Upload folder using huggingface_hub (#1)

- 727cbe92c46205e920ce41200eb70edde1d3032859e385b1528046d231300127 (e98409505a9a4edd23c735d630960730f5d30844)
- 9e4d8449a580d66917014edba4c9f9775cbb40970df2ba02f02e2c4dcfb10840 (d9622dc3ae1cef95f99c4790a41f082bd070b537)
- bd9777834c184d0a2047a8fb47549c691b9abcfdd2fdecdfa3a3f0259b526e88 (4a0edf75ca3995ed728598e56e675c7c3cfee778)
- 3542f13a79f069a192eff275127c3eef9ae2810cc75cbe532d9ba5f7a07238a7 (2f26511b8ef55ee0db54e3b088dc3c38408c85f8)
- 1cec57cb2612e067b0a49f643f773e59fe2397b753c3144cbd8c2c4e47d3fd52 (602d1f35280629655397b3b62e69290338d12ee9)
- df9218404066df5842bf1f034dfc67ec034fd989c39a12b7154c44d0cc604862 (773cb7dbd90b0af9f52380fd2a9ecbb7ca82ac78)
- e2ff689cc6844a0ad2aba43af3edea12a567c28184563a032ba6e25b10502382 (d9f5da3eb82d0f86a19bb736a503ceb6bdd54fe4)

config.json

@@ -0,0 +1,55 @@
+{
+  "_name_or_path": "meta-llama/Llama-3.3-70B-Instruct",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 80,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "quip_params": {
+    "K": 3,
+    "L": 16,
+    "V": 2,
+    "codebook": "bitshift",
+    "codebook_version": 0,
+    "decode_mode": "quantlut_sym",
+    "skip_list": [
+      "0_v"
+    ],
+    "split_for_tp": false,
+    "td_x": 16,
+    "td_y": 16,
+    "tlut_bits": 9
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 8.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.2",
+  "use_cache": true,
+  "vocab_size": 128256
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.46.2"
+}

model-00001-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bfd90cdcf248fdf1736cb3ea8da9cd73aba8ca4f5034135b7642c3133c15b38
+size 4919418488

model-00002-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fd2264e2fafd911a5e455d39a624414d444a4b9a8a241986a5df422e1755375
+size 4966090984

model-00003-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f764ccbdae1f07be0eeb559067c4b82d287f49d4de1ddfa0a877341a464c34d7
+size 4997540024

model-00004-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2debae60ac649cdc7b65dabd9f753f36649c1a201ecd2c7405b944232e9fef35
+size 4966091040

model-00005-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d442a71828d431a62d474eee8c01e06b8054ccc33d9e1ced431a96b5e28178e2
+size 4997540024

model-00006-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44802abe6eabbcf71d7a5734830d57d388aa1e71ad0da6c84e2ae843f46ae40c
+size 2980887216

model-00007-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0436c98c9b4f78e61f6769e588e0853179b970e2a084ae2886e2daebf1ce13d
+size 2101346432