chore: import upstream snapshot with attribution

2026-07-13 13:24:42 +08:00
commit caf324b09d
920 changed files with 335491 additions and 0 deletions
@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "beta1": 0.85,
+  "beta2": 0.85,
+  "bias": "none",
+  "corda_config": null,
+  "deltaT": 1,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_lora_weights": true,
+  "init_r": 64,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 8,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "orth_reg_weight": 0.5,
+  "peft_type": "ADALORA",
+  "r": 8,
+  "rank_pattern": null,
+  "revision": null,
+  "target_modules": null,
+  "target_r": 32,
+  "task_type": null,
+  "tfinal": 500,
+  "tinit": 200,
+  "total_step": 5000,
+  "use_dora": false,
+  "use_rslora": false
+}
@@ -0,0 +1,10 @@
+{
+  "peft_type": "ADAMSS",
+  "r": 100,
+  "num_subspaces": 32,
+  "subspace_rank": 1,
+  "target_modules": null,
+  "init_weights": "orthogonal",
+  "use_asa": false,
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,10 @@
+{
+  "model_id": "meta-llama/Llama-3.2-3B",
+  "max_steps": 5000,
+  "batch_size": 4,
+  "eval_steps": 250,
+  "optimizer_kwargs": {
+    "lr": 1e-4,
+    "weight_decay": 0.1
+  }
+}
@@ -0,0 +1,11 @@
+{
+  "adapter_layers": 28,
+  "adapter_len": 100,
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "inference_mode": false,
+  "peft_type": "ADAPTION_PROMPT",
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 5e-4
+  }
+}
+
@@ -0,0 +1,12 @@
+{
+    "adapter_layers": 28,
+    "adapter_len": 100,
+    "auto_mapping": null,
+    "base_model_name_or_path": null,
+    "inference_mode": false,
+    "peft_type": "ADAPTION_PROMPT",
+    "peft_version": "0.19.2.dev0@UNKNOWN",
+    "revision": null,
+    "target_modules": null,
+    "task_type": null
+}
@@ -0,0 +1,23 @@
+{
+    "model_id": "meta-llama/Llama-3.2-3B",
+    "dtype": "bfloat16",
+    "max_seq_length": 512,
+    "batch_size": 4,
+    "batch_size_eval": 64,
+    "max_steps": 3000,
+    "eval_steps": 500,
+    "compile": false,
+    "use_gc": true,
+    "use_amp": false,
+    "grad_norm_clip": 1.0,
+    "optimizer_type": "AdamW",
+    "optimizer_kwargs": {
+        "lr": 0.005,
+        "weight_decay": 0.1
+    },
+    "lr_scheduler": "cosine",
+    "generation_kwargs": {
+        "max_new_tokens": 256
+    },
+    "query_template": "Question: {query} Think step by step.\nAnswer:"
+    }
@@ -0,0 +1,14 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "inference_mode": false,
+  "init_weights": true,
+  "modules_to_save": null,
+  "peft_type": "BEFT",
+  "peft_version": "0.19.2.dev0@UNKNOWN",
+  "revision": null,
+  "target_modules": [
+    "v_proj"
+  ],
+  "task_type": null
+}
@@ -0,0 +1,20 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "boft_block_num": 0,
+  "boft_block_size": 4,
+  "boft_dropout": 0.0,
+  "boft_n_butterfly_factor": 1,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "BOFT",
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": false,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "block_size": 64,
+  "block_size_pattern": {},
+  "peft_type": "C3A",
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": null
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 3e-1,
+    "weight_decay": 1e-5
+  }
+}
@@ -0,0 +1,23 @@
+{
+  "alpha": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "decomposition_method": "qr",
+  "deft_dropout": 0.0,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_scale": 1.0,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "para": true,
+  "peft_type": "DEFT",
+  "peft_version": "0.19.2.dev0@UNKNOWN",
+  "r": 32,
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,23 @@
+{
+  "alpha": 128,
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "decomposition_method": "relu",
+  "deft_dropout": 0.0,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_scale": 1.0,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "para": false,
+  "peft_type": "DEFT",
+  "peft_version": "0.19.2.dev0@UNKNOWN",
+  "r": 32,
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,20 @@
+{
+  "lambda_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "delora_lambda": 15,
+  "module_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "DELORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
+
@@ -0,0 +1,23 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": false,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "n_frequency": 1000,
+  "n_frequency_pattern": {},
+  "peft_type": "FOURIERFT",
+  "random_loc_seed": 777,
+  "revision": null,
