Training in progress, step 1000, checkpoint

last-checkpoint/config.json

@@ -4,7 +4,7 @@
     "EsmForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.01,
-  "classifier_dropout": 0.0,
+  "classifier_dropout": 0.2,
   "emb_layer_norm_before": false,
   "esmfold_config": null,
   "hidden_act": "gelu",

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb21af9c25c39b2cb7668efddc847d2ba5af7361978f82387c531754ecc95b7a
+oid sha256:1c2b017b5fe60f0cbc2c91063a517d21b0d2ccfd36083a2d6e82ab7d3ad67efe
 size 135998360

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7be706b7ff7d94c86b7d21f0d4997a4cc156253d2c391d8cf855066e9c999995
+oid sha256:edc21caa0587cb4e7837fb779d2967a2e952edf1d73a4dc4d6acb6fe57dc9c06
 size 268169559

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d2452ab08f456f1369c7e1ada99b0eb5f5bbd423b73abc458a65e4ad932a03d
+oid sha256:4a02ebaae927db17be3e9d7c65bb8a88e4e7211cf37f4b41ea598035b1fc61b0
 size 623

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.09097296744585037,
+  "best_metric": 0.0718587264418602,
   "best_model_checkpoint": "mgh6/TCS_Pairing/checkpoint-1000",
   "epoch": 0.11934598400763814,
   "eval_steps": 1000,
@@ -10,23 +10,23 @@
   "log_history": [
     {
       "epoch": 0.12,
-      "learning_rate": 9.40327007996181e-06,
-      "loss": 0.1125,
+      "learning_rate": 9.602180053307873e-05,
+      "loss": 0.0929,
       "step": 1000
     },
     {
       "epoch": 0.12,
-      "eval_loss": 0.09097296744585037,
-      "eval_runtime": 142.0975,
-      "eval_samples_per_second": 178.765,
-      "eval_steps_per_second": 22.351,
+      "eval_loss": 0.0718587264418602,
+      "eval_runtime": 142.087,
+      "eval_samples_per_second": 178.778,
+      "eval_steps_per_second": 22.353,
       "step": 1000
     }
   ],
   "logging_steps": 1000,
-  "max_steps": 16758,
+  "max_steps": 25137,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 3,
   "save_steps": 1000,
   "total_flos": 2500930816896000.0,
   "train_batch_size": 8,

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:714bf81117703d9ab46d5e672bec52bde0e0ec1932ff5038a8492eca82e1b939
+oid sha256:ba2f5a5f4a22c32fb4c184c9f2e52c873b6a3af6d30c870c875739a2d0c9b17c
 size 4271