diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round10.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..43c5b73a1d0a86a21f198bef6406f136b022c02e
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f06a89dfcac5961ac5bdb293a2a3f574fcb9216e4593500ab6c2155ff3df0313
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round12.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9cdd2f16fa70aebaec4d2fe37636e4d9fc050928
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fe982dd9d5a3ae03280e3447d8063dcc43629d7dca6b8e90b03907c76b90a58
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round15.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round15.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f6b47e6a56d882849d9613fa0eef0899682fb289
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round15.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aedba7096d5b5406fd46033b0b93b844abbe3d7454189a2035107d0dbb00caaa
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round17.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round17.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9267382f11615a7b5ca809afe52900b3b79f4a58
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round17.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0526023fc6ccb898eb90c77e6bdc99b569624d6470ad5b920a20e689c289ad2b
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round2.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..55863cf2228aec2289bed847d2026e23be276129
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a263e7e7ee03bfcf87feac35953e8aa2fa2ce0c0e3c56c5dffbcf6c8a7d52721
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round20.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round20.pth
new file mode 100644
index 0000000000000000000000000000000000000000..8c3d9fc3a7bb3d389a5ecb7de29724858d033ab7
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round20.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8e10816fae9b78236f55f93f29814d10a19539fd654a0c3e9737cbd47cd8d9d
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round5.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..17f7860a79b43e056c4a34dff64cc07ac7ff59a0
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f4335cc459cd1552cda05f02b928ed4417cd0d799ff11e7ef4d08a16e0b8484
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round7.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3a9a0073e92f2d3a31756ec70dbaf30c23ae6828
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_client_model_round7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c206007eb99dbc9905215c8f5443524bbcf2827728a8be2f97d0edee647ca575
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_trainer_state.json b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..acd2e41d1419d0b075fa2fe12752a2fc4e909b9a
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/0_trainer_state.json
@@ -0,0 +1,189 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 43,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09302325581395349,
+      "grad_norm": 2.9576117992401123,
+      "learning_rate": 2e-05,
+      "loss": 0.2411,
+      "step": 2
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 1.2031710147857666,
+      "learning_rate": 2e-05,
+      "loss": 0.0578,
+      "step": 4
+    },
+    {
+      "epoch": 0.27906976744186046,
+      "grad_norm": 2.9653682708740234,
+      "learning_rate": 2e-05,
+      "loss": 0.3511,
+      "step": 6
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 4.665686130523682,
+      "learning_rate": 2e-05,
+      "loss": 0.1305,
+      "step": 8
+    },
+    {
+      "epoch": 0.46511627906976744,
+      "grad_norm": 2.068437337875366,
+      "learning_rate": 2e-05,
+      "loss": 0.1576,
+      "step": 10
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 0.7609916925430298,
+      "learning_rate": 2e-05,
+      "loss": 0.1547,
+      "step": 12
+    },
+    {
+      "epoch": 0.6511627906976745,
+      "grad_norm": 0.8752633333206177,
+      "learning_rate": 2e-05,
+      "loss": 0.2439,
+      "step": 14
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 0.7228463292121887,
+      "learning_rate": 2e-05,
+      "loss": 0.0869,
+      "step": 16
+    },
+    {
+      "epoch": 0.8372093023255814,
+      "grad_norm": 0.5220616459846497,
+      "learning_rate": 2e-05,
+      "loss": 0.1898,
+      "step": 18
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 0.5391299724578857,
+      "learning_rate": 2e-05,
+      "loss": 0.0884,
+      "step": 20
+    },
+    {
+      "epoch": 1.0232558139534884,
+      "grad_norm": 0.4390765428543091,
+      "learning_rate": 2e-05,
+      "loss": 0.038,
+      "step": 22
+    },
+    {
+      "epoch": 1.1162790697674418,
+      "grad_norm": 4.568640232086182,
+      "learning_rate": 2e-05,
+      "loss": 0.2474,
+      "step": 24
+    },
+    {
+      "epoch": 1.2093023255813953,
+      "grad_norm": 4.5196356773376465,
+      "learning_rate": 2e-05,
+      "loss": 0.3904,
+      "step": 26
+    },
+    {
+      "epoch": 1.302325581395349,
+      "grad_norm": 0.23186014592647552,
+      "learning_rate": 2e-05,
+      "loss": 0.101,
+      "step": 28
+    },
+    {
+      "epoch": 1.3953488372093024,
+      "grad_norm": 2.5624356269836426,
+      "learning_rate": 2e-05,
+      "loss": 0.1768,
+      "step": 30
+    },
+    {
+      "epoch": 1.4883720930232558,
+      "grad_norm": 0.4649677872657776,
+      "learning_rate": 2e-05,
+      "loss": 0.0719,
+      "step": 32
+    },
+    {
+      "epoch": 1.5813953488372092,
+      "grad_norm": 0.7176182270050049,
+      "learning_rate": 2e-05,
+      "loss": 0.1396,
+      "step": 34
+    },
+    {
+      "epoch": 1.6744186046511629,
+      "grad_norm": 2.2132463455200195,
+      "learning_rate": 2e-05,
+      "loss": 0.087,
+      "step": 36
+    },
+    {
+      "epoch": 1.7674418604651163,
+      "grad_norm": 1.706496238708496,
+      "learning_rate": 2e-05,
+      "loss": 0.2997,
+      "step": 38
+    },
+    {
+      "epoch": 1.8604651162790697,
+      "grad_norm": 0.48359882831573486,
+      "learning_rate": 2e-05,
+      "loss": 0.0237,
+      "step": 40
+    },
+    {
+      "epoch": 1.9534883720930232,
+      "grad_norm": 0.40984606742858887,
+      "learning_rate": 2e-05,
+      "loss": 0.0393,
+      "step": 42
+    },
+    {
+      "epoch": 2.0,
+      "step": 43,
+      "total_flos": 4777205812953088.0,
+      "train_loss": 0.15616962521575217,
+      "train_runtime": 217.1153,
+      "train_samples_per_second": 0.792,
+      "train_steps_per_second": 0.198
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 43,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4777205812953088.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round10.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9f6b8987f65f08f2381f74183a36e21f649baf85
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71c6fce6889d0763e53ec11b925ecf780872259115e0a4de278fecb1730f1c44
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round12.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..dfc39bd7a41f242a1297c2821dd73cb6b32967ab
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:818822b11989459a5fc595b2555903333d4fbd06ec209d9f8fd739991ecd393f
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round15.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round15.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e88e826b091094ff9b80e41a9de93de8fe9e8951
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round15.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c441dfd77124cafa5afe9a4f3e30bd524ebfeb667ce4cb78133a3d18595567b1