+  "scaling": 300,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": null
+}
@@ -0,0 +1,23 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": false,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "n_frequency": 5000,
+  "n_frequency_pattern": {},
+  "peft_type": "FOURIERFT",
+  "random_loc_seed": 777,
+  "revision": null,
+  "scaling": 300,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": null
+}
@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "frod_dropout": 0.0,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "FROD",
+  "projection_prng_key": 0,
+  "regularization_alpha": 0.001,
+  "revision": null,
+  "runtime_offload_base_weight": true,
+  "save_projection": true,
+  "sparse_rate": 0.01,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "frod_dropout": 0.0,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "FROD",
+  "projection_prng_key": 0,
+  "regularization_alpha": 0.001,
+  "revision": null,
+  "runtime_offload_base_weight": true,
+  "save_projection": true,
+  "sparse_rate": 0.02,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,5 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-5
+  }
+}
+
@@ -0,0 +1,18 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "config_A_B": "lora",
+  "config_C": "lora",
+  "config_D_E": "constant",
+  "inference_mode": false,
+  "init_weights": true,
+  "modules_to_save": null,
+  "peft_type": "GLORA",
+  "r": 8,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,18 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "config_A_B": "vector",
+  "config_C": "vector",
+  "config_D_E": "constant",
+  "inference_mode": false,
+  "init_weights": true,
+  "modules_to_save": null,
+  "peft_type": "GLORA",
+  "r": 8,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 5e-4,
+    "weight_decay": 0.0
+  }
+}
@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "alpha": 64,
+  "gralora_dropout": 0.0,
+  "gralora_k": 2,
+  "hybrid_r": 0,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "GRALORA",
+  "peft_version": "0.17.2.dev0@UNKNOWN",
+  "r": 32,
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,5 @@
+{
+  "optimizer_kwargs": {
+    "lr": 2e-4
+  }
+}
@@ -0,0 +1,20 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "hira_dropout": 0.0,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "HIRA",
+  "r": 32,
+  "rank_pattern": {},
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 4.5e-3,
+    "weight_decay": 0.0
+  }
+}
@@ -0,0 +1,14 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "feedforward_modules": ["down_proj", "gate_proj", "up_proj"],
+  "inference_mode": false,
+  "init_ia3_weights": true,
+  "modules_to_save": null,
+  "peft_type": "IA3",
+  "revision": null,
+  "target_modules": ["q_proj", "k_proj", "v_proj", "o_proj", "down_proj", "gate_proj", "up_proj"],
+  "task_type": null
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 5e-3,
+    "weight_decay": 0.0
+  }
+}
@@ -0,0 +1,14 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "feedforward_modules": null,
+  "inference_mode": false,
+  "init_ia3_weights": true,
+  "modules_to_save": null,
+  "peft_type": "IA3",
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,14 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "feedforward_modules": null,
+  "inference_mode": false,
+  "init_ia3_weights": true,
+  "modules_to_save": null,
+  "peft_type": "IA3",
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
+
@@ -0,0 +1,19 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_B": 2,
+  "peft_type": "LILY",
+  "peft_version": "0.18.2.dev0@UNKNOWN",
+  "r": 140,
+  "revision": null,
+  "scaling": 8.0,
+  "stride_A": 14,
+  "target_modules": ["gate_proj", "up_proj", "down_proj"],
+  "task_type": null
+}
@@ -0,0 +1,22 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_B": 2,
+  "peft_type": "LILY",
+  "peft_version": "0.18.2.dev0@UNKNOWN",
+  "r": 896,
+  "revision": null,
+  "scaling": 2.0,
+  "stride_A": 14,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null
+}
@@ -0,0 +1,11 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "exclude_modules": null,
+  "inference_mode": false,
+  "modules_to_save": null,
+  "peft_type": "LN_TUNING",
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,24 @@
+{
+  "alpha": 64,
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "module_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "LOHA",
+  "r": 32,
+  "rank_dropout": 0.0,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null,
+  "use_effective_conv2d": false
+}
@@ -0,0 +1,27 @@
+{
+  "alpha": 64,
