Upload folder using huggingface_hub

phase4_50steps_smoke/checkpoint-25/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

phase4_50steps_smoke/checkpoint-25/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "Qwen/Qwen2-0.5B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 896,
+  "initializer_range": 0.02,
+  "intermediate_size": 4864,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 24,
+  "model_type": "qwen2",
+  "num_attention_heads": 14,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.47.1",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

phase4_50steps_smoke/checkpoint-25/generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.47.1"
+}

phase4_50steps_smoke/checkpoint-25/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8cf44f285aa7ccf92adffcfebd3e138106614d299892c96bd5998835e7061ce
+size 988097824

phase4_50steps_smoke/checkpoint-25/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e32ab6bb959cfb92f7e33b01a73ddab6db6850036e7a2812bd617f5c8a4e0cc
+size 1976374202

phase4_50steps_smoke/checkpoint-25/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a225a70a9f40867bb4b43f22cb2f00d38476632e75acae9f866ea52253ec7eb
+size 14244

phase4_50steps_smoke/checkpoint-25/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0147ee358028ddb9cb7d88061e4ce5373f63bb85dc349b4dd3f9fceb5e922b44
+size 1064

phase4_50steps_smoke/checkpoint-25/special_tokens_map.json

@@ -0,0 +1,14 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|im_end|>"
+}

phase4_50steps_smoke/checkpoint-25/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17db16dda2e1867e40857a8ff53ecb50be68a9d5b5763bf7994271cc3b73dcb1
+size 11418430

phase4_50steps_smoke/checkpoint-25/tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|im_end|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

phase4_50steps_smoke/checkpoint-25/trainer_state.json

@@ -0,0 +1,93 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.8,
+  "eval_steps": 500,
+  "global_step": 25,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 87.5125,
+      "epoch": 0.16,
+      "grad_norm": 33.0,
+      "kl": 0.0008317360421642661,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.08881250247359276,
+      "reward_std": 0.1590106257237494,
+      "rewards/real_reward_fn": -0.08881250247359276,
+      "step": 5
+    },
+    {
+      "completion_length": 85.86875,
+      "epoch": 0.32,
+      "grad_norm": 55.25,
+      "kl": 0.001102981199801434,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.08131250264123083,
+      "reward_std": 0.1880020027048886,
+      "rewards/real_reward_fn": -0.08131250264123083,
+      "step": 10
+    },
+    {
+      "completion_length": 83.225,
+      "epoch": 0.48,
+      "grad_norm": 39.0,
+      "kl": 0.001013497701205779,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": 0.02462499784305692,
+      "reward_std": 0.13523416444659234,
+      "rewards/real_reward_fn": 0.02462499784305692,
+      "step": 15
+    },
+    {
+      "completion_length": 80.8125,
+      "epoch": 0.64,
+      "grad_norm": 27.125,
+      "kl": 0.0011422165334806778,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.08068750295788049,
+      "reward_std": 0.18747167382389307,
+      "rewards/real_reward_fn": -0.08068750295788049,
+      "step": 20
+    },
+    {
+      "completion_length": 80.29375,
+      "epoch": 0.8,
+      "grad_norm": 30.5,
+      "kl": 0.0010913557125604711,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.03431250210851431,
+      "reward_std": 0.18764845319092274,
+      "rewards/real_reward_fn": -0.03431250210851431,
+      "step": 25
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 50,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 25,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

phase4_50steps_smoke/checkpoint-25/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e3ee00a3958bd8e3dd9fd7996d93814add2487f8f288a5fd88363fd1225305d
+size 5560

phase4_50steps_smoke/checkpoint-50/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

phase4_50steps_smoke/checkpoint-50/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "Qwen/Qwen2-0.5B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 896,
+  "initializer_range": 0.02,
+  "intermediate_size": 4864,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 24,
+  "model_type": "qwen2",
+  "num_attention_heads": 14,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.47.1",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

phase4_50steps_smoke/checkpoint-50/generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.47.1"
+}

phase4_50steps_smoke/checkpoint-50/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67b6a26050b73c52cc643bddf5c0eb6780a4ebfeb59dffafa5f33b314e1a4249
+size 988097824

phase4_50steps_smoke/checkpoint-50/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f6c37c3d737412b0ea3a5186370ad1d09dce1f7227f23cdf723423eec5baafd
+size 1976374202

phase4_50steps_smoke/checkpoint-50/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14a6bb635f72161d724e2a0bc9be0fe4f34e8ab268b3521b539c9c44508c3fc2
+size 14244

