Training in progress, step 100, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b0fb13d10a5153c42be30667ec413318749436b697a6a62d3dd8e7e8cd181234
+oid sha256:54617ace9be9da2e4e57ae55a0f5329011112b5c8868c9fffa532ea69e4c2d98
 size 29034840

last-checkpoint/global_step100/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ddc2817afdc9225f515b78a9eb8a53a0f3315e3cdd9dea905c90813f2839bdd
+size 43429616

last-checkpoint/global_step100/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1f3250f9636418835e6fb0cb6591fe5d43889ab3ec1d5034336bd3337377ae7
+size 43429616

last-checkpoint/global_step100/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4cce653c94d71ed469a09c7d917fa914f90a11a71b34f8eeffd05e35fc4c4515
+size 43429616

last-checkpoint/global_step100/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4098ce82fff11005a17eecd3482bcc1b0d125c2c4454f4f2c5da52a26a9e46e4
+size 43429616

last-checkpoint/global_step100/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5d10e19daf23d9a5519105cd8f02b133fd832afdbceaae32bd68ecfb18273f2
+size 637299

last-checkpoint/global_step100/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e70d7ddedc58dfdc0ec49a115d1b1e0af3fb290d2785ce1448b9276a2c88d36
+size 637171

last-checkpoint/global_step100/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ea9021c0781ce4f2bf1279fa5cd2b7bb74cadac2cd292ff41e5bda0c40f8f90
+size 637171

last-checkpoint/global_step100/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db256fcff0519e97f39b24fc7b31436bff588b11e7bcc1b1e3b7bba4345a7d67
+size 637171

last-checkpoint/latest

@@ -1 +1 @@
-global_step50
+global_step100

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbe0d720c4c75a6a04213fa3b64bacbe794718a53e2b56ebb67a1a795014dfad
+oid sha256:308f94f9a5c24e1bad5c393d56ae7af7782600f4e791d9c6ac35b22fff2105b6
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72452d3138d0ca2ff89429e3294a834ae7a68e8596fc757735ca56ae52509d57
+oid sha256:b056f3c23cb32dc77a2ec9e7651e0b64e4440e21f0fdf969b86bfc56a1cbdf06
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f36e306fb8ebcf53a167bfd6c9af74db410a269ada1e619e3e816f5269543b9d
+oid sha256:f3f8a05714bc528f4885a2816181652f2303b3e8150f89b56aaee6bec56aa520
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb47ce0c6f815a6f8302b0e3819b4c2315ca71dae3138d97fdceb765cdd0a039
+oid sha256:4f755bd3c330281961e5c03af9d10ce8c1e1678619d384f6f1fd5fd7dce2ff50
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ab82860e8f4753962498aa91f545ddf5d94711cc852b62536116736b2908eac
+oid sha256:4e451a9e086b06d7c667be8442b2115f5c088953bade0b625e61f2ce5c7fd404
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.33834776282310486,
-  "best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-50",
-  "epoch": 0.025753283543651816,
+  "best_metric": 0.18663176894187927,
+  "best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-100",
+  "epoch": 0.05150656708730363,
   "eval_steps": 50,
-  "global_step": 50,
+  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -96,11 +96,100 @@
       "eval_steps_per_second": 0.316,
       "num_input_tokens_seen": 499200,
       "step": 50
+    },
+    {
+      "epoch": 0.028328611898016998,
+      "grad_norm": 1.4016556961184263,
+      "learning_rate": 3.235294117647059e-05,
+      "loss": 0.3182,
+      "num_input_tokens_seen": 549120,
+      "step": 55
+    },
+    {
+      "epoch": 0.03090394025238218,
+      "grad_norm": 0.6437613769459606,
+      "learning_rate": 3.529411764705883e-05,
+      "loss": 0.3294,
+      "num_input_tokens_seen": 599040,
+      "step": 60
+    },
+    {
+      "epoch": 0.03347926860674736,
+      "grad_norm": 0.7389008951321312,
+      "learning_rate": 3.8235294117647055e-05,
+      "loss": 0.3097,
+      "num_input_tokens_seen": 648960,
+      "step": 65
+    },
+    {
+      "epoch": 0.036054596961112545,
+      "grad_norm": 0.771553860801019,
+      "learning_rate": 4.11764705882353e-05,
+      "loss": 0.3008,
+      "num_input_tokens_seen": 698880,
+      "step": 70
+    },
+    {
+      "epoch": 0.03862992531547772,
+      "grad_norm": 0.6965369148334918,
+      "learning_rate": 4.411764705882353e-05,
+      "loss": 0.3278,
+      "num_input_tokens_seen": 748800,
+      "step": 75
+    },
+    {
+      "epoch": 0.04120525366984291,
+      "grad_norm": 0.912943461315541,
+      "learning_rate": 4.705882352941177e-05,
+      "loss": 0.3074,
+      "num_input_tokens_seen": 798720,
+      "step": 80
+    },
+    {
+      "epoch": 0.043780582024208085,
+      "grad_norm": 0.8407481737577445,
+      "learning_rate": 5e-05,
+      "loss": 0.3423,
+      "num_input_tokens_seen": 848640,
+      "step": 85
+    },
+    {
+      "epoch": 0.04635591037857327,
+      "grad_norm": 0.9112879058417015,
+      "learning_rate": 5.294117647058824e-05,
+      "loss": 0.3008,
+      "num_input_tokens_seen": 898560,
+      "step": 90
+    },
+    {
+      "epoch": 0.04893123873293845,
+      "grad_norm": 2.391489040464162,
+      "learning_rate": 5.588235294117647e-05,
+      "loss": 0.2815,
+      "num_input_tokens_seen": 948480,
+      "step": 95
+    },
+    {
+      "epoch": 0.05150656708730363,
+      "grad_norm": 2.155211791607199,
+      "learning_rate": 5.882352941176471e-05,
+      "loss": 0.2274,
+      "num_input_tokens_seen": 998400,
+      "step": 100
+    },
+    {
+      "epoch": 0.05150656708730363,
+      "eval_loss": 0.18663176894187927,
+      "eval_runtime": 18.9199,
+      "eval_samples_per_second": 3.171,
+      "eval_steps_per_second": 0.793,
+      "num_input_tokens_seen": 998400,
+      "step": 100
     }
   ],
   "logging_steps": 5,
   "max_steps": 3400,
-  "num_input_tokens_seen": 499200,
+  "num_input_tokens_seen": 998400,
   "num_train_epochs": 2,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -115,7 +204,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 32884807958528.0,
+  "total_flos": 65824787529728.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null