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "decompose_both": false,
+  "decompose_factor": -1,
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "module_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "LOKR",
+  "r": 32,
+  "rank_dropout": 0.0,
+  "rank_dropout_scale": false,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null,
+  "use_effective_conv2d": false
+}
@@ -0,0 +1,30 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 20,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 10,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": ["gate_proj", "up_proj", "down_proj"],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
@@ -0,0 +1,48 @@
+{
+    "alpha_pattern": {},
+    "auto_mapping": null,
+    "base_model_name_or_path": null,
+    "bias": "none",
+    "corda_config": null,
+    "eva_config": null,
+    "exclude_modules": null,
+    "fan_in_fan_out": false,
+    "inference_mode": false,
+    "init_lora_weights": true,
+    "layer_replication": null,
+    "layers_pattern": null,
+    "layers_to_transform": null,
+    "loftq_config": {},
+    "lora_alpha": 28,
+    "lora_bias": false,
+    "lora_dropout": 0.0,
+    "megatron_config": null,
+    "megatron_core": "megatron.core",
+    "modules_to_save": null,
+    "peft_type": "LORA",
+    "r": 14,
+    "rank_pattern": {},
+    "revision": null,
+    "target_modules": [
+        "gate_proj",
+        "up_proj",
+        "down_proj"
+    ],
+    "task_type": "CAUSAL_LM",
+    "use_dora": false,
+    "use_rslora": false,
+    "use_bdlora": {
+        "target_modules_bd_a": [
+            "o_proj",
+            "down_proj"
+        ],
+        "target_modules_bd_b": [
+            "q_proj",
+            "v_proj",
+            "k_proj",
+            "up_proj",
+            "gate_proj"
+        ],
+        "nblocks": 2
+    }
+}
@@ -0,0 +1,30 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM",
+  "use_dora": true,
+  "use_rslora": false
+}
@@ -0,0 +1,30 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
@@ -0,0 +1,9 @@
+{
+  "optimizer_type": "lora-fa",
+  "optimizer_kwargs": {
+    "r": 32,
+    "lora_alpha": 64,
+    "lr": 1e-4,
+    "weight_decay": 0.1
+  }
+}
@@ -0,0 +1,30 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_lora_weights": "mica",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
@@ -0,0 +1,30 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
@@ -0,0 +1,30 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": true
+}
@@ -0,0 +1,30 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": null,
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
@@ -0,0 +1,18 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": "bat",
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "mini_r": 1,
+  "miss_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "MISS",
+  "r": 64,
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,18 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "mini_r": 1,
+  "miss_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "MISS",
+  "r": 64,
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,18 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": "mini",
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "mini_r": 64,
+  "miss_dropout": 0.0,
+  "modules_to_save": null,
+  "peft_type": "MISS",
+  "r": 64,
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,27 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "block_share": false,
+  "coft": false,
+  "eps": 6e-05,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "module_dropout": 0.0,
+  "modules_to_save": null,
+  "oft_block_size": 0,
+  "peft_type": "OFT",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null
+}
@@ -0,0 +1,28 @@
+{
+  "task_type": null,
+  "peft_type": "OSF",
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Llama-3.2-3B",
+  "revision": null,
+  "inference_mode": false,
+  "effective_rank": null,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "down_proj",
+    "up_proj"
+  ],
+  "rank_pattern": {
+    "q_proj": 2944,
+    "o_proj": 2944,
+    "k_proj": 896,
+    "v_proj": 896,
+    "gate_proj": 2944,
+    "down_proj": 2944,
+    "up_proj": 2944
+  }
+}
+
@@ -0,0 +1,7 @@
+{
+  "autocast_adapter_dtype": false,
+  "optimizer_kwargs": {
+    "lr": 5e-5
+  }
+}
+
@@ -0,0 +1,22 @@
+{
+  "act_fn": "relu",
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "depth": 0,
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "PEANUT",
+  "peft_version": "0.18.2.dev0@UNKNOWN",
+  "r": 1,
+  "revision": null,