phase4_50steps_smoke/checkpoint-50/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:236d066e7dadb3b02c6a940d84a1d2ac49885f42b02673c132b47bc8e889c5e6
+size 1064

phase4_50steps_smoke/checkpoint-50/special_tokens_map.json

@@ -0,0 +1,14 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|im_end|>"
+}

phase4_50steps_smoke/checkpoint-50/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17db16dda2e1867e40857a8ff53ecb50be68a9d5b5763bf7994271cc3b73dcb1
+size 11418430

phase4_50steps_smoke/checkpoint-50/tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|im_end|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

phase4_50steps_smoke/checkpoint-50/trainer_state.json

@@ -0,0 +1,153 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.576,
+  "eval_steps": 500,
+  "global_step": 50,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "completion_length": 87.5125,
+      "epoch": 0.16,
+      "grad_norm": 33.0,
+      "kl": 0.0008317360421642661,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.08881250247359276,
+      "reward_std": 0.1590106257237494,
+      "rewards/real_reward_fn": -0.08881250247359276,
+      "step": 5
+    },
+    {
+      "completion_length": 85.86875,
+      "epoch": 0.32,
+      "grad_norm": 55.25,
+      "kl": 0.001102981199801434,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.08131250264123083,
+      "reward_std": 0.1880020027048886,
+      "rewards/real_reward_fn": -0.08131250264123083,
+      "step": 10
+    },
+    {
+      "completion_length": 83.225,
+      "epoch": 0.48,
+      "grad_norm": 39.0,
+      "kl": 0.001013497701205779,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": 0.02462499784305692,
+      "reward_std": 0.13523416444659234,
+      "rewards/real_reward_fn": 0.02462499784305692,
+      "step": 15
+    },
+    {
+      "completion_length": 80.8125,
+      "epoch": 0.64,
+      "grad_norm": 27.125,
+      "kl": 0.0011422165334806778,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.08068750295788049,
+      "reward_std": 0.18747167382389307,
+      "rewards/real_reward_fn": -0.08068750295788049,
+      "step": 20
+    },
+    {
+      "completion_length": 80.29375,
+      "epoch": 0.8,
+      "grad_norm": 30.5,
+      "kl": 0.0010913557125604711,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.03431250210851431,
+      "reward_std": 0.18764845319092274,
+      "rewards/real_reward_fn": -0.03431250210851431,
+      "step": 25
+    },
+    {
+      "completion_length": 80.575,
+      "epoch": 0.96,
+      "grad_norm": 58.0,
+      "kl": 0.0010487768857274204,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": 0.024874999094754456,
+      "reward_std": 0.1568009190261364,
+      "rewards/real_reward_fn": 0.024874999094754456,
+      "step": 30
+    },
+    {
+      "completion_length": 76.90441176470588,
+      "epoch": 1.096,
+      "grad_norm": 35.0,
+      "kl": 0.0010246854611015057,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.05720588389564963,
+      "reward_std": 0.14079742714324417,
+      "rewards/real_reward_fn": -0.05720588389564963,
+      "step": 35
+    },
+    {
+      "completion_length": 83.84375,
+      "epoch": 1.256,
+      "grad_norm": 28.125,
+      "kl": 0.001108163884782698,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.02056250227615237,
+      "reward_std": 0.14504527123644947,
+      "rewards/real_reward_fn": -0.02056250227615237,
+      "step": 40
+    },
+    {
+      "completion_length": 79.25,
+      "epoch": 1.416,
+      "grad_norm": 32.0,
+      "kl": 0.0010782188706798478,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": 0.021437497437000276,
+      "reward_std": 0.17562763625755906,
+      "rewards/real_reward_fn": 0.021437497437000276,
+      "step": 45
+    },
+    {
+      "completion_length": 86.08125,
+      "epoch": 1.576,
+      "grad_norm": 35.75,
+      "kl": 0.0010450664689415135,
+      "learning_rate": 5e-07,
+      "loss": 0.0,
+      "reward": -0.032250001840293405,
+      "reward_std": 0.16334165595471858,
+      "rewards/real_reward_fn": -0.032250001840293405,
+      "step": 50
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 50,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 25,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

phase4_50steps_smoke/checkpoint-50/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e3ee00a3958bd8e3dd9fd7996d93814add2487f8f288a5fd88363fd1225305d
+size 5560

