Fine-tuned model update

checkpoint-1605/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a78c45c29ec4d43b9c9d84054394a0b682489b9ba02524e706f2df280b697526
+oid sha256:0b892a1da419ea5e2308ca459691451091eb08ff29ec453dc60c71d90f9cda75
 size 1112223464

checkpoint-1605/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e5fb5006a7d66c0f5968f9dd26f810ece329a7fd2aca9b05f1dd33648e843d0
+oid sha256:f4f818316d94db9569a51254ccd85ca7b72b7a420be492312d80f30e9983bdda
 size 2224566458

checkpoint-1605/trainer_state.json

@@ -10,250 +10,250 @@
   "log_history": [
     {
       "epoch": 0.09345794392523364,
-      "grad_norm": 0.008999723941087723,
+      "grad_norm": 0.012311354279518127,
       "learning_rate": 4.844236760124611e-05,
-      "loss": 0.243,
+      "loss": 0.252,
       "step": 50
     },
     {
       "epoch": 0.18691588785046728,
-      "grad_norm": 0.009431353770196438,
+      "grad_norm": 0.11957651376724243,
       "learning_rate": 4.6884735202492216e-05,
-      "loss": 0.0069,
+      "loss": 0.0064,
       "step": 100
     },
     {
       "epoch": 0.2803738317757009,
-      "grad_norm": 0.00998706091195345,
+      "grad_norm": 0.0069496892392635345,
       "learning_rate": 4.532710280373832e-05,
-      "loss": 0.0004,
+      "loss": 0.0007,
       "step": 150
     },
     {
       "epoch": 0.37383177570093457,
-      "grad_norm": 0.007257778197526932,
+      "grad_norm": 0.0055118128657341,
       "learning_rate": 4.376947040498442e-05,
       "loss": 0.0035,
       "step": 200
     },
     {
       "epoch": 0.4672897196261682,
-      "grad_norm": 0.020342009142041206,
+      "grad_norm": 0.007157029118388891,
       "learning_rate": 4.221183800623053e-05,
-      "loss": 0.0067,
+      "loss": 0.0066,
       "step": 250
     },
     {
       "epoch": 0.5607476635514018,
-      "grad_norm": 0.004194826819002628,
+      "grad_norm": 0.005454342812299728,
       "learning_rate": 4.0654205607476636e-05,
-      "loss": 0.0004,
+      "loss": 0.0003,
       "step": 300
     },
     {
       "epoch": 0.6542056074766355,
-      "grad_norm": 0.0033228793181478977,
+      "grad_norm": 0.0042994399555027485,
       "learning_rate": 3.909657320872274e-05,
       "loss": 0.0002,
       "step": 350
     },
     {
       "epoch": 0.7476635514018691,
-      "grad_norm": 0.002507598605006933,
+      "grad_norm": 0.0036757723428308964,
       "learning_rate": 3.753894080996885e-05,
       "loss": 0.0001,
       "step": 400
     },
     {
       "epoch": 0.8411214953271028,
-      "grad_norm": 0.004089244641363621,
+      "grad_norm": 0.004236909560859203,
       "learning_rate": 3.5981308411214956e-05,
-      "loss": 0.0036,
+      "loss": 0.0034,
       "step": 450
     },
     {
       "epoch": 0.9345794392523364,
-      "grad_norm": 0.004642685875296593,
+      "grad_norm": 0.004730775952339172,
       "learning_rate": 3.442367601246106e-05,
-      "loss": 0.0036,
+      "loss": 0.0035,
       "step": 500
     },
     {
       "epoch": 1.0,
-      "eval_loss": 8.867182731628418,
-      "eval_runtime": 0.7286,
-      "eval_samples_per_second": 6.862,
-      "eval_steps_per_second": 1.372,
+      "eval_loss": 8.711264610290527,
+      "eval_runtime": 0.6907,
+      "eval_samples_per_second": 7.239,
+      "eval_steps_per_second": 1.448,
       "step": 535
     },
     {
       "epoch": 1.02803738317757,
-      "grad_norm": 0.026588544249534607,
+      "grad_norm": 0.022852910682559013,
       "learning_rate": 3.286604361370716e-05,
-      "loss": 0.0031,
+      "loss": 0.003,
       "step": 550
     },
     {
       "epoch": 1.1214953271028036,
-      "grad_norm": 0.0032198007684201,
+      "grad_norm": 0.0033658354077488184,
       "learning_rate": 3.130841121495327e-05,
-      "loss": 0.0002,
+      "loss": 0.0003,
       "step": 600
     },
     {
       "epoch": 1.2149532710280373,
-      "grad_norm": 0.0027408436872065067,
+      "grad_norm": 0.003417043248191476,
       "learning_rate": 2.9750778816199376e-05,
       "loss": 0.0001,
       "step": 650
     },
     {
       "epoch": 1.308411214953271,
-      "grad_norm": 0.001963774673640728,
+      "grad_norm": 0.002329085720703006,
       "learning_rate": 2.8193146417445482e-05,
       "loss": 0.0001,
       "step": 700
     },
     {
       "epoch": 1.4018691588785046,
-      "grad_norm": 0.0030643907375633717,
+      "grad_norm": 0.003157567698508501,
       "learning_rate": 2.663551401869159e-05,
       "loss": 0.0036,
       "step": 750
     },
     {
       "epoch": 1.4953271028037383,
-      "grad_norm": 0.0027519240975379944,
+      "grad_norm": 0.002345046494156122,
       "learning_rate": 2.5077881619937692e-05,
-      "loss": 0.001,
+      "loss": 0.002,
       "step": 800
     },
     {