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round17.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round17.pth
new file mode 100644
index 0000000000000000000000000000000000000000..df360d80eb06e5af939e27a6a7617f15b09e5664
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round17.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76093b70c93eb8976c70ba3b52470568b0f3d0896ec2dec92615014a4a25ae3e
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round2.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a21be7dab2c1a9b4bd539bb647c1179bbfd8350c
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27e267c53893530bd8784e742b230c103f0af6e3be5e71601f6620f4c0100934
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round20.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round20.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b6974a850332bf7e78861119e009141d09eebae4
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round20.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:225727d72f520f2af545e163e03222702bf08e8084ca44d12dd0337c3ad2899d
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round5.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..fa31e85fe10a4bcfea5873d74d1064b53c22a218
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51d18abf0982eaad9817b60cffb3c40f99da4761429ac0c24f6149861969da4c
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round7.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..54db12159d7b653e27d9bc79685a903f0d8b5f72
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_client_model_round7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e2e06f880660e7b5ffca36b45892d7f2ed89797c6ff687b9dc8da9431886cfa
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_trainer_state.json b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..24df07b7aa67c1eaf81494b9d3f1673a5fe278f3
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/1_trainer_state.json
@@ -0,0 +1,189 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 43,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09302325581395349,
+      "grad_norm": 0.3085244596004486,
+      "learning_rate": 2e-05,
+      "loss": 0.1226,
+      "step": 2
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 3.019911289215088,
+      "learning_rate": 2e-05,
+      "loss": 0.2883,
+      "step": 4
+    },
+    {
+      "epoch": 0.27906976744186046,
+      "grad_norm": 1.109711766242981,
+      "learning_rate": 2e-05,
+      "loss": 0.1198,
+      "step": 6
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 0.4532826840877533,
+      "learning_rate": 2e-05,
+      "loss": 0.0175,
+      "step": 8
+    },
+    {
+      "epoch": 0.46511627906976744,
+      "grad_norm": 0.8678146600723267,
+      "learning_rate": 2e-05,
+      "loss": 0.0323,
+      "step": 10
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 2.618272066116333,
+      "learning_rate": 2e-05,
+      "loss": 0.1189,
+      "step": 12
+    },
+    {
+      "epoch": 0.6511627906976745,
+      "grad_norm": 0.20907337963581085,
+      "learning_rate": 2e-05,
+      "loss": 0.0141,
+      "step": 14
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 1.467644214630127,
+      "learning_rate": 2e-05,
+      "loss": 0.1211,
+      "step": 16
+    },
+    {
+      "epoch": 0.8372093023255814,
+      "grad_norm": 1.8505585193634033,
+      "learning_rate": 2e-05,
+      "loss": 0.0966,
+      "step": 18
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 0.5534336566925049,
+      "learning_rate": 2e-05,
+      "loss": 0.0249,
+      "step": 20
+    },
+    {
+      "epoch": 1.0232558139534884,
+      "grad_norm": 0.42154374718666077,
+      "learning_rate": 2e-05,
+      "loss": 0.0169,
+      "step": 22
+    },
+    {
+      "epoch": 1.1162790697674418,
+      "grad_norm": 0.37012800574302673,
+      "learning_rate": 2e-05,
+      "loss": 0.0338,
+      "step": 24
+    },
+    {
+      "epoch": 1.2093023255813953,
+      "grad_norm": 0.9166113138198853,
+      "learning_rate": 2e-05,
+      "loss": 0.0618,
+      "step": 26
+    },
+    {
+      "epoch": 1.302325581395349,
+      "grad_norm": 0.27272695302963257,
+      "learning_rate": 2e-05,
+      "loss": 0.0284,
+      "step": 28
+    },
+    {
+      "epoch": 1.3953488372093024,
+      "grad_norm": 0.2518395781517029,
+      "learning_rate": 2e-05,
+      "loss": 0.0126,
+      "step": 30
+    },
+    {
+      "epoch": 1.4883720930232558,
+      "grad_norm": 0.7915253043174744,
+      "learning_rate": 2e-05,
+      "loss": 0.0801,
+      "step": 32
+    },
+    {
+      "epoch": 1.5813953488372092,
+      "grad_norm": 0.902931809425354,
+      "learning_rate": 2e-05,
+      "loss": 0.0334,
+      "step": 34
+    },
+    {
+      "epoch": 1.6744186046511629,
+      "grad_norm": 0.3951447308063507,
+      "learning_rate": 2e-05,
+      "loss": 0.0307,
+      "step": 36
+    },
+    {
+      "epoch": 1.7674418604651163,
+      "grad_norm": 2.1907150745391846,
+      "learning_rate": 2e-05,
+      "loss": 0.0596,
+      "step": 38
+    },
+    {
+      "epoch": 1.8604651162790697,
+      "grad_norm": 1.2620713710784912,
+      "learning_rate": 2e-05,
+      "loss": 0.0869,
+      "step": 40
+    },
+    {
+      "epoch": 1.9534883720930232,
+      "grad_norm": 0.3642316162586212,
+      "learning_rate": 2e-05,
+      "loss": 0.0304,
+      "step": 42
+    },
+    {
+      "epoch": 2.0,
+      "step": 43,
+      "total_flos": 4812138065428480.0,
+      "train_loss": 0.06693689213242641,
+      "train_runtime": 216.4073,
+      "train_samples_per_second": 0.795,
+      "train_steps_per_second": 0.199
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 43,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4812138065428480.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round10.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..4da0480a0fd26020a39d322da019b509cb51c19e
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69816f63dda6857999b2188db1a0aaaa9147aab897dcee7573e9d55dbdf1b2cc
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round12.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e8662966e341488d2ea553527ea2dc2bcf3d98ec
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c914d80153a80299ed8993a6ceb64539f8af74f9b6e40638c543863a69ad54f
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round15.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round15.pth
new file mode 100644
index 0000000000000000000000000000000000000000..50d207e9107a0be914b09ac4cfc33cf0dc8e5adc
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round15.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ce361c6c22db8c7c4ebfeaae20c30cb50de1598a3657cb2b4d2a0239088930b
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round17.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round17.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c4e1a34470c8327acc095111e77191b94112b92e
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round17.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b726647932691999b54f2ab9ec9197c040c249348e7cb4864fddba8cca057773
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round2.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..dadbbfb8741ccbb6db99c5860ab18d78d3e820f1
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63647af6cd32eb3f3541fd71185343a2b826fe5badc33e786d96673a7068f25e
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round20.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round20.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5d3cdb6a916c6bb0889c251fef282a94f04b9d0e