training_log_phase4.json

@@ -0,0 +1,131 @@
+[
+  {
+    "loss": 0.0,
+    "grad_norm": 33.0,
+    "learning_rate": 5e-07,
+    "completion_length": 87.5125,
+    "rewards/real_reward_fn": -0.08881250247359276,
+    "reward": -0.08881250247359276,
+    "reward_std": 0.1590106257237494,
+    "kl": 0.0008317360421642661,
+    "epoch": 0.16,
+    "step": 5
+  },
+  {
+    "loss": 0.0,
+    "grad_norm": 55.25,
+    "learning_rate": 5e-07,
+    "completion_length": 85.86875,
+    "rewards/real_reward_fn": -0.08131250264123083,
+    "reward": -0.08131250264123083,
+    "reward_std": 0.1880020027048886,
+    "kl": 0.001102981199801434,
+    "epoch": 0.32,
+    "step": 10
+  },
+  {
+    "loss": 0.0,
+    "grad_norm": 39.0,
+    "learning_rate": 5e-07,
+    "completion_length": 83.225,
+    "rewards/real_reward_fn": 0.02462499784305692,
+    "reward": 0.02462499784305692,
+    "reward_std": 0.13523416444659234,
+    "kl": 0.001013497701205779,
+    "epoch": 0.48,
+    "step": 15
+  },
+  {
+    "loss": 0.0,
+    "grad_norm": 27.125,
+    "learning_rate": 5e-07,
+    "completion_length": 80.8125,
+    "rewards/real_reward_fn": -0.08068750295788049,
+    "reward": -0.08068750295788049,
+    "reward_std": 0.18747167382389307,
+    "kl": 0.0011422165334806778,
+    "epoch": 0.64,
+    "step": 20
+  },
+  {
+    "loss": 0.0,
+    "grad_norm": 30.5,
+    "learning_rate": 5e-07,
+    "completion_length": 80.29375,
+    "rewards/real_reward_fn": -0.03431250210851431,
+    "reward": -0.03431250210851431,
+    "reward_std": 0.18764845319092274,
+    "kl": 0.0010913557125604711,
+    "epoch": 0.8,
+    "step": 25
+  },
+  {
+    "loss": 0.0,
+    "grad_norm": 58.0,
+    "learning_rate": 5e-07,
+    "completion_length": 80.575,
+    "rewards/real_reward_fn": 0.024874999094754456,
+    "reward": 0.024874999094754456,
+    "reward_std": 0.1568009190261364,
+    "kl": 0.0010487768857274204,
+    "epoch": 0.96,
+    "step": 30
+  },
+  {
+    "loss": 0.0,
+    "grad_norm": 35.0,
+    "learning_rate": 5e-07,
+    "completion_length": 76.90441176470588,
+    "rewards/real_reward_fn": -0.05720588389564963,
+    "reward": -0.05720588389564963,
+    "reward_std": 0.14079742714324417,
+    "kl": 0.0010246854611015057,
+    "epoch": 1.096,
+    "step": 35
+  },
+  {
+    "loss": 0.0,
+    "grad_norm": 28.125,
+    "learning_rate": 5e-07,
+    "completion_length": 83.84375,
+    "rewards/real_reward_fn": -0.02056250227615237,
+    "reward": -0.02056250227615237,
+    "reward_std": 0.14504527123644947,
+    "kl": 0.001108163884782698,
+    "epoch": 1.256,
+    "step": 40
+  },
+  {
+    "loss": 0.0,
+    "grad_norm": 32.0,
+    "learning_rate": 5e-07,
+    "completion_length": 79.25,
+    "rewards/real_reward_fn": 0.021437497437000276,
+    "reward": 0.021437497437000276,
+    "reward_std": 0.17562763625755906,
+    "kl": 0.0010782188706798478,
+    "epoch": 1.416,
+    "step": 45
+  },
+  {
+    "loss": 0.0,
+    "grad_norm": 35.75,
+    "learning_rate": 5e-07,
+    "completion_length": 86.08125,
+    "rewards/real_reward_fn": -0.032250001840293405,
+    "reward": -0.032250001840293405,
+    "reward_std": 0.16334165595471858,
+    "kl": 0.0010450664689415135,
+    "epoch": 1.576,
+    "step": 50
+  },
+  {
+    "train_runtime": 1158.1834,
+    "train_samples_per_second": 0.691,
+    "train_steps_per_second": 0.043,
+    "total_flos": 0.0,
+    "train_loss": 4.1337845323141666e-05,
+    "epoch": 1.576,
+    "step": 50
+  }
+]