Training in progress, step 950, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84586103a211c06663bd404e55c4d764d0a39628f772147e29017239dc0ff834
+oid sha256:ed912ae9f29651ceebc41713ab5b91988158a52f69ac82baf6102389d31521a7
 size 29034840

last-checkpoint/global_step950/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b809ff8945c89100055098c0e58e465735de8a9f55ea94fa5fcdd14223305ef
+size 43429616

last-checkpoint/global_step950/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:d260f3afadca21e3a4fdba818360a0d7ec292faf80eef232e413b07af69ffc10
+size 43429616

last-checkpoint/global_step950/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:e642dbc5359877c528369fb0e55fd945f101471de9fa1da28289e1ea1879197e
+size 43429616

last-checkpoint/global_step950/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93d3075ceb4a740b9de02bd73f8b6e74568b2448b5e5d20bdfc72aefd6a43ab0
+size 43429616

last-checkpoint/global_step950/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de8d2452415f67f6cb314134176f636d47ded9477de6e168c2925a95f1e50399
+size 637299

last-checkpoint/global_step950/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59a0c15d11bab9ab36ba888fa09e8a77ddc15c19181e193c722b9f734d07accd
+size 637171

last-checkpoint/global_step950/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8010ba1747feb31117a6b69f86638d1fc0741ad088c9024682efeba498962221
+size 637171

last-checkpoint/global_step950/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39c9dd240d6e20c6b0f5d0f32f43209c894bb5b7adbab13d992cf542fcdcdb1c
+size 637171

last-checkpoint/latest

@@ -1 +1 @@
-global_step900
+global_step950

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8f22ced19e790cc864cefe3b7c711d9ae631c44f95d42fb4829688cc3de0153
+oid sha256:7f2439da621f14c22b4f733e91bfc9de6b506d28d7b8d6f3eaca2e0b4f24c078
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e0407513eba77d34cbf3adf0e59a58bd80716f4f00f414854253637e82be43d
+oid sha256:c9e3fb386557f376b8946af5b8c91f9418f374dddb2ad9da4868b1ef16778c32
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6060636c023258ce9b965e244b8a58b4c99d5784dde4405b39737550ef50cd4f
+oid sha256:dc7774d06045635bece9e960378fdc6913bf7bbbc903444cc570d1ca6ac25645
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c24ccdfdcde39cb2265c82c50c36ffdfcc670f757aba4bcf4bb0fdc6d1373c4c
+oid sha256:d98c54a80a914fecf43d06ea81432499f46e70664f1d04651bf339163e30fa9e
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:feb4015894f59edc29c71bc4938b5d4ab98daad34a38a3d387b308b3b1d4b280
+oid sha256:b9f224baf5bd2044314606c1d88f84cce32f1b37c43c15835b14e72f6a72a4fc
 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": 0.4635591037857327,
+  "epoch": 0.4893123873293845,
   "eval_steps": 50,
-  "global_step": 900,
+  "global_step": 950,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1609,11 +1609,100 @@
       "eval_steps_per_second": 0.778,
       "num_input_tokens_seen": 8985600,
       "step": 900
+    },
+    {
+      "epoch": 0.46613443214009787,
+      "grad_norm": 2.090931980208863,
+      "learning_rate": 8.775851227988656e-05,
+      "loss": 0.0318,
+      "num_input_tokens_seen": 9035520,
+      "step": 905
+    },
+    {
+      "epoch": 0.46870976049446306,
+      "grad_norm": 1.7373295104389102,
+      "learning_rate": 8.759866980070963e-05,
+      "loss": 0.0635,
+      "num_input_tokens_seen": 9085440,
+      "step": 910
+    },
+    {
+      "epoch": 0.47128508884882825,
+      "grad_norm": 0.1557668148789241,
+      "learning_rate": 8.743793810744654e-05,
+      "loss": 0.0035,
+      "num_input_tokens_seen": 9135360,
+      "step": 915
+    },
+    {
+      "epoch": 0.4738604172031934,
+      "grad_norm": 0.5241949871459053,
+      "learning_rate": 8.727632100142551e-05,
+      "loss": 0.0047,
+      "num_input_tokens_seen": 9185280,
+      "step": 920
+    },
+    {
+      "epoch": 0.4764357455575586,
+      "grad_norm": 2.329131853129593,
+      "learning_rate": 8.711382230491493e-05,
+      "loss": 0.0194,
+      "num_input_tokens_seen": 9235200,
+      "step": 925
+    },
+    {
+      "epoch": 0.47901107391192377,
+      "grad_norm": 1.0158880737206768,
+      "learning_rate": 8.695044586103296e-05,
+      "loss": 0.0234,
+      "num_input_tokens_seen": 9285120,
+      "step": 930
+    },
+    {
+      "epoch": 0.48158640226628896,
+      "grad_norm": 0.016729230547496875,
+      "learning_rate": 8.678619553365659e-05,
+      "loss": 0.0253,
+      "num_input_tokens_seen": 9335040,
+      "step": 935
+    },
+    {
+      "epoch": 0.48416173062065415,
+      "grad_norm": 2.937987778998638,
+      "learning_rate": 8.662107520733027e-05,
+      "loss": 0.0191,
+      "num_input_tokens_seen": 9384960,
+      "step": 940
+    },
+    {
+      "epoch": 0.4867370589750193,
+      "grad_norm": 0.20640046184009653,
+      "learning_rate": 8.64550887871741e-05,
+      "loss": 0.0469,
+      "num_input_tokens_seen": 9434880,
+      "step": 945
+    },
+    {
+      "epoch": 0.4893123873293845,
+      "grad_norm": 1.9952606902760353,
+      "learning_rate": 8.628824019879137e-05,
+      "loss": 0.0415,
+      "num_input_tokens_seen": 9484800,
+      "step": 950
+    },
+    {
+      "epoch": 0.4893123873293845,
+      "eval_loss": 0.007216573692858219,
+      "eval_runtime": 19.2716,
+      "eval_samples_per_second": 3.113,
+      "eval_steps_per_second": 0.778,
+      "num_input_tokens_seen": 9484800,
+      "step": 950
     }
   ],
   "logging_steps": 5,
   "max_steps": 3400,
-  "num_input_tokens_seen": 8985600,
+  "num_input_tokens_seen": 9484800,
   "num_train_epochs": 2,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -1628,7 +1717,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 592864460668928.0,
+  "total_flos": 625804440240128.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null