Training in progress, step 3, checkpoint

last-checkpoint/adapter_config.json

@@ -20,10 +20,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "dense",
-    "query_key_value",
     "dense_h_to_4h",
-    "dense_4h_to_h"
+    "dense_4h_to_h",
+    "query_key_value",
+    "dense"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4aa70cea4b1cc9d96e392c895307f7eedb0937f41e829fbf4e2100dc757bbd7c
+oid sha256:3c05cc0b163026ca4dd2f7ad428f0b454113c3e12711501af2ad78c95d40696a
 size 3152296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86e8d5f05b68d7f2cab479d6959be4f080fea281130fece63715175019ec7796
+oid sha256:8cc763d5997639d658a81914c963d0688ab037315f2e12c5cda0f6b9f3548700
 size 1656058

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:456bf484f2805c5eb5383d929dec834f74d12009334db2c37db29af2053d4bfc
+oid sha256:fe474c19ed55a700df6d0f4840ca2274d4f5b8c39547d46aa2909b1442c2b257
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0099cb7287625b29b67c4fcf42ff20fae623b429bfb10f5ac695bc54f2be54fd
+oid sha256:ffd8c58e5d02492554dbaa495f8cf80dff41fabc0e1288cb2fd18a103d402219
 size 1064

last-checkpoint/trainer_state.json

@@ -1,16 +1,16 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 8.899535370090886e-05,
+  "epoch": 4.449767685045443e-05,
   "eval_steps": 3,
-  "global_step": 6,
+  "global_step": 3,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.4832558950151478e-05,
-      "grad_norm": 1051.87255859375,
+      "grad_norm": 992.8629150390625,
       "learning_rate": 2e-05,
       "loss": 95.6979,
       "step": 1
@@ -18,61 +18,32 @@
     {
       "epoch": 1.4832558950151478e-05,
       "eval_loss": 23.534069061279297,
-      "eval_runtime": 1833.9767,
-      "eval_samples_per_second": 15.478,
-      "eval_steps_per_second": 7.739,
+      "eval_runtime": 1783.7836,
+      "eval_samples_per_second": 15.914,
+      "eval_steps_per_second": 7.957,
       "step": 1
     },
     {
       "epoch": 2.9665117900302956e-05,
-      "grad_norm": 1240.6348876953125,
+      "grad_norm": 1293.0372314453125,
       "learning_rate": 4e-05,
       "loss": 96.301,
       "step": 2
     },
     {
       "epoch": 4.449767685045443e-05,
-      "grad_norm": 1472.569580078125,
+      "grad_norm": 1802.606689453125,
       "learning_rate": 6e-05,
-      "loss": 94.7081,
+      "loss": 94.739,
       "step": 3
     },
     {
       "epoch": 4.449767685045443e-05,
-      "eval_loss": 23.525732040405273,
-      "eval_runtime": 1617.8878,
-      "eval_samples_per_second": 17.546,
-      "eval_steps_per_second": 8.773,
-      "step": 3
-    },
-    {
-      "epoch": 5.933023580060591e-05,
-      "grad_norm": 1074.8275146484375,
-      "learning_rate": 8e-05,
-      "loss": 93.5059,
-      "step": 4
-    },
-    {
-      "epoch": 7.416279475075739e-05,
-      "grad_norm": 952.0130615234375,
-      "learning_rate": 0.0001,
-      "loss": 94.6268,
-      "step": 5
-    },
-    {
-      "epoch": 8.899535370090886e-05,
-      "grad_norm": 1127.652587890625,
-      "learning_rate": 0.00012,
-      "loss": 94.5666,
-      "step": 6
-    },
-    {
-      "epoch": 8.899535370090886e-05,
-      "eval_loss": 23.497276306152344,
-      "eval_runtime": 1782.5657,
-      "eval_samples_per_second": 15.925,
+      "eval_loss": 23.527429580688477,
+      "eval_runtime": 1782.403,
+      "eval_samples_per_second": 15.926,
       "eval_steps_per_second": 7.963,
-      "step": 6
+      "step": 3
     }
   ],
   "logging_steps": 1,
@@ -92,7 +63,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 53623740432384.0,
+  "total_flos": 26811870216192.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8bbc8d65aeab4741d54437118f2d0637fe81692f25988a2c5190b8981bd90935
+oid sha256:58a37180f364930b594f306b55ac126974852e8447dfe8f2fffd0623e6ff8dc8
 size 6776