upload lora

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_128/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 16,
+    "target_modules": [
+        "k_proj",
+        "o_proj",
+        "down_proj",
+        "gate_proj",
+        "q_proj",
+        "up_proj",
+        "v_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 32,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_128/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f1fb99524d62fb79af445666801a84e91e21c6b661b2dec1cffb3e8b622c19c
+size 119801496

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_192/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 16,
+    "target_modules": [
+        "k_proj",
+        "o_proj",
+        "down_proj",
+        "gate_proj",
+        "q_proj",
+        "up_proj",
+        "v_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 32,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_192/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a451728cf477be4a4e15b4fe2d64302710961042989101c16480abfe62b8c26
+size 119801496

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_256/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 16,
+    "target_modules": [
+        "o_proj",
+        "q_proj",
+        "up_proj",
+        "k_proj",
+        "v_proj",
+        "gate_proj",
+        "down_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 32,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_256/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbfd16f689eb5119e11269f253dc9eb1504a96f73091e1d6932d8c462108a762
+size 119801496

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_320/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 16,
+    "target_modules": [
+        "o_proj",
+        "q_proj",
+        "up_proj",
+        "k_proj",
+        "v_proj",
+        "gate_proj",
+        "down_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 32,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_320/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d50ef6d9c84b34bdda22be8f3d0989baa73604480b42c2e1bbc74da24af7632
+size 119801496

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_384/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 16,
+    "target_modules": [
+        "o_proj",
+        "q_proj",
+        "up_proj",
+        "k_proj",
+        "v_proj",
+        "gate_proj",
+        "down_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 32,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_384/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a64b26eec62e2b067f83546607c803010d245e51fab76a260968b944ef1ea14e
+size 119801496

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_448/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 16,
+    "target_modules": [
+        "o_proj",
+        "q_proj",
+        "up_proj",
+        "k_proj",
+        "v_proj",
+        "gate_proj",
+        "down_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 32,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_448/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d768ffde3ce3412b87033edf42df9896bf5a964c16885a1c9ab6eb6a74375123
+size 119801496

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_512/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 16,
+    "target_modules": [
+        "o_proj",
+        "q_proj",
+        "up_proj",
+        "k_proj",
+        "v_proj",
+        "gate_proj",
+        "down_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 32,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_512/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fcbae757ee3552da3cdc0e7775969069cd616937d3d9413bfeec09be238ba52
+size 119801496

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_64/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 16,
+    "target_modules": [
+        "k_proj",
+        "o_proj",
+        "down_proj",
+        "gate_proj",
+        "q_proj",
+        "up_proj",
+        "v_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 32,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank16/global_step_64/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2578414694c142126fdba73a11718d68178ca5b6f4f6dbb851a58e561f7eaca7
+size 119801496

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_128/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 32,
+    "target_modules": [
+        "k_proj",
+        "up_proj",
+        "down_proj",
+        "v_proj",
+        "q_proj",
+        "o_proj",
+        "gate_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 64,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_128/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fe6185fe4d2ee1e35ffacfd2e07c7a58583bfd4d83e8f31d09cb046b165a3a7
+size 239536248

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_192/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 32,
+    "target_modules": [
+        "k_proj",
+        "up_proj",
+        "down_proj",
+        "v_proj",
+        "q_proj",
+        "o_proj",
+        "gate_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 64,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_192/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecabed98da06e091980486ca97ffd64bcefca86da6da62c84d6fd89ce91e29be
+size 239536248

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_256/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 32,
+    "target_modules": [
+        "k_proj",
+        "up_proj",
+        "down_proj",
+        "v_proj",
+        "q_proj",
+        "o_proj",
+        "gate_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 64,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_256/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c697eea35c14797c8141135eb4d6dbb3baf5f6be92ac69134e7bce9874e01deb
+size 239536248

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_320/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 32,
+    "target_modules": [
+        "k_proj",
+        "up_proj",
+        "down_proj",
+        "v_proj",
+        "q_proj",
+        "o_proj",
+        "gate_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 64,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_320/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:278824427889e589d4fecd55b33fb38189408f865295f8e2649ca8d2d97b99f1
+size 239536248

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_384/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 32,
+    "target_modules": [
+        "k_proj",
+        "up_proj",
+        "down_proj",
+        "v_proj",
+        "q_proj",
+        "o_proj",
+        "gate_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 64,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_384/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:497e9f4fb588330bde941e871067729c531f1f19b99b622ef6ee7f1fadcbfbc6
+size 239536248

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_448/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 32,
+    "target_modules": [
+        "k_proj",
+        "up_proj",
+        "down_proj",
+        "v_proj",
+        "q_proj",
+        "o_proj",
+        "gate_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 64,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_448/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:776e22339116930f3018a2f3b8fb3c2697ef89123100f610aaaeecdcf42a291d
+size 239536248

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_512/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 32,
+    "target_modules": [
+        "k_proj",
+        "up_proj",
+        "down_proj",
+        "v_proj",
+        "q_proj",
+        "o_proj",
+        "gate_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 64,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_512/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52afade49505cacb116c3c7c814606d6775ef04d16cb73588036b135f4221c0b
+size 239536248

