Training in progress, epoch 3, checkpoint

checkpoint-243/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e54982edf4e429f30750c11cbc11500cd2273acd13dda3b5b312ccf2785d8a89
+oid sha256:8e62698666c26f0113bedfd7a96468065c3cfdb715cc9ca53c409c6fd2dfe401
 size 592324828

checkpoint-243/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1be6b69814b8aecab7c71b56c11c28dac0b4e7452c8079d9f462fc628ccf7639
+oid sha256:85e5597180e12da6c3ba8c8f36fd77839bcb000a5d5fcc96939e7135a114be15
 size 1014657786

checkpoint-243/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4510120abb9e19965e0fc90a9342c8b839c7b334c4b005e1a74bbda01d4f51cc
+oid sha256:becd6836611c76f35df7d32580a0d0ecb16ccf6774d717a60f2504beb710244c
 size 14244

checkpoint-243/trainer_state.json

@@ -11,135 +11,135 @@
     {
       "epoch": 1.0,
       "eval_Claim": {
-        "f1-score": 0.44734338904742493,
-        "precision": 0.5477537437603993,
-        "recall": 0.3780431786862655,
-        "support": 4354.0
+        "f1-score": 0.38183790485045715,
+        "precision": 0.6497890295358649,
+        "recall": 0.27035330261136714,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.6504965622612681,
-        "precision": 0.5898856944925528,
-        "recall": 0.7249893571732652,
-        "support": 2349.0
+        "f1-score": 0.6266258607498087,
+        "precision": 0.5535653937140926,
+        "recall": 0.7219039224327898,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.8859784283513097,
-        "precision": 0.8689081979599547,
-        "recall": 0.9037328094302554,
-        "support": 10180.0
+        "f1-score": 0.8545803971812939,
+        "precision": 0.8443037974683544,
+        "recall": 0.8651102464332037,
+        "support": 8481.0
       },
       "eval_Premise": {
-        "f1-score": 0.8720120805863504,
-        "precision": 0.859258183929738,
-        "recall": 0.8851502916105877,
-        "support": 13374.0
+        "f1-score": 0.878949075575738,
+        "precision": 0.8314310260186549,
+        "recall": 0.9322278794550709,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.8059953068711373,
-      "eval_loss": 0.47828415036201477,
+      "eval_accuracy": 0.7960859220535504,
+      "eval_loss": 0.5359914898872375,
       "eval_macro avg": {
-        "f1-score": 0.7139576150615883,
-        "precision": 0.7164514550356611,
-        "recall": 0.7229789092250934,
-        "support": 30257.0
-      },
-      "eval_runtime": 4.8422,
-      "eval_samples_per_second": 16.728,
-      "eval_steps_per_second": 2.272,
+        "f1-score": 0.6854983095893245,
+        "precision": 0.7197723116842417,
+        "recall": 0.697398837733108,
+        "support": 29841.0
+      },
+      "eval_runtime": 4.8191,
+      "eval_samples_per_second": 16.601,
+      "eval_steps_per_second": 2.075,
       "eval_weighted avg": {
-        "f1-score": 0.7984036588902531,
-        "precision": 0.7967665566250599,
-        "recall": 0.8059953068711373,
-        "support": 30257.0
+        "f1-score": 0.7769240850962494,
+        "precision": 0.7862232339538372,
+        "recall": 0.7960859220535504,
+        "support": 29841.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_Claim": {
-        "f1-score": 0.5787561000934482,
-        "precision": 0.5281409891984082,
-        "recall": 0.6401010564997703,
-        "support": 4354.0
+        "f1-score": 0.6161300742974051,
+        "precision": 0.6048625792811839,
+        "recall": 0.6278253236778583,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.7702650787552824,
-        "precision": 0.7017850892544627,
-        "recall": 0.8535547041294168,
-        "support": 2349.0
+        "f1-score": 0.7920215150156881,
+        "precision": 0.8057455540355677,
+        "recall": 0.7787571617452622,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9095337508698677,
-        "precision": 0.9206077681626081,
-        "recall": 0.8987229862475442,
-        "support": 10180.0
+        "f1-score": 0.8922796352583586,
+        "precision": 0.9209436566695947,
+        "recall": 0.865346067680698,
+        "support": 8481.0
       },
       "eval_Premise": {