+  "scaling": 32.0,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null
+}
@@ -0,0 +1,22 @@
+{
+  "act_fn": "relu",
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "depth": 0,
+  "exclude_modules": null,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "PEANUT",
+  "peft_version": "0.18.2.dev0@UNKNOWN",
+  "r": 32,
+  "revision": null,
+  "scaling": 2.0,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null
+}
@@ -0,0 +1,15 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "encoder_hidden_size": 3072,
+  "inference_mode": false,
+  "num_attention_heads": 24,
+  "num_layers": 28,
+  "num_transformer_submodules": 1,
+  "num_virtual_tokens": 200,
+  "peft_type": "PREFIX_TUNING",
+  "prefix_projection": false,
+  "revision": null,
+  "task_type": "CAUSAL_LM",
+  "token_dim": 3072
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
+
@@ -0,0 +1,16 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "encoder_hidden_size": 3072,
+  "inference_mode": false,
+  "num_attention_heads": 24,
+  "num_layers": 28,
+  "num_transformer_submodules": 1,
+  "num_virtual_tokens": 200,
+  "init_weights": "zero",
+  "peft_type": "PREFIX_TUNING",
+  "prefix_projection": false,
+  "revision": null,
+  "task_type": "CAUSAL_LM",
+  "token_dim": 3072
+}
@@ -0,0 +1,5 @@
+{
+  "optimizer_kwargs": {
+    "lr": 0.005
+  }
+}
@@ -0,0 +1,16 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "encoder_hidden_size": 3072,
+  "inference_mode": false,
+  "num_attention_heads": 24,
+  "num_layers": 28,
+  "num_transformer_submodules": 1,
+  "num_virtual_tokens": 50,
+  "init_weights": "zero",
+  "peft_type": "PREFIX_TUNING",
+  "prefix_projection": false,
+  "revision": null,
+  "task_type": "CAUSAL_LM",
+  "token_dim": 3072
+}
@@ -0,0 +1,6 @@
+{
+  "init_kv_cache_prefix": "You are a helpful math assistant. Solve mathematical problems by thinking step by step. First, identify what the problem is asking. Then, break down the solution into clear logical steps. Show all your work and calculations. Finally, verify that your answer makes sense.",
+  "optimizer_kwargs": {
+    "lr": 0.005
+  }
+}
@@ -0,0 +1,17 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "inference_mode": false,
+  "num_attention_heads": 24,
+  "num_layers": 28,
+  "num_transformer_submodules": 1,
+  "num_virtual_tokens": 200,
+  "peft_type": "PROMPT_TUNING",
+  "prompt_tuning_init": "RANDOM",
+  "prompt_tuning_init_text": null,
+  "revision": null,
+  "task_type": "CAUSAL_LM",
+  "token_dim": 3072,
+  "tokenizer_kwargs": null,
+  "tokenizer_name_or_path": null
+}
@@ -0,0 +1,17 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "inference_mode": false,
+  "num_attention_heads": 24,
+  "num_layers": 28,
+  "num_transformer_submodules": 1,
+  "num_virtual_tokens": 200,
+  "peft_type": "PROMPT_TUNING",
+  "prompt_tuning_init": "RANDOM",
+  "prompt_tuning_init_text": null,
+  "revision": null,
+  "task_type": "CAUSAL_LM",
+  "token_dim": 3072,
+  "tokenizer_kwargs": null,
+  "tokenizer_name_or_path": null
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
+
@@ -0,0 +1,17 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "inference_mode": false,
+  "num_attention_heads": 24,
+  "num_layers": 28,
+  "num_transformer_submodules": 1,
+  "num_virtual_tokens": 200,
+  "peft_type": "PROMPT_TUNING",
+  "prompt_tuning_init": "SAMPLE_VOCAB",
+  "prompt_tuning_init_text": null,
+  "revision": null,
+  "task_type": "CAUSAL_LM",
+  "token_dim": 3072,
+  "tokenizer_kwargs": null,
+  "tokenizer_name_or_path": null
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
+
@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "r": 256,
+  "psoft_alpha": 256,
+  "psoft_dropout": 0.0,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "ab_svd_init": "psoft_init",
+  "psoft_svd": "full",
+  "psoft_svd_lowrank_niter": 10,
+  "psoft_orth": true,
+  "psoft_mag_a": true,
+  "psoft_mag_b": true,
+  "use_cayley_neumann": false,
+  "num_cayley_neumann_terms": 5,
+  "cayley_neumann_eps": null,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "PSOFT",
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "r": 256,
+  "psoft_alpha": 256,
+  "psoft_dropout": 0.0,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "ab_svd_init": "psoft_init",
+  "psoft_svd": "lowrank",
+  "psoft_svd_lowrank_niter": 10,
+  "psoft_orth": true,
+  "psoft_mag_a": true,
+  "psoft_mag_b": true,
+  "use_cayley_neumann": true,
+  "num_cayley_neumann_terms": 5,
+  "cayley_neumann_eps": null,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "PSOFT",
