,
+      "step": 146
+    },
+    {
+      "epoch": 0.3828125,
+      "grad_norm": 0.1582839530592079,
+      "learning_rate": 7.556787143660521e-05,
+      "loss": 2.3499,
+      "step": 147
+    },
+    {
+      "epoch": 0.3854166666666667,
+      "grad_norm": 0.15670436913091046,
+      "learning_rate": 7.522166804352226e-05,
+      "loss": 2.4295,
+      "step": 148
+    },
+    {
+      "epoch": 0.3880208333333333,
+      "grad_norm": 0.14730612216140554,
+      "learning_rate": 7.487395835119231e-05,
+      "loss": 2.3101,
+      "step": 149
+    },
+    {
+      "epoch": 0.390625,
+      "grad_norm": 0.15120580103738476,
+      "learning_rate": 7.452476826029011e-05,
+      "loss": 2.4888,
+      "step": 150
+    },
+    {
+      "epoch": 0.3932291666666667,
+      "grad_norm": 0.15081791485599214,
+      "learning_rate": 7.417412378176446e-05,
+      "loss": 2.3946,
+      "step": 151
+    },
+    {
+      "epoch": 0.3958333333333333,
+      "grad_norm": 0.13244742107918075,
+      "learning_rate": 7.382205103490043e-05,
+      "loss": 2.2704,
+      "step": 152
+    },
+    {
+      "epoch": 0.3984375,
+      "grad_norm": 0.14571632653572605,
+      "learning_rate": 7.346857624537407e-05,
+      "loss": 2.4644,
+      "step": 153
+    },
+    {
+      "epoch": 0.4010416666666667,
+      "grad_norm": 0.15252862923031207,
+      "learning_rate": 7.311372574329854e-05,
+      "loss": 2.554,
+      "step": 154
+    },
+    {
+      "epoch": 0.4036458333333333,
+      "grad_norm": 0.14927980854491554,
+      "learning_rate": 7.275752596126308e-05,
+      "loss": 2.3804,
+      "step": 155
+    },
+    {
+      "epoch": 0.40625,
+      "grad_norm": 0.1406590744757482,
+      "learning_rate": 7.240000343236385e-05,
+      "loss": 2.2504,
+      "step": 156
+    },
+    {
+      "epoch": 0.40625,
+      "eval_loss": 2.408146619796753,
+      "eval_runtime": 65.4856,
+      "eval_samples_per_second": 1.222,
+      "eval_steps_per_second": 0.153,
+      "step": 156
     }
   ],
   "logging_steps": 1,
@@ -877,7 +1158,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 38677656895488.0,
+  "total_flos": 51570209193984.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null