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round20.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbae85d3b932e88792d442e91b69a8dd7af46ddbc2f0e332cb4c695c7a9f454b
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round5.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a7323fef1173a173a0daa3c777c1e0b1aebd8cd7
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc53b17de3dc585196fdcf469a3d94372c45f6eff9d977f4523242386413acb0
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round7.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e78b2be4fd00331ed051ac1ef50845e4c45fb7d2
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_client_model_round7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:638b6048a232e5fd62e1b30f85d1598fe8d2c095bbb76d1ca75681d6b97b62b3
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_trainer_state.json b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..7577933677e92ea08346422f4e8ccabedb727696
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/2_trainer_state.json
@@ -0,0 +1,189 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 43,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09302325581395349,
+      "grad_norm": 3.3456127643585205,
+      "learning_rate": 2e-05,
+      "loss": 0.2545,
+      "step": 2
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 0.6647293567657471,
+      "learning_rate": 2e-05,
+      "loss": 0.143,
+      "step": 4
+    },
+    {
+      "epoch": 0.27906976744186046,
+      "grad_norm": 2.341078519821167,
+      "learning_rate": 2e-05,
+      "loss": 0.1575,
+      "step": 6
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 1.038511037826538,
+      "learning_rate": 2e-05,
+      "loss": 0.1649,
+      "step": 8
+    },
+    {
+      "epoch": 0.46511627906976744,
+      "grad_norm": 1.1091891527175903,
+      "learning_rate": 2e-05,
+      "loss": 0.1359,
+      "step": 10
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 0.9437009692192078,
+      "learning_rate": 2e-05,
+      "loss": 0.11,
+      "step": 12
+    },
+    {
+      "epoch": 0.6511627906976745,
+      "grad_norm": 2.193312168121338,
+      "learning_rate": 2e-05,
+      "loss": 0.2129,
+      "step": 14
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 1.4751211404800415,
+      "learning_rate": 2e-05,
+      "loss": 0.0952,
+      "step": 16
+    },
+    {
+      "epoch": 0.8372093023255814,
+      "grad_norm": 0.38253253698349,
+      "learning_rate": 2e-05,
+      "loss": 0.0647,
+      "step": 18
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 0.6969690322875977,
+      "learning_rate": 2e-05,
+      "loss": 0.1394,
+      "step": 20
+    },
+    {
+      "epoch": 1.0232558139534884,
+      "grad_norm": 0.7709684371948242,
+      "learning_rate": 2e-05,
+      "loss": 0.1008,
+      "step": 22
+    },
+    {
+      "epoch": 1.1162790697674418,
+      "grad_norm": 2.740997552871704,
+      "learning_rate": 2e-05,
+      "loss": 0.1238,
+      "step": 24
+    },
+    {
+      "epoch": 1.2093023255813953,
+      "grad_norm": 1.5686851739883423,
+      "learning_rate": 2e-05,
+      "loss": 0.1016,
+      "step": 26
+    },
+    {
+      "epoch": 1.302325581395349,
+      "grad_norm": 0.3873344659805298,
+      "learning_rate": 2e-05,
+      "loss": 0.0211,
+      "step": 28
+    },
+    {
+      "epoch": 1.3953488372093024,
+      "grad_norm": 0.8399427533149719,
+      "learning_rate": 2e-05,
+      "loss": 0.1423,
+      "step": 30
+    },
+    {
+      "epoch": 1.4883720930232558,
+      "grad_norm": 2.8620800971984863,
+      "learning_rate": 2e-05,
+      "loss": 0.2682,
+      "step": 32
+    },
+    {
+      "epoch": 1.5813953488372092,
+      "grad_norm": 1.715096116065979,
+      "learning_rate": 2e-05,
+      "loss": 0.1543,
+      "step": 34
+    },
+    {
+      "epoch": 1.6744186046511629,
+      "grad_norm": 0.39780232310295105,
+      "learning_rate": 2e-05,
+      "loss": 0.0503,
+      "step": 36
+    },
+    {
+      "epoch": 1.7674418604651163,
+      "grad_norm": 1.9032905101776123,
+      "learning_rate": 2e-05,
+      "loss": 0.1529,
+      "step": 38
+    },
+    {
+      "epoch": 1.8604651162790697,
+      "grad_norm": 1.9974933862686157,
+      "learning_rate": 2e-05,
+      "loss": 0.1527,
+      "step": 40
+    },
+    {
+      "epoch": 1.9534883720930232,
+      "grad_norm": 0.41782402992248535,
+      "learning_rate": 2e-05,
+      "loss": 0.1376,
+      "step": 42
+    },
+    {
+      "epoch": 2.0,
+      "step": 43,
+      "total_flos": 4715460834623488.0,
+      "train_loss": 0.13745101662569267,
+      "train_runtime": 217.324,
+      "train_samples_per_second": 0.791,
+      "train_steps_per_second": 0.198
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 43,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4715460834623488.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round10.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5b21b5be3aab5882b048175bb708f4479ad136e4
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9f0a3b90e19eabbfc0e5635fe7059549c618b6e036f9f45a4372451a8a36e5e
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round12.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d2c71b14723f88612f80100dcf9ad6a10a2772e7
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4567d214400f00e8531f2289e87b461d7e2ff36d3599953485e74c95a50ae33
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round15.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round15.pth
new file mode 100644
index 0000000000000000000000000000000000000000..12b1bbe6f9111d18a499b27ab4174586709d1c7e
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round15.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36f52ffe528820e313e8399cc62c04cbc2bf7042f11dc9809454f8d9ac8e6419
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round17.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round17.pth
new file mode 100644
index 0000000000000000000000000000000000000000..14322fe73a0876b30b2eff25d8e23afb73655ea5
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round17.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95607ac19caf68871278ba85509065e63ada58d495744bd25ca2ae6c2627a16d
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round2.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6ee60c12dcb5e8fe9c84af689f625cd121013d84
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5921fb1eef7a3d375c934d78dd90029f1c52ed7f25c757f29f83bb17cced3e9d
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round20.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round20.pth
new file mode 100644
index 0000000000000000000000000000000000000000..158a1a6f29bb1786900c887ec7d18ee350f8fecf
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round20.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b1eefd262a4d3ee7c54828027c97f4b11f4186d76647bcae6b7ff7378e1a742
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round5.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..04ac03642a9d7545140198f6d2b088b0dd1f1fdb
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fc91a80a0564d2d9cc2000161fd3302ed3cd3d82edead6b1e60e8270032e1a3
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round7.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..fdc6f78d89e99197747afc1b76a68a8e67fb63ea
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_client_model_round7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ccdf785607bf48d74b05613167b1393b6d33de3717fd8db901693e4439448fa
