Training in progress, step 650, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:085286a0e6090c82fcc13eea53f92c3af07a3dd530dc3c67c22de5abfe705112
+oid sha256:f082218d4450a6fd5cba15403298e1e103fd66cae7d3efcef2dce424d9043bc9
 size 29034840

last-checkpoint/global_step650/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81db18808ef854cee5581ca4f17863358b719281e053752be4de4707e39e43c0
+size 43429616

last-checkpoint/global_step650/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:e857fd390a161f2feb8f5e6a6760418412e53d89c4b054f94b6a141334063763
+size 43429616

last-checkpoint/global_step650/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:fedb43a5f62ade4a23cb9ee6ed4887f51051ae484e0a31b7e386eaf51cd04ee2
+size 43429616

last-checkpoint/global_step650/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:c0c36d41d30e3fabe68d4d64ac13a18e9b9378b59d3facbd4e9ec1d8d22d58c1
+size 43429616

last-checkpoint/global_step650/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f9450a0ced61d6c33a66cdf1b2511022f0ec24416b17e1f16ed49dc98ff62dd
+size 637299

last-checkpoint/global_step650/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d01b9093c4ff061c6453535b14beda5a9d96a531472a58effac685152d63de95
+size 637171

last-checkpoint/global_step650/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a3bfb0ef98b44969e473f6d5c11a92b5a607c48dc2a278b5d72591611de4079
+size 637171

last-checkpoint/global_step650/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a41c42f45b9260afbaae285e2a90de1413db2108ea931952bc69942ea6f19e1e
+size 637171

last-checkpoint/latest

@@ -1 +1 @@
-global_step600
+global_step650

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a81e3916b1392c4c49afb171dee5415c15f5a5a5af8749b28195fcfa0596699c
+oid sha256:8044e4c53158c210a17648ba8f2dc2d25a25bbfc55f686015542618eb652a33e
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a781038dd714b87b8adb1aac8dbc8217ceb607428a992133954ad522365236e
+oid sha256:4cd85d7fa425e7888c973f1c2985ac15ca21b5e6171fe140a401c2bc75ca46ff
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9446c3db15f382a5546f13622787fc99392a5e0bc8a9ca2da1838de7ab621a37
+oid sha256:d7915667371a58f1598639e0d1c20a0c59c783c14580cd040a6631eb4ea2311e
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f11e7a6b3faa884fc23044e3772ff9dd72c257f02e121665061e2a03d518bd9
+oid sha256:35dd78929ad7f0fbf37fdb1284e8edf0424350f6e6ce1cd5a3ee78979af3d3cb
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b76b388bede074656df32b92902ac42b965557bfee0c930366af07d8382b1b4a
+oid sha256:8cd94ecf5c982ee0e060d3e07a575ce03dc3b0f289b5e32a1f65d3b6366a8a0e
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": 0.0033526704646646976,
   "best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-550",
-  "epoch": 0.3090394025238218,
+  "epoch": 0.3347926860674736,
   "eval_steps": 50,
-  "global_step": 600,
+  "global_step": 650,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1075,11 +1075,100 @@
       "eval_steps_per_second": 0.795,
       "num_input_tokens_seen": 5990400,
       "step": 600
+    },
+    {
+      "epoch": 0.311614730878187,
+      "grad_norm": 0.06328082378687645,
+      "learning_rate": 9.55911532374151e-05,
+      "loss": 0.0207,
+      "num_input_tokens_seen": 6040320,
+      "step": 605
+    },
+    {
+      "epoch": 0.31419005923255217,
+      "grad_norm": 1.846961607001629,
+      "learning_rate": 9.549077815930636e-05,
+      "loss": 0.023,
+      "num_input_tokens_seen": 6090240,
+      "step": 610
+    },
+    {
+      "epoch": 0.31676538758691736,
+      "grad_norm": 0.07122251919235507,
+      "learning_rate": 9.538932721758474e-05,
+      "loss": 0.017,
+      "num_input_tokens_seen": 6140160,
+      "step": 615
+    },
+    {
+      "epoch": 0.3193407159412825,
+      "grad_norm": 1.9564639611342651,
+      "learning_rate": 9.528680281157999e-05,
+      "loss": 0.009,
+      "num_input_tokens_seen": 6190080,
+      "step": 620
+    },
+    {
+      "epoch": 0.3219160442956477,
+      "grad_norm": 0.43628623054662674,
+      "learning_rate": 9.518320736600943e-05,
+      "loss": 0.0421,
+      "num_input_tokens_seen": 6240000,
+      "step": 625
+    },
+    {
+      "epoch": 0.3244913726500129,
+      "grad_norm": 0.7626652394047067,
+      "learning_rate": 9.507854333092063e-05,
+      "loss": 0.0324,
+      "num_input_tokens_seen": 6289920,
+      "step": 630
+    },
+    {
+      "epoch": 0.32706670100437807,
+      "grad_norm": 0.5823990007842583,
+      "learning_rate": 9.497281318163346e-05,
+      "loss": 0.0139,
+      "num_input_tokens_seen": 6339840,
+      "step": 635
+    },
+    {
+      "epoch": 0.32964202935874326,
+      "grad_norm": 1.6726620150276597,
+      "learning_rate": 9.486601941868154e-05,
+      "loss": 0.0375,
+      "num_input_tokens_seen": 6389760,
+      "step": 640
+    },
+    {
+      "epoch": 0.3322173577131084,
+      "grad_norm": 0.08456122780891835,
+      "learning_rate": 9.475816456775313e-05,
+      "loss": 0.0618,
+      "num_input_tokens_seen": 6439680,
+      "step": 645
+    },
+    {
+      "epoch": 0.3347926860674736,
+      "grad_norm": 0.3287256358007373,
+      "learning_rate": 9.464925117963133e-05,
+      "loss": 0.0327,
+      "num_input_tokens_seen": 6489600,
+      "step": 650
+    },
+    {
+      "epoch": 0.3347926860674736,
+      "eval_loss": 0.02095886506140232,
+      "eval_runtime": 19.1993,
+      "eval_samples_per_second": 3.125,
+      "eval_steps_per_second": 0.781,
+      "num_input_tokens_seen": 6489600,
+      "step": 650
     }
   ],
   "logging_steps": 5,
   "max_steps": 3400,
-  "num_input_tokens_seen": 5990400,
+  "num_input_tokens_seen": 6489600,
   "num_train_epochs": 2,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -1094,7 +1183,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 395224583241728.0,
+  "total_flos": 428164562812928.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null