       "epoch": 1.588785046728972,
-      "grad_norm": 0.002502395771443844,
+      "grad_norm": 0.0028426579665392637,
       "learning_rate": 2.3520249221183802e-05,
       "loss": 0.0001,
       "step": 850
     },
     {
       "epoch": 1.6822429906542056,
-      "grad_norm": 0.0024411738850176334,
+      "grad_norm": 0.0021719676442444324,
       "learning_rate": 2.196261682242991e-05,
-      "loss": 0.0008,
+      "loss": 0.001,
       "step": 900
     },
     {
       "epoch": 1.7757009345794392,
-      "grad_norm": 0.003065146040171385,
+      "grad_norm": 0.0028401941526681185,
       "learning_rate": 2.0404984423676012e-05,
-      "loss": 0.0034,
+      "loss": 0.0021,
       "step": 950
     },
     {
       "epoch": 1.8691588785046729,
-      "grad_norm": 0.0014599463902413845,
+      "grad_norm": 0.0017638171557337046,
       "learning_rate": 1.884735202492212e-05,
       "loss": 0.0001,
       "step": 1000
     },
     {
       "epoch": 1.9626168224299065,
-      "grad_norm": 0.0023229168727993965,
+      "grad_norm": 0.0020422961097210646,
       "learning_rate": 1.7289719626168225e-05,
-      "loss": 0.0038,
+      "loss": 0.001,
       "step": 1050
     },
     {
       "epoch": 2.0,
-      "eval_loss": 7.628686428070068,
-      "eval_runtime": 0.3776,
-      "eval_samples_per_second": 13.241,
-      "eval_steps_per_second": 2.648,
+      "eval_loss": 5.994241237640381,
+      "eval_runtime": 0.3635,
+      "eval_samples_per_second": 13.756,
+      "eval_steps_per_second": 2.751,
       "step": 1070
     },
     {
       "epoch": 2.05607476635514,
-      "grad_norm": 0.0019721754360944033,
+      "grad_norm": 0.0015974253183230758,
       "learning_rate": 1.573208722741433e-05,
-      "loss": 0.0001,
+      "loss": 0.0004,
       "step": 1100
     },
     {
       "epoch": 2.149532710280374,
-      "grad_norm": 0.002189771970734,
+      "grad_norm": 0.0014293135609477758,
       "learning_rate": 1.4174454828660435e-05,
-      "loss": 0.0034,
+      "loss": 0.0008,
       "step": 1150
     },
     {
       "epoch": 2.2429906542056073,
-      "grad_norm": 0.0022546553518623114,
+      "grad_norm": 0.0013810868840664625,
       "learning_rate": 1.2616822429906542e-05,
       "loss": 0.0001,
       "step": 1200
     },
     {
       "epoch": 2.336448598130841,
-      "grad_norm": 0.0020222938619554043,
+      "grad_norm": 0.0013669389300048351,
       "learning_rate": 1.1059190031152649e-05,
       "loss": 0.0007,
       "step": 1250
     },
     {
       "epoch": 2.4299065420560746,
-      "grad_norm": 0.0017899508820846677,
+      "grad_norm": 0.0012841916177421808,
       "learning_rate": 9.501557632398754e-06,
       "loss": 0.0001,
       "step": 1300
     },
     {
       "epoch": 2.5233644859813085,
-      "grad_norm": 0.00168142793700099,
+      "grad_norm": 0.0013251251075416803,
       "learning_rate": 7.94392523364486e-06,
       "loss": 0.0001,
       "step": 1350
     },
     {
       "epoch": 2.616822429906542,
-      "grad_norm": 0.0016604465199634433,
+      "grad_norm": 0.001289376406930387,
       "learning_rate": 6.386292834890965e-06,
       "loss": 0.0008,
       "step": 1400
     },
     {
       "epoch": 2.710280373831776,
-      "grad_norm": 0.0016119988868013024,
+      "grad_norm": 0.0012487081112340093,
       "learning_rate": 4.828660436137072e-06,
-      "loss": 0.0001,
+      "loss": 0.0,
       "step": 1450
     },
     {
       "epoch": 2.803738317757009,
-      "grad_norm": 0.005219247657805681,
+      "grad_norm": 0.0013417797163128853,
       "learning_rate": 3.2710280373831774e-06,
-      "loss": 0.0043,
+      "loss": 0.0019,
       "step": 1500
     },
     {
       "epoch": 2.897196261682243,
-      "grad_norm": 0.0015339914243668318,
+      "grad_norm": 0.001086659263819456,
       "learning_rate": 1.7133956386292835e-06,
-      "loss": 0.0007,
+      "loss": 0.0006,
       "step": 1550
     },
     {
       "epoch": 2.9906542056074765,
-      "grad_norm": 0.001584639772772789,
+      "grad_norm": 0.001263592392206192,
       "learning_rate": 1.5576323987538942e-07,
-      "loss": 0.0008,
+      "loss": 0.0007,
       "step": 1600
     },
     {
       "epoch": 3.0,
-      "eval_loss": 6.102351665496826,
-      "eval_runtime": 0.3512,
-      "eval_samples_per_second": 14.238,
-      "eval_steps_per_second": 2.848,
+      "eval_loss": 6.119048595428467,
+      "eval_runtime": 0.3387,
+      "eval_samples_per_second": 14.762,
+      "eval_steps_per_second": 2.952,
       "step": 1605
     }
   ],

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a78c45c29ec4d43b9c9d84054394a0b682489b9ba02524e706f2df280b697526
+oid sha256:0b892a1da419ea5e2308ca459691451091eb08ff29ec453dc60c71d90f9cda75
 size 1112223464