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_trainer_state.json b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..9c993bf480388a7d48deb7744c141e979994057e
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/3_trainer_state.json
@@ -0,0 +1,189 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 43,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09302325581395349,
+      "grad_norm": 1.1141589879989624,
+      "learning_rate": 2e-05,
+      "loss": 0.1236,
+      "step": 2
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 1.2319393157958984,
+      "learning_rate": 2e-05,
+      "loss": 0.3126,
+      "step": 4
+    },
+    {
+      "epoch": 0.27906976744186046,
+      "grad_norm": 2.423560857772827,
+      "learning_rate": 2e-05,
+      "loss": 0.4022,
+      "step": 6
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 2.308655023574829,
+      "learning_rate": 2e-05,
+      "loss": 0.2367,
+      "step": 8
+    },
+    {
+      "epoch": 0.46511627906976744,
+      "grad_norm": 1.6019493341445923,
+      "learning_rate": 2e-05,
+      "loss": 0.1691,
+      "step": 10
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 2.4790854454040527,
+      "learning_rate": 2e-05,
+      "loss": 0.2446,
+      "step": 12
+    },
+    {
+      "epoch": 0.6511627906976745,
+      "grad_norm": 1.2402669191360474,
+      "learning_rate": 2e-05,
+      "loss": 0.1289,
+      "step": 14
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 4.58521842956543,
+      "learning_rate": 2e-05,
+      "loss": 0.1723,
+      "step": 16
+    },
+    {
+      "epoch": 0.8372093023255814,
+      "grad_norm": 1.7242616415023804,
+      "learning_rate": 2e-05,
+      "loss": 0.1184,
+      "step": 18
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 2.655269145965576,
+      "learning_rate": 2e-05,
+      "loss": 0.244,
+      "step": 20
+    },
+    {
+      "epoch": 1.0232558139534884,
+      "grad_norm": 2.253636121749878,
+      "learning_rate": 2e-05,
+      "loss": 0.2518,
+      "step": 22
+    },
+    {
+      "epoch": 1.1162790697674418,
+      "grad_norm": 3.0500082969665527,
+      "learning_rate": 2e-05,
+      "loss": 0.2864,
+      "step": 24
+    },
+    {
+      "epoch": 1.2093023255813953,
+      "grad_norm": 2.2204771041870117,
+      "learning_rate": 2e-05,
+      "loss": 0.31,
+      "step": 26
+    },
+    {
+      "epoch": 1.302325581395349,
+      "grad_norm": 2.925663948059082,
+      "learning_rate": 2e-05,
+      "loss": 0.5858,
+      "step": 28
+    },
+    {
+      "epoch": 1.3953488372093024,
+      "grad_norm": 1.716472864151001,
+      "learning_rate": 2e-05,
+      "loss": 0.2467,
+      "step": 30
+    },
+    {
+      "epoch": 1.4883720930232558,
+      "grad_norm": 2.464635133743286,
+      "learning_rate": 2e-05,
+      "loss": 0.2649,
+      "step": 32
+    },
+    {
+      "epoch": 1.5813953488372092,
+      "grad_norm": 0.828945517539978,
+      "learning_rate": 2e-05,
+      "loss": 0.0731,
+      "step": 34
+    },
+    {
+      "epoch": 1.6744186046511629,
+      "grad_norm": 1.1858415603637695,
+      "learning_rate": 2e-05,
+      "loss": 0.111,
+      "step": 36
+    },
+    {
+      "epoch": 1.7674418604651163,
+      "grad_norm": 3.732938766479492,
+      "learning_rate": 2e-05,
+      "loss": 0.2349,
+      "step": 38
+    },
+    {
+      "epoch": 1.8604651162790697,
+      "grad_norm": 2.46227765083313,
+      "learning_rate": 2e-05,
+      "loss": 0.3129,
+      "step": 40
+    },
+    {
+      "epoch": 1.9534883720930232,
+      "grad_norm": 1.6540902853012085,
+      "learning_rate": 2e-05,
+      "loss": 0.2932,
+      "step": 42
+    },
+    {
+      "epoch": 2.0,
+      "step": 43,
+      "total_flos": 4880262747914240.0,
+      "train_loss": 0.25222200571104536,
+      "train_runtime": 215.4858,
+      "train_samples_per_second": 0.798,
+      "train_steps_per_second": 0.2
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 43,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4880262747914240.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round10.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..43dbb67231859b2e116349f6131ec90ba64a09d0
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:676d1a1cc1b175d0235d62fc017e315c318119b6f9d2b694a759bf9012621fa2
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round12.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..1df2363101cf2ac85b1919c04dd25603f6ff8279
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4244a475c759d82fa5ffbcbfc2deb96f9ee806361d03aa645aebf32bc2d0d139
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round15.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round15.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6f4a7a174a3527952f16549fb8e97ca12ce94fa7
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round15.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae5a9f63d9f5f4fd3301f8e4874b04b5995dadb0557e81e711abcd08c7b252f6
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round17.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round17.pth
new file mode 100644
index 0000000000000000000000000000000000000000..88ef1659d6f9bfab8800a47b5f475caaf28af7df
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round17.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe972cdc2d4d4445927a1f957a4a63e44b72e60301b94c3115794d59381c74a8
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round2.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ad36f8cf32e7c0fd3f7fc199ff3422a923b13cb0
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3239853222047f1365deb3434639266349a53ca0b9050cb2fde4032b546dc38e
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round20.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round20.pth
new file mode 100644
index 0000000000000000000000000000000000000000..bec45f04aea25a59586f6d6e747106d07cabe712
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round20.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fe8e87567dbde70e61d4b3893fe944c4cfa319c4867e6672286ef519491e3ee
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round5.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..badb832b67e8da5c1f6975c14b1ffb97fc3520d0
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64cf784b6a5fa4dcec46cd536c3ea92812bcec7bee6eeb3a6b752d0d9a8adf67
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round7.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9e4929615716e47281eb07abecbb80482a99641c
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_client_model_round7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04dc540d0b92cd9e8c8b92930521667e6c090ca32a551758b702ae8e34e21395
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_trainer_state.json b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..65e4e6cdc96594115bb3638dd0613fc5fff92f7a
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/4_trainer_state.json
@@ -0,0 +1,189 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 43,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09302325581395349,
+      "grad_norm": 2.0882959365844727,
+      "learning_rate": 2e-05,
+      "loss": 0.3258,
+      "step": 2
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 1.28811776638031,
+      "learning_rate": 2e-05,
+      "loss": 0.1932,
+      "step": 4
+    },
+    {
+      "epoch": 0.27906976744186046,
+      "grad_norm": 0.822692334651947,
+      "learning_rate": 2e-05,
+      "loss": 0.0951,
+      "step": 6
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 1.5997800827026367,
+      "learning_rate": 2e-05,
+      "loss": 0.2066,
+      "step": 8
+    },
+    {