-        "f1-score": 0.8736648538675224,
-        "precision": 0.9162905211325401,
-        "recall": 0.8348287722446538,
-        "support": 13374.0
+        "f1-score": 0.8914289590611538,
+        "precision": 0.8790554552143957,
+        "recall": 0.9041557726709784,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.8297584030141786,
-      "eval_loss": 0.4241589605808258,
+      "eval_accuracy": 0.8413927147213566,
+      "eval_loss": 0.443877637386322,
       "eval_macro avg": {
-        "f1-score": 0.7830549458965301,
-        "precision": 0.7667060919370048,
-        "recall": 0.8068018797803463,
-        "support": 30257.0
-      },
-      "eval_runtime": 4.8636,
-      "eval_samples_per_second": 16.654,
-      "eval_steps_per_second": 2.262,
+        "f1-score": 0.7979650459081514,
+        "precision": 0.8026518113001855,
+        "recall": 0.7940210814436992,
+        "support": 29841.0
+      },
+      "eval_runtime": 4.8087,
+      "eval_samples_per_second": 16.637,
+      "eval_steps_per_second": 2.08,
       "eval_weighted avg": {
-        "f1-score": 0.8352680063880269,
-        "precision": 0.8452350051608073,
-        "recall": 0.8297584030141786,
-        "support": 30257.0
+        "f1-score": 0.842071433389123,
+        "precision": 0.8435143116581857,
+        "recall": 0.8413927147213566,
+        "support": 29841.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_Claim": {
-        "f1-score": 0.5728465083303793,
-        "precision": 0.5899245558530056,
-        "recall": 0.5567294441892513,
-        "support": 4354.0
+        "f1-score": 0.6042799305957202,
+        "precision": 0.6389432485322897,
+        "recall": 0.5731841123546193,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.7862876979822088,
-        "precision": 0.8017699115044248,
-        "recall": 0.7713920817369093,
-        "support": 2349.0
+        "f1-score": 0.8349557522123894,
+        "precision": 0.8382940915148823,
+        "recall": 0.8316438959894227,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.914532821534425,
-        "precision": 0.9048168445341794,
-        "recall": 0.9244597249508841,
-        "support": 10180.0
+        "f1-score": 0.8941106484235574,
+        "precision": 0.9022691799735862,
+        "recall": 0.8860983374602052,
+        "support": 8481.0
       },
       "eval_Premise": {
-        "f1-score": 0.8880533114924986,
-        "precision": 0.8843330614665975,
-        "recall": 0.8918049947659639,
-        "support": 13374.0
+        "f1-score": 0.8917763490086511,
+        "precision": 0.8729980887102089,
+        "recall": 0.9113802119168846,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.8452258981392735,
-      "eval_loss": 0.41067537665367126,
+      "eval_accuracy": 0.8464863778023525,
+      "eval_loss": 0.4555923044681549,
       "eval_macro avg": {
-        "f1-score": 0.7904300848348779,
-        "precision": 0.7952110933395518,
-        "recall": 0.7860965614107521,
-        "support": 30257.0
-      },
-      "eval_runtime": 4.8684,
-      "eval_samples_per_second": 16.638,
-      "eval_steps_per_second": 2.259,
+        "f1-score": 0.8062806700600795,
+        "precision": 0.8131261521827418,
+        "recall": 0.8005766394302829,
+        "support": 29841.0
+      },
+      "eval_runtime": 4.7811,
+      "eval_samples_per_second": 16.733,
+      "eval_steps_per_second": 2.092,
       "eval_weighted avg": {
-        "f1-score": 0.8437033615676307,
-        "precision": 0.8424495118392472,
-        "recall": 0.8452258981392735,
-        "support": 30257.0
+        "f1-score": 0.8442159482345275,
+        "precision": 0.8429359878649869,
+        "recall": 0.8464863778023525,
+        "support": 29841.0
       },
       "step": 243
     }
@@ -149,7 +149,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 430026551602200.0,
+  "total_flos": 431366198180400.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

checkpoint-243/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ce83aa27c1a93c128c1dae6a1646206ab5d85d935efad925ac4e4beb0be29a0
+oid sha256:19f05e33d3889e7c1ce1bf18bf9e16659f52e74109a304d618b5d09955736add
 size 4856