Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,12 +20,12 @@
   "rank_pattern": {},
   "revision": "unsloth",
   "target_modules": [
-    "k_proj",
+    "down_proj",
+    "o_proj",
     "v_proj",
     "up_proj",
-    "down_proj",
+    "k_proj",
     "q_proj",
-    "o_proj",
     "gate_proj"
   ],
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53555ee8a4eb110c4bdce3fd50ed70e1c37d2228502979597728308e7dbc4c8a
+oid sha256:f79c51cd876c8b5b6f891bdff8fd168dd531ab20717f8389982574d19f0dbea0
 size 1912664024

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1645b3ce8c7b797e757add674024ac2f3c2bdadc80e8a598a873bcc34e1ba413
+oid sha256:2bf021cd132f16efcbc21f9c136335d5522103e91e16b4f5562176a75367b8c4
 size 958697364

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:caf99753320d5a922d21785c9acff8aad90fb47892dad1f221c94717bf3a04b1
+oid sha256:376fa8cdca401e75879a433c4cf57e5c31b5973d7635f7e690ffeb05b513101b
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:21226719b9e25fe4304ec632d533e056eb568a7272feb0ef7163c162da3912fa
+oid sha256:d4039ca40b296c95519ba3582af20e2ac85fa629c760519260b975e4ffc9aa9b
 size 1064

last-checkpoint/tokenizer_config.json

@@ -122,7 +122,7 @@
   "legacy": false,
   "model_max_length": 4096,
   "pad_token": "<|placeholder6|>",
-  "padding_side": "right",
+  "padding_side": "left",
   "sp_model_kwargs": {},
   "tokenizer_class": "LlamaTokenizer",
   "unk_token": "<unk>",

