Upload MambaForCausalLM

config.json

@@ -0,0 +1,62 @@
+{
+  "_name_or_path": "./checkpoint-800",
+  "alpha_pattern": {},
+  "architectures": [
+    "MambaForCausalLM"
+  ],
+  "auto_mapping": null,
+  "base_model_name_or_path": "state-spaces/mamba-370m-hf",
+  "bias": "none",
+  "bos_token_id": 0,
+  "conv_kernel": 4,
+  "eos_token_id": 0,
+  "expand": 2,
+  "fan_in_fan_out": false,
+  "hidden_act": "silu",
+  "hidden_size": 768,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "initializer_range": 0.1,
+  "intermediate_size": 1536,
+  "layer_norm_epsilon": 1e-05,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 8,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "model_type": "mamba",
+  "modules_to_save": null,
+  "num_hidden_layers": 32,
+  "pad_token_id": 0,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "rescale_prenorm_residual": false,
+  "residual_in_fp32": true,
+  "revision": null,
+  "state_size": 16,
+  "target_modules": [
+    "embeddings",
+    "x_proj",
+    "out_proj",
+    "in_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "time_step_floor": 0.0001,
+  "time_step_init_scheme": "random",
+  "time_step_max": 0.1,
+  "time_step_min": 0.001,
+  "time_step_rank": 48,
+  "time_step_scale": 1.0,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "use_bias": false,
+  "use_cache": true,
+  "use_conv_bias": true,
+  "use_dora": false,
+  "use_rslora": false,
+  "vocab_size": 50280
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "pad_token_id": 0,
+  "transformers_version": "4.40.1"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e6548525a9b565075da7876e18d12ea00a40667f54ac0ae5012e1f50b8bb767
+size 637268960