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,17 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "encoder_dropout": 0.0,
+  "encoder_hidden_size": 3072,
+  "encoder_num_layers": 2,
+  "encoder_reparameterization_type": "MLP",
+  "inference_mode": false,
+  "num_attention_heads": 24,
+  "num_layers": 28,
+  "num_transformer_submodules": 1,
+  "num_virtual_tokens": 20,
+  "peft_type": "P_TUNING",
+  "revision": null,
+  "task_type": "CAUSAL_LM",
+  "token_dim": 3072
+}
@@ -0,0 +1,20 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "d_initial": 0.1,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "PVERA",
+  "projection_prng_key": 0,
+  "r": 256,
+  "revision": null,
+  "save_projection": true,
+  "target_modules": null,
+  "task_type": null,
+  "pvera_dropout": 0.0
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 3e-3
+  }
+}
+
@@ -0,0 +1,20 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "d_initial": 0.1,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "PVERA",
+  "projection_prng_key": 0,
+  "r": 64,
+  "revision": null,
+  "save_projection": true,
+  "target_modules": null,
+  "task_type": null,
+  "pvera_dropout": 0.0
+}
@@ -0,0 +1,5 @@
+{
+  "optimizer_kwargs": {
+    "lr": 3e-3
+  }
+}
@@ -0,0 +1,22 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "RANDLORA",
+  "projection_prng_key": 0,
+  "r": 32,
+  "randlora_alpha": 640,
+  "randlora_dropout": 0.0,
+  "revision": null,
+  "save_projection": true,
+  "sparse": false,
+  "target_modules": null,
+  "task_type": null,
+  "very_sparse": false
+}
@@ -0,0 +1,12 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "group_size": 64,
+  "inference_mode": false,
+  "init_weights": true,
+  "peft_type": "ROAD",
+  "revision": null,
+  "target_modules": null,
+  "task_type": null,
+  "variant": "road_2"
+}
@@ -0,0 +1,5 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
@@ -0,0 +1,15 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "mask_type": "random",
+  "modules_to_save": null,
+  "peft_type": "SHIRA",
+  "r": 32,
+  "random_seed": 42,
+  "revision": null,
+  "target_modules": null,
+  "task_type": null
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 3e-4
+  }
+}
+
@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "TINYLORA",
+  "projection_seed": 42,
+  "r": 2,
+  "revision": null,
+  "save_projection": true,
+  "target_modules": null,
+  "task_type": null,
+  "tinylora_dropout": 0.0,
+  "u": 64,
+  "weight_tying": 0.0
+}
@@ -0,0 +1,5 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "TINYLORA",
+  "projection_seed": 42,
+  "r": 2,
+  "revision": null,
+  "save_projection": true,
+  "target_modules": null,
+  "task_type": null,
+  "tinylora_dropout": 0.0,
+  "u": 64,
+  "weight_tying": 1.0
+}
@@ -0,0 +1,5 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
@@ -0,0 +1,7 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "peft_type": "TRAINABLE_TOKENS",
+  "token_indices": [128000, 128001],
+  "task_type": "CAUSAL_LM"
+}
@@ -0,0 +1,5 @@
+{
+  "optimizer_kwargs": {
+    "lr": 0.2
+  }
+}
@@ -0,0 +1,24 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_theta_d_bound": 0.02,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "UNILORA",
+  "proj_seed": 42,
+  "r": 32,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "theta_d_length": 256,
+  "unilora_dropout": 0.0
+}
@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_logits_std": 0.1,
+  "init_vector_bank_bound": 0.02,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "num_vectors": 256,
+  "peft_type": "VBLORA",
+  "r": 4,
+  "revision": null,
+  "save_only_topk_weights": false,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": null,
+  "topk": 2,
+  "vblora_dropout": 0.0,
+  "vector_length": 256
+}
@@ -0,0 +1,20 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "d_initial": 0.1,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "peft_type": "VERA",
+  "projection_prng_key": 0,
+  "r": 256,
+  "revision": null,
+  "save_projection": true,
+  "target_modules": null,
+  "task_type": null,
+  "vera_dropout": 0.0
+}
@@ -0,0 +1,6 @@
+{
+  "optimizer_kwargs": {
+    "lr": 1e-3
+  }
+}
+
@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": false,
+  "init_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "modules_to_save": null,
+  "n_frequency": 5000,
+  "n_frequency_pattern": {},
+  "peft_type": "WAVEFT",
+  "proportional_parameters": false,
+  "random_loc_seed": 777,
+  "revision": null,
+  "scaling": 25.0,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_idwt": true,
+  "wavelet_family": "db1"
+}