+      "epoch": 0.46511627906976744,
+      "grad_norm": 2.9270811080932617,
+      "learning_rate": 2e-05,
+      "loss": 0.2142,
+      "step": 10
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 0.9111597537994385,
+      "learning_rate": 2e-05,
+      "loss": 0.0939,
+      "step": 12
+    },
+    {
+      "epoch": 0.6511627906976745,
+      "grad_norm": 1.769669771194458,
+      "learning_rate": 2e-05,
+      "loss": 0.1228,
+      "step": 14
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 1.8244539499282837,
+      "learning_rate": 2e-05,
+      "loss": 0.1983,
+      "step": 16
+    },
+    {
+      "epoch": 0.8372093023255814,
+      "grad_norm": 1.7385451793670654,
+      "learning_rate": 2e-05,
+      "loss": 0.2394,
+      "step": 18
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 2.4531519412994385,
+      "learning_rate": 2e-05,
+      "loss": 0.1592,
+      "step": 20
+    },
+    {
+      "epoch": 1.0232558139534884,
+      "grad_norm": 0.8544302582740784,
+      "learning_rate": 2e-05,
+      "loss": 0.1533,
+      "step": 22
+    },
+    {
+      "epoch": 1.1162790697674418,
+      "grad_norm": 2.170783519744873,
+      "learning_rate": 2e-05,
+      "loss": 0.3049,
+      "step": 24
+    },
+    {
+      "epoch": 1.2093023255813953,
+      "grad_norm": 1.9732646942138672,
+      "learning_rate": 2e-05,
+      "loss": 0.3371,
+      "step": 26
+    },
+    {
+      "epoch": 1.302325581395349,
+      "grad_norm": 3.642188549041748,
+      "learning_rate": 2e-05,
+      "loss": 0.3442,
+      "step": 28
+    },
+    {
+      "epoch": 1.3953488372093024,
+      "grad_norm": 1.494310975074768,
+      "learning_rate": 2e-05,
+      "loss": 0.13,
+      "step": 30
+    },
+    {
+      "epoch": 1.4883720930232558,
+      "grad_norm": 0.3682941496372223,
+      "learning_rate": 2e-05,
+      "loss": 0.0284,
+      "step": 32
+    },
+    {
+      "epoch": 1.5813953488372092,
+      "grad_norm": 1.2264622449874878,
+      "learning_rate": 2e-05,
+      "loss": 0.1392,
+      "step": 34
+    },
+    {
+      "epoch": 1.6744186046511629,
+      "grad_norm": 0.5091031193733215,
+      "learning_rate": 2e-05,
+      "loss": 0.1509,
+      "step": 36
+    },
+    {
+      "epoch": 1.7674418604651163,
+      "grad_norm": 1.4979541301727295,
+      "learning_rate": 2e-05,
+      "loss": 0.1525,
+      "step": 38
+    },
+    {
+      "epoch": 1.8604651162790697,
+      "grad_norm": 1.4245574474334717,
+      "learning_rate": 2e-05,
+      "loss": 0.2237,
+      "step": 40
+    },
+    {
+      "epoch": 1.9534883720930232,
+      "grad_norm": 1.7625281810760498,
+      "learning_rate": 2e-05,
+      "loss": 0.2382,
+      "step": 42
+    },
+    {
+      "epoch": 2.0,
+      "step": 43,
+      "total_flos": 5466572382535680.0,
+      "train_loss": 0.19268665757290152,
+      "train_runtime": 218.8116,
+      "train_samples_per_second": 0.786,
+      "train_steps_per_second": 0.197
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 43,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5466572382535680.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round10.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e8143e72fa013e502d4f080b05361ecb42773878
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88808d32046e53379114e941c371bd75493bde5bb1542d9334164b318c408b0a
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round12.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..bcc445fca28d2781843d2b93a98fb0e2736b6b7a
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50aa5dce46ef29bd53549ca4bb111aa001b4b81fcbdcd42593a2e89c24d1e609
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round15.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round15.pth
new file mode 100644
index 0000000000000000000000000000000000000000..067fceaf6ef83c4bf52a4bc94575a9e37b7e982e
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round15.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1d29c2144b0318477b6e17f2c8a95e0b1b6823db4c651660c3937d120884472
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round17.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round17.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6ff06203fe7f99ee234df7166bf74c6a5e129b1a
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round17.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b310990f72f64fe576d22dcaac6e3f75e347ddddd2ac489381cbd61f1b47a75
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round2.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c18efa264a3de95fbe957637a5d0638b4af015bb
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a072c7cecc25d135822bb331490b5401228a3292ef9795ef78ae35d285a7e882
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round20.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round20.pth
new file mode 100644
index 0000000000000000000000000000000000000000..29fa2b118528b3808e254b82cfb3932e0b49c0e3
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round20.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e3da873f58556abc6180f4a8a563ba802aaa5284c849a77851f73f9db432eeb
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round5.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f37104a5e65a1f682f52cf825a24c671095db476
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21786aae4d0cd71a441646d83f6a1dbc6b235317f77082997b684305b8c29a46
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round7.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5117ad6a78b695053cfcc478d8528dbcd26de779
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_client_model_round7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57f963e28f484f6b261d664c1383881d9dd12f4c581f80ded30a5b423f232bd1
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_trainer_state.json b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..332138f36d1da0a87e3dfd4cce1d342cb6eb43aa
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/5_trainer_state.json
@@ -0,0 +1,189 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 43,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09302325581395349,
+      "grad_norm": 0.45722439885139465,
+      "learning_rate": 2e-05,
+      "loss": 0.1039,
+      "step": 2
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 3.8990814685821533,
+      "learning_rate": 2e-05,
+      "loss": 0.4251,
+      "step": 4
+    },
+    {
+      "epoch": 0.27906976744186046,
+      "grad_norm": 3.4248876571655273,
+      "learning_rate": 2e-05,
+      "loss": 0.2447,
+      "step": 6
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 4.074819564819336,
+      "learning_rate": 2e-05,
+      "loss": 0.1802,
+      "step": 8
+    },
+    {
+      "epoch": 0.46511627906976744,
+      "grad_norm": 0.28555965423583984,
+      "learning_rate": 2e-05,
+      "loss": 0.0223,
+      "step": 10
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 2.2729122638702393,
+      "learning_rate": 2e-05,
+      "loss": 0.0879,
+      "step": 12
+    },
+    {
+      "epoch": 0.6511627906976745,
+      "grad_norm": 1.4464470148086548,
+      "learning_rate": 2e-05,
+      "loss": 0.0681,
+      "step": 14
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 1.4769107103347778,
+      "learning_rate": 2e-05,
+      "loss": 0.2656,
+      "step": 16
+    },
+    {
+      "epoch": 0.8372093023255814,
+      "grad_norm": 0.6722472906112671,
+      "learning_rate": 2e-05,
+      "loss": 0.0413,
+      "step": 18
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 3.9609217643737793,
+      "learning_rate": 2e-05,
+      "loss": 0.3035,
+      "step": 20
