Upload PhiMoEForCausalLM

config.json

@@ -16,9 +16,9 @@
   "hidden_size": 3072,
   "initializer_range": 0.02,
   "input_jitter_noise": 0.01,
-  "intermediate_size": 4096,
+  "intermediate_size": 2240,
   "lm_head_bias": true,
-  "max_position_embeddings": 131072,
+  "max_position_embeddings": 4096,
   "model_type": "phimoe",
   "num_attention_heads": 32,
   "num_experts_per_tok": 2,
@@ -32,7 +32,7 @@
   "rope_theta": 10000.0,
   "router_aux_loss_coef": 0.0,
   "router_jitter_noise": 0.01,
-  "sliding_window": 131072,
+  "sliding_window": 4096,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.41.2",

model-00001-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4936082815d6ecad882993aba412267415e8b63641948814c208c0dac7e966d
+size 4987367352

model-00002-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88044b49f86be88317fa71a4687c38f1600525b220ea953c1ce57960a90b8248
+size 4996831624

model-00003-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3662118b09dbf0b540ac61fe3861d980c5fbade8b5f618727b0657b122b40e7
+size 4996831808

model-00004-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c30d3b366b6eec144f4d4b7a312452daa74185694b29a47791464dd820f443b
+size 4996831800

model-00005-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b69d3d760e43b9c3b90be3e135cde5a94b291313d8475224dca5d852ffdc5933
+size 3070130352