Upload folder using huggingface_hub

devkit/models--meta-llama--Llama-3.1-8B-Instruct_language_embeddings.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:460a5cfdeefd5d7bb4f7f762e8b16b192fbdb3a8f3f220c3000c96c02b5adb06
+size 2101346432

devkit/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82e9d31979e92ab929cd544440f129d9ecd797b69e327f80f17e1c50d5551b55
+size 2183982

devkit/vlm_config.json

@@ -0,0 +1,80 @@
+{
+    "model_name": "models--meta-llama--Llama-3.1-8B-Instruct",
+    "model_type": "llm-llama3.1",
+    "vm_cfg": null,
+    "mm_cfg": null,
+    "lm_cfg": {
+        "model_type": "llama",
+        "data_type": "bfloat16",
+        "arch": "llama",
+        "gen": "3.1",
+        "size": "8b",
+        "token_cfg": {
+            "tokenizer_type": "tiktoken",
+            "tokenizer_path": "/project/mlasw/users/amy.chen/debug/models--meta-llama--Llama-3.1-8B-Instruct/blobs/82e9d31979e92ab929cd544440f129d9ecd797b69e327f80f17e1c50d5551b55",
+            "vocab_size": 128256,
+            "special_tokens": {
+                "bos_token_id": 128000,
+                "eos_token_id": [
+                    128001,
+                    128008,
+                    128009
+                ]
+            }
+        },
+        "rope_cfg": {
+            "rope_theta": 500000,
+            "rope_local_base_freq": 500000,
+            "rope_scaling": {
+                "factor": 8.0,
+                "low_freq_factor": 1.0,
+                "high_freq_factor": 4.0,
+                "original_max_position_embeddings": 8192,
+                "long_factor": null,
+                "short_factor": null,
+                "rope_type": "llama3"
+            }
+        },
+        "attn_cfg": {
+            "num_attention_heads": 32,
+            "num_key_value_heads": 8,
+            "head_dim": 128,
+            "swa_enable": false,
+            "swa_ratio": 0,
+            "sliding_window": 0,
+            "attention_bias": false,
+            "attention_dropout": 0.0,
+            "query_pre_attn_scalar": 0
+        },
+        "mlp_cfg": {
+            "intermediate_size": 14336,
+            "act": "silu",
+            "num_layers": 3,
+            "mlp_bias": false
+        },
+        "hidden_size": 4096,
+        "num_hidden_layers": 32,
+        "max_position_embeddings": 2048,
+        "rms_norm_eps": 1e-05,
+        "layer_norms": [
+            "pre_attn",
+            "post_attn"
+        ],
+        "attn_logit_softcapping": null,
+        "final_logit_softcapping": null,
+        "lm_head_num_splits": 2,
+        "lm_head_split_dim": 64128
+    },
+    "pipeline_cfg": {
+        "system_prompt": null,
+        "max_num_tokens": 1024,
+        "input_token_group_size": 128,
+        "input_token_group_offsets": [
+            0,
+            128,
+            256
+        ],
+        "future_token_mask_size": 128
+    },
+    "language_model_name": "models--meta-llama--Llama-3.1-8B-Instruct_language"
+}