+    },
+    {
+      "epoch": 1.0232558139534884,
+      "grad_norm": 0.3582138419151306,
+      "learning_rate": 2e-05,
+      "loss": 0.1244,
+      "step": 22
+    },
+    {
+      "epoch": 1.1162790697674418,
+      "grad_norm": 0.7322174906730652,
+      "learning_rate": 2e-05,
+      "loss": 0.2526,
+      "step": 24
+    },
+    {
+      "epoch": 1.2093023255813953,
+      "grad_norm": 0.20428186655044556,
+      "learning_rate": 2e-05,
+      "loss": 0.0239,
+      "step": 26
+    },
+    {
+      "epoch": 1.302325581395349,
+      "grad_norm": 0.345355749130249,
+      "learning_rate": 2e-05,
+      "loss": 0.044,
+      "step": 28
+    },
+    {
+      "epoch": 1.3953488372093024,
+      "grad_norm": 1.054936170578003,
+      "learning_rate": 2e-05,
+      "loss": 0.084,
+      "step": 30
+    },
+    {
+      "epoch": 1.4883720930232558,
+      "grad_norm": 4.094634532928467,
+      "learning_rate": 2e-05,
+      "loss": 0.1025,
+      "step": 32
+    },
+    {
+      "epoch": 1.5813953488372092,
+      "grad_norm": 5.273200988769531,
+      "learning_rate": 2e-05,
+      "loss": 0.5042,
+      "step": 34
+    },
+    {
+      "epoch": 1.6744186046511629,
+      "grad_norm": 0.8009881377220154,
+      "learning_rate": 2e-05,
+      "loss": 0.1144,
+      "step": 36
+    },
+    {
+      "epoch": 1.7674418604651163,
+      "grad_norm": 0.6966337561607361,
+      "learning_rate": 2e-05,
+      "loss": 0.1874,
+      "step": 38
+    },
+    {
+      "epoch": 1.8604651162790697,
+      "grad_norm": 2.884883403778076,
+      "learning_rate": 2e-05,
+      "loss": 0.1102,
+      "step": 40
+    },
+    {
+      "epoch": 1.9534883720930232,
+      "grad_norm": 1.31296968460083,
+      "learning_rate": 2e-05,
+      "loss": 0.0504,
+      "step": 42
+    },
+    {
+      "epoch": 2.0,
+      "step": 43,
+      "total_flos": 4774350557282304.0,
+      "train_loss": 0.15653900767481604,
+      "train_runtime": 217.4032,
+      "train_samples_per_second": 0.791,
+      "train_steps_per_second": 0.198
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 43,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4774350557282304.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round10.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9055d93dd5a20e91c00e34007a1928ffd93fce8f
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2d42d214d61d0a917da1b08bd9154134cddc1fc91e1f043baad755e95cceab4
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round12.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3a1fa47badef12c0f3f93346b21434276ea8c030
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84bf10b30840ee671bdd5d3917ebbb978d08a48da0b7f03c9c6b79c79c3240f0
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round15.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round15.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e1d56a3d3ed810c7cf80efac4f7fa25f00df77dc
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round15.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a90bbd20c0e33b1d98200b3e479975e941bcb1fa08adcc9f8513747838be9be5
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round17.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round17.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5cca7911b32448c62c9599f5f5ef487d92ec0200
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round17.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aac8d2f7910003842ce097774317513ba45bbfa971abae855b94a4cd36f54df9
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round2.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c259cbf2e725a2a7a1e57ae771e42012812382a9
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2df2e6dabf5effcee5e2dea203559d9ac93bcf6383a7e34a21ca0f4d64b3eba4
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round20.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round20.pth
new file mode 100644
index 0000000000000000000000000000000000000000..8580a0f57fc3497f673ef3b9c73c0ae51e41584e
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round20.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:805e5c0000ed719cd27756fb1170a41b1b67e445fd46649471326e07c54e1995
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round5.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3f8bfbfe7bd14fc3571088e9b916c07a46e726a6
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4318289d727442e991e6a3a82208da7090ea0e838128caff2844dd0fab08a24
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round7.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..71771486c86278b8ad09d55530d712258bce72b2
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_client_model_round7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6ee375e567f879ef13bcee35187eaedb1d5b3525d6ffbed51fb4ed6fa338a2f
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_trainer_state.json b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..6e1f8d2c2a1160b9ddf454baf5c7f243dd5f2057
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/6_trainer_state.json
@@ -0,0 +1,189 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 43,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09302325581395349,
+      "grad_norm": 0.9547498226165771,
+      "learning_rate": 2e-05,
+      "loss": 0.1891,
+      "step": 2
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 1.5149000883102417,
+      "learning_rate": 2e-05,
+      "loss": 0.2614,
+      "step": 4
+    },
+    {
+      "epoch": 0.27906976744186046,
+      "grad_norm": 1.0263214111328125,
+      "learning_rate": 2e-05,
+      "loss": 0.2237,
+      "step": 6
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 1.6056371927261353,
+      "learning_rate": 2e-05,
+      "loss": 0.2599,
+      "step": 8
+    },
+    {
+      "epoch": 0.46511627906976744,
+      "grad_norm": 0.9209157228469849,
+      "learning_rate": 2e-05,
+      "loss": 0.195,
+      "step": 10
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 2.0802102088928223,
+      "learning_rate": 2e-05,
+      "loss": 0.2057,
+      "step": 12
+    },
+    {
+      "epoch": 0.6511627906976745,
+      "grad_norm": 2.8346526622772217,
+      "learning_rate": 2e-05,
+      "loss": 0.3626,
+      "step": 14
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 2.0434837341308594,
+      "learning_rate": 2e-05,
+      "loss": 0.1595,
+      "step": 16
+    },
+    {
+      "epoch": 0.8372093023255814,
+      "grad_norm": 0.992615282535553,
+      "learning_rate": 2e-05,
+      "loss": 0.1827,
+      "step": 18
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 2.1075265407562256,
+      "learning_rate": 2e-05,
+      "loss": 0.2105,
+      "step": 20
+    },
+    {
+      "epoch": 1.0232558139534884,
+      "grad_norm": 2.0729548931121826,
+      "learning_rate": 2e-05,
+      "loss": 0.3808,
+      "step": 22
+    },
+    {
+      "epoch": 1.1162790697674418,
+      "grad_norm": 1.4767122268676758,
+      "learning_rate": 2e-05,
+      "loss": 0.1218,
+      "step": 24
+    },
+    {
+      "epoch": 1.2093023255813953,
+      "grad_norm": 2.4659721851348877,
+      "learning_rate": 2e-05,
+      "loss": 0.1984,
+      "step": 26
+    },
+    {
+      "epoch": 1.302325581395349,
+      "grad_norm": 2.5538926124572754,
+      "learning_rate": 2e-05,
+      "loss": 0.2617,
+      "step": 28
+    },
+    {
+      "epoch": 1.3953488372093024,
+      "grad_norm": 0.35006552934646606,
+      "learning_rate": 2e-05,
+      "loss": 0.1934,
+      "step": 30
+    },
+    {
+      "epoch": 1.4883720930232558,
+      "grad_norm": 5.821071624755859,
