Upload task output 1

adapter_config.json

@@ -29,13 +29,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "q_proj",
+    "v_proj",
     "down_proj",
-    "up_proj",
     "k_proj",
-    "gate_proj",
-    "q_proj",
     "o_proj",
-    "v_proj"
+    "up_proj",
+    "gate_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a51e4d58b1e4af75220adfa4b19eb8e54a528c771c7995176479f54a4662aee
+oid sha256:38704008a55135f8c68ad0c90eaa5bfdebc06173357d68ab8ccc64d7e5c514b3
 size 957942768

loss.txt

@@ -1 +1 @@
-4,-0.57375
+9,no_eval

trainer_state.json

@@ -2,49 +2,50 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
+  "epoch": 0.011523687580025609,
   "eval_steps": 500,
-  "global_step": 4,
+  "global_step": 9,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.0,
-      "eval_clip_ratio/high_max": 0.0,
-      "eval_clip_ratio/high_mean": 0.0,
-      "eval_clip_ratio/low_mean": 0.0,
-      "eval_clip_ratio/low_min": 0.0,
-      "eval_clip_ratio/region_mean": 0.0,
-      "eval_completions/clipped_ratio": 0.0,
-      "eval_completions/max_length": 2.98,
-      "eval_completions/max_terminated_length": 2.98,
-      "eval_completions/mean_length": 2.32,
-      "eval_completions/mean_terminated_length": 2.32,
-      "eval_completions/min_length": 2.0,
-      "eval_completions/min_terminated_length": 2.0,
-      "eval_entropy": 0.6514365351200104,
-      "eval_frac_reward_zero_std": 0.44,
-      "eval_loss": 0.015936430543661118,
-      "eval_num_tokens": 439873.0,
-      "eval_reward": 0.57375,
-      "eval_reward_std": 0.48721748173236845,
-      "eval_rewards/env_game_reward/mean": 0.57375,
-      "eval_rewards/env_game_reward/std": 0.48721749722957614,
-      "eval_runtime": 71.8753,
-      "eval_samples_per_second": 2.783,
-      "eval_sampling/importance_sampling_ratio/max": 1.3798925983905792,
-      "eval_sampling/importance_sampling_ratio/mean": 0.9876983833312988,
-      "eval_sampling/importance_sampling_ratio/min": 0.7033022457361221,
-      "eval_sampling/sampling_logp_difference/max": 0.4170952755212784,
-      "eval_sampling/sampling_logp_difference/mean": 0.09031737592071294,
-      "eval_steps_per_second": 0.348,
-      "step": 4
+      "clip_ratio/high_max": 0.01083222816232592,
+      "clip_ratio/high_mean": 0.007730815443210304,
+      "clip_ratio/low_mean": 0.008320118731353432,
+      "clip_ratio/low_min": 0.004644901840947568,
+      "clip_ratio/region_mean": 0.016050934419035913,
+      "completions/clipped_ratio": 0.0,
+      "completions/max_length": 374.6,
+      "completions/max_terminated_length": 374.6,
+      "completions/mean_length": 288.41796875,
+      "completions/mean_terminated_length": 288.41796875,
+      "completions/min_length": 145.2,
+      "completions/min_terminated_length": 145.2,
+      "entropy": 0.35683331191539763,
+      "epoch": 0.006402048655569782,
+      "frac_reward_zero_std": 0.4546875,
+      "grad_norm": 0.04971550032496452,
+      "kl": 0.011301717348396778,
+      "learning_rate": 1.137216e-06,
+      "loss": 0.0003326917765662074,
+      "num_tokens": 1074086.0,
+      "reward": 0.28232031464576723,
+      "reward_std": 0.25428222119808197,
+      "rewards/env_goofspiel_reward/mean": 0.2823203206062317,
+      "rewards/env_goofspiel_reward/std": 0.38380778431892393,
+      "sampling/importance_sampling_ratio/max": 2.474874496459961,
+      "sampling/importance_sampling_ratio/mean": 0.972897756099701,
+      "sampling/importance_sampling_ratio/min": 0.00847407877445221,
+      "sampling/sampling_logp_difference/max": 8.352801895141601,
+      "sampling/sampling_logp_difference/mean": 0.09170940220355987,
+      "step": 5,
+      "step_time": 27.68519940240021
     }
   ],
   "logging_steps": 5,
-  "max_steps": 12,
-  "num_input_tokens_seen": 439873,
+  "max_steps": 2343,
+  "num_input_tokens_seen": 1930059,
   "num_train_epochs": 3,
   "save_steps": 500,
   "stateful_callbacks": {
@@ -54,13 +55,13 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
   "total_flos": 0.0,
-  "train_batch_size": 42,
+  "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86716a0f9d6726730f9bb33b6ff3e2ea21914baec34e4348e4636d0c05bbd3ef
-size 7889
+oid sha256:c896334b84a5cf87c711da6ada923aa8cad601284747089a6a46ea1af55c0030
+size 7185