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_64/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 32,
+    "target_modules": [
+        "k_proj",
+        "up_proj",
+        "down_proj",
+        "v_proj",
+        "q_proj",
+        "o_proj",
+        "gate_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 64,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank32/global_step_64/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2a7cdcd365628213d2cfa6afbdbaaa5b34e1c55c4a5151c57bb13aa64856509
+size 239536248

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_128/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 64,
+    "target_modules": [
+        "gate_proj",
+        "k_proj",
+        "o_proj",
+        "v_proj",
+        "up_proj",
+        "down_proj",
+        "q_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 128,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_128/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7b9b3c8427ecaff63b4a3b32591942eb14858e5c67f3255a7a9e6acbb3800fd
+size 479005032

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_192/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 64,
+    "target_modules": [
+        "gate_proj",
+        "k_proj",
+        "o_proj",
+        "v_proj",
+        "up_proj",
+        "down_proj",
+        "q_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 128,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_192/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf347ef2c78a2814e2ee63f9822ded3cccfa0c9aaf3c1334705c5fe54f255ccf
+size 479005032

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_256/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 64,
+    "target_modules": [
+        "gate_proj",
+        "k_proj",
+        "o_proj",
+        "v_proj",
+        "up_proj",
+        "down_proj",
+        "q_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 128,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_256/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13a6e06cf3d61d65e5e63e48a105986493761c23d3906e901f79836b4bd85faa
+size 479005032

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_320/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 64,
+    "target_modules": [
+        "gate_proj",
+        "k_proj",
+        "o_proj",
+        "v_proj",
+        "up_proj",
+        "down_proj",
+        "q_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 128,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_320/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fef5d4bfd5ed243c1ac6ae94430c697b6d0e5cba2f589f132720fb3afd8b8d19
+size 479005032

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_384/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 64,
+    "target_modules": [
+        "gate_proj",
+        "k_proj",
+        "o_proj",
+        "v_proj",
+        "up_proj",
+        "down_proj",
+        "q_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 128,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_384/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcf380ab14b47f94e7bed10897afba3e9e3e87fe7081aae6839d2d4ae939e008
+size 479005032

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_448/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 64,
+    "target_modules": [
+        "gate_proj",
+        "k_proj",
+        "o_proj",
+        "v_proj",
+        "up_proj",
+        "down_proj",
+        "q_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 128,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_448/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b63300a374fa4a7f0f060c811c0f4ea0671f39a97b985f27d058115353531bae
+size 479005032

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_512/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 64,
+    "target_modules": [
+        "gate_proj",
+        "k_proj",
+        "o_proj",
+        "v_proj",
+        "up_proj",
+        "down_proj",
+        "q_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 128,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_512/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78e0bb7767cbde6789b99110af8b00a98c6bdd147d23e9f31e13bb078519dc78
+size 479005032

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_64/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 64,
+    "target_modules": [
+        "gate_proj",
+        "k_proj",
+        "o_proj",
+        "v_proj",
+        "up_proj",
+        "down_proj",
+        "q_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 128,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank64/global_step_64/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d71fd9a27da282dd2122f197317a221938aaa9f391f506020a8daa0408722ff
+size 479005032

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank8/global_step_128/actor/lora_adapter/adapter_config.json

@@ -0,0 +1,49 @@
+{
+    "task_type": "CAUSAL_LM",
+    "peft_type": "LORA",
+    "auto_mapping": null,
+    "peft_version": "0.18.1",
+    "base_model_name_or_path": "/dev/shm/verl-cache/b3c28c1b99a08b84eb28d5733b49c01c/aa8e72537993ba99e69dfaafa59ed015b17504d1",
+    "revision": null,
+    "inference_mode": false,
+    "r": 8,
+    "target_modules": [
+        "up_proj",
+        "k_proj",
+        "gate_proj",
+        "o_proj",
+        "q_proj",
+        "v_proj",
+        "down_proj"
+    ],
+    "exclude_modules": null,
+    "lora_alpha": 16,
+    "lora_dropout": 0.0,
+    "fan_in_fan_out": false,
+    "bias": "none",
+    "use_rslora": false,
+    "modules_to_save": null,
+    "init_lora_weights": true,
+    "layers_to_transform": null,
+    "layers_pattern": null,
+    "rank_pattern": {},
+    "alpha_pattern": {},
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "trainable_token_indices": null,
+    "loftq_config": {},
+    "eva_config": null,
+    "corda_config": null,
+    "use_dora": false,
+    "alora_invocation_tokens": null,
+    "use_qalora": false,
+    "qalora_group_size": 16,
+    "layer_replication": null,
+    "runtime_config": {
+        "ephemeral_gpu_offload": false
+    },
+    "lora_bias": false,
+    "target_parameters": null,
+    "arrow_config": null,
+    "ensure_weight_tying": false
+}

qwen-3b-lora/Qwen/Qwen2.5-3B-Instruct-polaris-GRPO-LoRA-rank8/global_step_128/actor/lora_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2baeef11332894cb3bc6cc15b91c60c0e3581dad844a84a0ed9befe37740788
+size 59933600