last-checkpoint/trainer_state.json

@@ -1,170 +1,50 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.012270691453463402,
+  "epoch": 0.0030676728633658506,
   "eval_steps": 20,
-  "global_step": 200,
+  "global_step": 50,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.0012270691453463403,
-      "grad_norm": 0.05003603920340538,
-      "learning_rate": 0.00019993864026834657,
-      "loss": 2.6971,
+      "grad_norm": NaN,
+      "learning_rate": 0.0,
+      "loss": 3.3245,
       "step": 20
     },
     {
       "epoch": 0.0012270691453463403,
-      "eval_loss": 2.296663284301758,
-      "eval_runtime": 26.0065,
-      "eval_samples_per_second": 3.845,
-      "eval_steps_per_second": 0.5,
+      "eval_loss": 1.7450172901153564,
+      "eval_runtime": 24.6271,
+      "eval_samples_per_second": 4.061,
+      "eval_steps_per_second": 0.528,
       "step": 20
     },
     {
       "epoch": 0.0024541382906926807,
-      "grad_norm": 0.07180392742156982,
-      "learning_rate": 0.00019985682729280866,
-      "loss": 2.202,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0001999263532588683,
+      "loss": 3.5708,
       "step": 40
     },
     {
       "epoch": 0.0024541382906926807,
-      "eval_loss": 2.067845582962036,
-      "eval_runtime": 26.2005,
-      "eval_samples_per_second": 3.817,
-      "eval_steps_per_second": 0.496,
+      "eval_loss": 1.7450109720230103,
+      "eval_runtime": 26.043,
+      "eval_samples_per_second": 3.84,
+      "eval_steps_per_second": 0.499,
       "step": 40
-    },
-    {
-      "epoch": 0.003681207436039021,
-      "grad_norm": 0.08042994141578674,
-      "learning_rate": 0.00019977501431727072,
-      "loss": 2.0527,
-      "step": 60
-    },
-    {
-      "epoch": 0.003681207436039021,
-      "eval_loss": 1.9334843158721924,
-      "eval_runtime": 25.7281,
-      "eval_samples_per_second": 3.887,
-      "eval_steps_per_second": 0.505,
-      "step": 60
-    },
-    {
-      "epoch": 0.004908276581385361,
-      "grad_norm": 0.08663463592529297,
-      "learning_rate": 0.0001996932013417328,
-      "loss": 1.9392,
-      "step": 80
-    },
-    {
-      "epoch": 0.004908276581385361,
-      "eval_loss": 1.8688364028930664,
-      "eval_runtime": 26.145,
-      "eval_samples_per_second": 3.825,
-      "eval_steps_per_second": 0.497,
-      "step": 80
-    },
-    {
-      "epoch": 0.006135345726731701,
-      "grad_norm": 0.08462908118963242,
-      "learning_rate": 0.00019961138836619487,
-      "loss": 1.8772,
-      "step": 100
-    },
-    {
-      "epoch": 0.006135345726731701,
-      "eval_loss": 1.8178932666778564,
-      "eval_runtime": 26.3004,
-      "eval_samples_per_second": 3.802,
-      "eval_steps_per_second": 0.494,
-      "step": 100
-    },
-    {
-      "epoch": 0.007362414872078042,
-      "grad_norm": 0.11603377014398575,
-      "learning_rate": 0.000199529575390657,
-      "loss": 1.7827,
-      "step": 120
-    },
-    {
-      "epoch": 0.007362414872078042,
-      "eval_loss": 1.773740291595459,
-      "eval_runtime": 26.4277,
-      "eval_samples_per_second": 3.784,
-      "eval_steps_per_second": 0.492,
-      "step": 120
-    },
-    {
-      "epoch": 0.008589484017424381,
-      "grad_norm": 0.10753527283668518,
-      "learning_rate": 0.00019944776241511905,
-      "loss": 1.8118,
-      "step": 140
-    },
-    {
-      "epoch": 0.008589484017424381,
-      "eval_loss": 1.7376396656036377,
-      "eval_runtime": 26.2097,
-      "eval_samples_per_second": 3.815,
-      "eval_steps_per_second": 0.496,
-      "step": 140
-    },
-    {
-      "epoch": 0.009816553162770723,
-      "grad_norm": 0.09732338786125183,
-      "learning_rate": 0.00019936594943958114,
-      "loss": 1.7397,
-      "step": 160
-    },
-    {
-      "epoch": 0.009816553162770723,
-      "eval_loss": 1.7010047435760498,
-      "eval_runtime": 25.8266,
-      "eval_samples_per_second": 3.872,
-      "eval_steps_per_second": 0.503,
-      "step": 160
-    },
-    {
-      "epoch": 0.011043622308117063,
-      "grad_norm": 0.11193066835403442,
-      "learning_rate": 0.0001992841364640432,
-      "loss": 1.7868,
-      "step": 180
-    },
-    {
-      "epoch": 0.011043622308117063,
-      "eval_loss": 1.6764620542526245,
-      "eval_runtime": 26.1757,
-      "eval_samples_per_second": 3.82,
-      "eval_steps_per_second": 0.497,
-      "step": 180
-    },
-    {
-      "epoch": 0.012270691453463402,
-      "grad_norm": 0.10227545350790024,
-      "learning_rate": 0.0001992023234885053,
-      "loss": 1.7799,
-      "step": 200
-    },
-    {
-      "epoch": 0.012270691453463402,
-      "eval_loss": 1.656005859375,
-      "eval_runtime": 26.2305,
-      "eval_samples_per_second": 3.812,
-      "eval_steps_per_second": 0.496,
-      "step": 200
     }
   ],
   "logging_steps": 20,
-  "max_steps": 48897,
+  "max_steps": 16299,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 1,
   "save_steps": 50,
-  "total_flos": 1.41326994253824e+16,
+  "total_flos": 3253280243097600.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e05a5a439e5cc5d47d8ef9a0303b090d4a155ad12a03f09a5d603bb3e1e1d88
+oid sha256:564d5c4c71e774ca6ed6b6056431bf55581adabc7617b07cefd97a1c116e6504
 size 4920