+      "learning_rate": 2e-05,
+      "loss": 0.4144,
+      "step": 32
+    },
+    {
+      "epoch": 1.5813953488372092,
+      "grad_norm": 0.8612791895866394,
+      "learning_rate": 2e-05,
+      "loss": 0.2032,
+      "step": 34
+    },
+    {
+      "epoch": 1.6744186046511629,
+      "grad_norm": 2.9951696395874023,
+      "learning_rate": 2e-05,
+      "loss": 0.2346,
+      "step": 36
+    },
+    {
+      "epoch": 1.7674418604651163,
+      "grad_norm": 0.7142083644866943,
+      "learning_rate": 2e-05,
+      "loss": 0.1884,
+      "step": 38
+    },
+    {
+      "epoch": 1.8604651162790697,
+      "grad_norm": 3.3161582946777344,
+      "learning_rate": 2e-05,
+      "loss": 0.2789,
+      "step": 40
+    },
+    {
+      "epoch": 1.9534883720930232,
+      "grad_norm": 1.4730931520462036,
+      "learning_rate": 2e-05,
+      "loss": 0.1537,
+      "step": 42
+    },
+    {
+      "epoch": 2.0,
+      "step": 43,
+      "total_flos": 9449926933086208.0,
+      "train_loss": 0.2294954255569813,
+      "train_runtime": 245.4222,
+      "train_samples_per_second": 0.701,
+      "train_steps_per_second": 0.175
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 43,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 9449926933086208.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round10.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c7deee2dc667e90502cf785d1aadee3a303385cd
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89183b5ac13038416218f13a28c181d81597894d0065dd12c88cf759ba1d8a0e
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round12.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e3855891cae22ec2fe32c035b156bf1171453838
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70561fb945e1a59ff68851e97141c657a89b3193554442df4f2857c050c6c126
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round15.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round15.pth
new file mode 100644
index 0000000000000000000000000000000000000000..23f9eccb0e7911aabff902d2fe0f7a40a4207fe2
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round15.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87354d8d7b6b1e81ff225f696d9e1bfdd7a9e99fe960bb9d4cfdaa7583ede409
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round17.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round17.pth
new file mode 100644
index 0000000000000000000000000000000000000000..dd60c9394ee47c359a6db9eb89c00ffeb9370cec
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round17.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5067df62a58a5b8615ff58ff4417bfe36018ca2d477da6fe1fc1844c1a6ddbd8
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round2.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2278b7ae37af60389bc5872da6a42be3bffe3fbe
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24d7d73e01ec10575100ac088b532476db2ebb67fae104a1c29c784c2b76aaad
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round20.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round20.pth
new file mode 100644
index 0000000000000000000000000000000000000000..596d42e81dc47e1c8a84df54dc2f12f00859464c
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round20.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e154c8245837fffa9b6f8e471b96b74f25911302cb587036729ca70a05136e2
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round5.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..00a0f79927e216b23b49530bfaba7b01e9f435ae
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16c268790f1966cbada6d9d73bbda335412d9f75640d41a3638c9afc3cade527
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round7.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d6780b0bf066930bb07bd4ccb48dfa4073060e36
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_client_model_round7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f78a4a96cde618f336c0d7405daba91413c2c220af48e062969b8f7ba3ba1250
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_trainer_state.json b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..6eb22f5f4b937ccfdab6cfc57a72823fd09358a3
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/7_trainer_state.json
@@ -0,0 +1,189 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 43,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09302325581395349,
+      "grad_norm": 2.5116467475891113,
+      "learning_rate": 2e-05,
+      "loss": 0.2517,
+      "step": 2
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 1.789315104484558,
+      "learning_rate": 2e-05,
+      "loss": 0.1989,
+      "step": 4
+    },
+    {
+      "epoch": 0.27906976744186046,
+      "grad_norm": 1.3634082078933716,
+      "learning_rate": 2e-05,
+      "loss": 0.2068,
+      "step": 6
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 0.7581627368927002,
+      "learning_rate": 2e-05,
+      "loss": 0.0652,
+      "step": 8
+    },
+    {
+      "epoch": 0.46511627906976744,
+      "grad_norm": 2.145380973815918,
+      "learning_rate": 2e-05,
+      "loss": 0.076,
+      "step": 10
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 0.7811837792396545,
+      "learning_rate": 2e-05,
+      "loss": 0.0858,
+      "step": 12
+    },
+    {
+      "epoch": 0.6511627906976745,
+      "grad_norm": 2.235487461090088,
+      "learning_rate": 2e-05,
+      "loss": 0.0809,
+      "step": 14
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 0.37225276231765747,
+      "learning_rate": 2e-05,
+      "loss": 0.0379,
+      "step": 16
+    },
+    {
+      "epoch": 0.8372093023255814,
+      "grad_norm": 2.32248854637146,
+      "learning_rate": 2e-05,
+      "loss": 0.1777,
+      "step": 18
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 5.746020317077637,
+      "learning_rate": 2e-05,
+      "loss": 0.235,
+      "step": 20
+    },
+    {
+      "epoch": 1.0232558139534884,
+      "grad_norm": 3.4946515560150146,
+      "learning_rate": 2e-05,
+      "loss": 0.1718,
+      "step": 22
+    },
+    {
+      "epoch": 1.1162790697674418,
+      "grad_norm": 3.0465049743652344,
+      "learning_rate": 2e-05,
+      "loss": 0.1924,
+      "step": 24
+    },
+    {
+      "epoch": 1.2093023255813953,
+      "grad_norm": 0.3472822606563568,
+      "learning_rate": 2e-05,
+      "loss": 0.0368,
+      "step": 26
+    },
+    {
+      "epoch": 1.302325581395349,
+      "grad_norm": 5.356133937835693,
+      "learning_rate": 2e-05,
+      "loss": 0.6351,
+      "step": 28
+    },
+    {
+      "epoch": 1.3953488372093024,
+      "grad_norm": 3.610119581222534,
+      "learning_rate": 2e-05,
+      "loss": 0.2331,
+      "step": 30
+    },
+    {
+      "epoch": 1.4883720930232558,
+      "grad_norm": 3.7931079864501953,
+      "learning_rate": 2e-05,
+      "loss": 0.3142,
+      "step": 32
+    },
+    {
+      "epoch": 1.5813953488372092,
+      "grad_norm": 3.1257500648498535,
+      "learning_rate": 2e-05,
+      "loss": 0.2056,
+      "step": 34
+    },
+    {
+      "epoch": 1.6744186046511629,
+      "grad_norm": 3.575251817703247,
+      "learning_rate": 2e-05,
+      "loss": 0.1245,
+      "step": 36
+    },
+    {
+      "epoch": 1.7674418604651163,
+      "grad_norm": 3.165757894515991,
+      "learning_rate": 2e-05,
+      "loss": 0.1002,
+      "step": 38
+    },
+    {
+      "epoch": 1.8604651162790697,
+      "grad_norm": 2.5444176197052,
+      "learning_rate": 2e-05,
+      "loss": 0.388,
+      "step": 40
+    },
+    {
+      "epoch": 1.9534883720930232,
+      "grad_norm": 3.0468590259552,
+      "learning_rate": 2e-05,
+      "loss": 0.462,
+      "step": 42
+    },
+    {
+      "epoch": 2.0,
+      "step": 43,
+      "total_flos": 4941383202766848.0,
+      "train_loss": 0.20255567861157794,
+      "train_runtime": 216.7983,
+      "train_samples_per_second": 0.793,
+      "train_steps_per_second": 0.198
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 43,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4941383202766848.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round10.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round10.pth
new file mode 100644
index 0000000000000000000000000000000000000000..41b7de0f4e242931f9e6d52dfa7efe7afcd331c3
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round10.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:862b15aaf682632e34b3ce85d02934ae738a016acf4ae52845355c51eb77c710
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round12.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round12.pth
new file mode 100644
index 0000000000000000000000000000000000000000..988a5f515de3aec37a0ce2405e43a3f6f9063b11
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round12.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56f080e7bedecda9b0fb967e42b9929362e0496267dc033546a34ec8fac5147d
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round15.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round15.pth
new file mode 100644
index 0000000000000000000000000000000000000000..29fd9644c74b8ebd0ffcc3280bcf8f326d1e3416
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round15.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b7e435469e3477b59c28dd09888e5a1fe6f97405da68367162470297513eb7f
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round17.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round17.pth
new file mode 100644
index 0000000000000000000000000000000000000000..23a1e238163a9cb2615a294c3eae0a0d24b17959
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round17.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e5208d9fec17b0a70ca800760fd264a45fc8f267eb5775095d500b40c3d17d5
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round2.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..0c41c348dcc778dc617d75aa02a7035dfe83476f
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4042a11cfc3af5dddc98db063be144b1fcaf8b1d96479014badcabee59bf7e3
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round20.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round20.pth
new file mode 100644
index 0000000000000000000000000000000000000000..460befbd11506261c610aabddfd0d94672b683d1
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round20.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5088eda43240d4843056b24d608e970f442eca0d3af7eb8936d9ea6d3fa7d8e2
+size 1167513110
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round5.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..fb13e4a9eeca72536716af6ab6107a804495314b
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f1a1d9483e00dcfef3e5985c41a255ff431217b8eae7048443427acf4dcd27b
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round7.pth b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round7.pth
new file mode 100644
index 0000000000000000000000000000000000000000..09b34ce72b6aaa52b527a36cc85bc983387439ca
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_client_model_round7.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bffd6d9d481df3d6bb8ae0bb6ec4ff9e77f2cbd23d085e04d55ac2186aaf477a
+size 1167511866
diff --git a/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_trainer_state.json b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..bfb0552947ce462ba3cab2446131e2bcee822a68
--- /dev/null
+++ b/client_states_feddat_bs4_saveoptim_lr2e-5_sc310_4tasks_5rounds_fixitr43_T0125_decay099/8_trainer_state.json
@@ -0,0 +1,189 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 43,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09302325581395349,
+      "grad_norm": 2.1495094299316406,
+      "learning_rate": 2e-05,
+      "loss": 0.287,
+      "step": 2
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 2.306636095046997,
+      "learning_rate": 2e-05,
+      "loss": 0.1049,
+      "step": 4
+    },
+    {
+      "epoch": 0.27906976744186046,
+      "grad_norm": 1.7588932514190674,
+      "learning_rate": 2e-05,
+      "loss": 0.1201,
+      "step": 6
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 2.4194767475128174,
+      "learning_rate": 2e-05,
+      "loss": 0.3482,
+      "step": 8
+    },
+    {
+      "epoch": 0.46511627906976744,
+      "grad_norm": 0.4276074469089508,
+      "learning_rate": 2e-05,
+      "loss": 0.058,
+      "step": 10
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 1.4213718175888062,
+      "learning_rate": 2e-05,
+      "loss": 0.2725,
+      "step": 12
+    },
+    {
+      "epoch": 0.6511627906976745,
+      "grad_norm": 0.5836634039878845,
+      "learning_rate": 2e-05,
+      "loss": 0.0932,
+      "step": 14
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 4.312954902648926,
+      "learning_rate": 2e-05,
+      "loss": 0.231,
+      "step": 16
+    },
+    {
+      "epoch": 0.8372093023255814,
+      "grad_norm": 0.3371574878692627,
+      "learning_rate": 2e-05,
+      "loss": 0.189,
+      "step": 18
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 4.521411895751953,
+      "learning_rate": 2e-05,
+      "loss": 0.4993,
+      "step": 20
+    },
+    {
+      "epoch": 1.0232558139534884,
+      "grad_norm": 1.1006470918655396,
+      "learning_rate": 2e-05,
+      "loss": 0.0529,
+      "step": 22
+    },
+    {
+      "epoch": 1.1162790697674418,
+      "grad_norm": 1.5914112329483032,
+      "learning_rate": 2e-05,
+      "loss": 0.084,
+      "step": 24
+    },
+    {
+      "epoch": 1.2093023255813953,
+      "grad_norm": 0.4478868544101715,
+      "learning_rate": 2e-05,
+      "loss": 0.0506,
+      "step": 26
+    },
+    {
+      "epoch": 1.302325581395349,
+      "grad_norm": 0.35286229848861694,
+      "learning_rate": 2e-05,
+      "loss": 0.0284,
+      "step": 28
+    },
+    {
+      "epoch": 1.3953488372093024,
+      "grad_norm": 1.8138010501861572,
+      "learning_rate": 2e-05,
+      "loss": 0.1689,
+      "step": 30
+    },
+    {
+      "epoch": 1.4883720930232558,
+      "grad_norm": 1.7487231492996216,
+      "learning_rate": 2e-05,
+      "loss": 0.4923,
+      "step": 32
+    },
+    {
+      "epoch": 1.5813953488372092,
+      "grad_norm": 2.4762122631073,
+      "learning_rate": 2e-05,
+      "loss": 0.1159,
+      "step": 34
+    },
+    {
+      "epoch": 1.6744186046511629,
+      "grad_norm": 1.0733447074890137,
+      "learning_rate": 2e-05,
+      "loss": 0.3211,
+      "step": 36
+    },
+    {
+      "epoch": 1.7674418604651163,
+      "grad_norm": 0.4013538956642151,
+      "learning_rate": 2e-05,
+      "loss": 0.0307,
+      "step": 38
+    },
+    {
+      "epoch": 1.8604651162790697,
+      "grad_norm": 5.619275093078613,
+      "learning_rate": 2e-05,
+      "loss": 0.4352,
+      "step": 40
+    },
+    {
+      "epoch": 1.9534883720930232,
+      "grad_norm": 0.5066792368888855,
+      "learning_rate": 2e-05,
+      "loss": 0.3619,
+      "step": 42
+    },
+    {
+      "epoch": 2.0,
+      "step": 43,
+      "total_flos": 4817714786861056.0,
+      "train_loss": 0.2102257262828738,
+      "train_runtime": 215.7907,
+      "train_samples_per_second": 0.797,
+      "train_steps_per_second": 0.199
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 43,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4817714786861056.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}