Upload folder using huggingface_hub

README.md

@@ -0,0 +1,26 @@
+---
+library_name: keras-nlp
+pipeline_tag: text-generation
+---
+This is a [`Phi3` model](https://keras.io/api/keras_nlp/models/phi3) uploaded using the KerasNLP library and can be used with JAX, TensorFlow, and PyTorch backends.
+This model is related to a `CausalLM` task.
+
+Model config:
+* **name:** phi3_backbone_1
+* **trainable:** True
+* **vocabulary_size:** 32064
+* **num_layers:** 32
+* **num_query_heads:** 32
+* **hidden_dim:** 3072
+* **intermediate_dim:** 8192
+* **num_key_value_heads:** 32
+* **layer_norm_epsilon:** 1e-05
+* **dropout:** 0.0
+* **max_sequence_length:** 4096
+* **pretraining_sequence_length:** 4096
+* **rope_max_wavelength:** 10000.0
+* **rope_scaling_type:** None
+* **rope_scaling_short_factor:** None
+* **rope_scaling_long_factor:** None
+
+This model card has been generated automatically and should be completed by the model author. See [Model Cards documentation](https://huggingface.co/docs/hub/model-cards) for more information.

config.json

@@ -0,0 +1,23 @@
+{
+    "module": "keras_nlp.src.models.phi3.phi3_backbone",
+    "class_name": "Phi3Backbone",
+    "config": {
+        "name": "phi3_backbone_1",
+        "trainable": true,
+        "vocabulary_size": 32064,
+        "num_layers": 32,
+        "num_query_heads": 32,
+        "hidden_dim": 3072,
+        "intermediate_dim": 8192,
+        "num_key_value_heads": 32,
+        "layer_norm_epsilon": 1e-05,
+        "dropout": 0.0,
+        "max_sequence_length": 4096,
+        "pretraining_sequence_length": 4096,
+        "rope_max_wavelength": 10000.0,
+        "rope_scaling_type": null,
+        "rope_scaling_short_factor": null,
+        "rope_scaling_long_factor": null
+    },
+    "registered_name": "keras_nlp>Phi3Backbone"
+}

metadata.json

@@ -0,0 +1,6 @@
+{
+    "keras_version": "3.4.1",
+    "keras_nlp_version": "0.14.0",
+    "parameter_count": 3821079552,
+    "date_saved": "2024-07-26@02:34:09"
+}

model.weights.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2bf084f5e43a263be8bf03751088b2532ebb306bb81ffc36908a598ff00ec7b
+size 15285416504

preprocessor.json

@@ -0,0 +1,39 @@
+{
+    "module": "keras_nlp.src.models.phi3.phi3_causal_lm_preprocessor",
+    "class_name": "Phi3CausalLMPreprocessor",
+    "config": {
+        "name": "phi3_causal_lm_preprocessor",
+        "trainable": true,
+        "dtype": {
+            "module": "keras",
+            "class_name": "DTypePolicy",
+            "config": {
+                "name": "mixed_bfloat16"
+            },
+            "registered_name": null
+        },
+        "tokenizer": {
+            "module": "keras_nlp.src.models.phi3.phi3_tokenizer",
+            "class_name": "Phi3Tokenizer",
+            "config": {
+                "name": "phi3_tokenizer",
+                "trainable": true,
+                "dtype": {
+                    "module": "keras",
+                    "class_name": "DTypePolicy",
+                    "config": {
+                        "name": "int32"
+                    },
+                    "registered_name": null
+                },
+                "proto": null,
+                "sequence_length": null
+            },
+            "registered_name": "keras_nlp>Phi3Tokenizer"
+        },
+        "sequence_length": 1024,
+        "add_start_token": true,
+        "add_end_token": false
+    },
+    "registered_name": "keras_nlp>Phi3CausalLMPreprocessor"
+}

task.json

@@ -0,0 +1,70 @@
+{
+    "module": "keras_nlp.src.models.phi3.phi3_causal_lm",
+    "class_name": "Phi3CausalLM",
+    "config": {
+        "backbone": {
+            "module": "keras_nlp.src.models.phi3.phi3_backbone",
+            "class_name": "Phi3Backbone",
+            "config": {
+                "name": "phi3_backbone_1",
+                "trainable": true,
+                "vocabulary_size": 32064,
+                "num_layers": 32,
+                "num_query_heads": 32,
+                "hidden_dim": 3072,
+                "intermediate_dim": 8192,
+                "num_key_value_heads": 32,
+                "layer_norm_epsilon": 1e-05,
+                "dropout": 0.0,
+                "max_sequence_length": 4096,
+                "pretraining_sequence_length": 4096,
+                "rope_max_wavelength": 10000.0,
+                "rope_scaling_type": null,
+                "rope_scaling_short_factor": null,
+                "rope_scaling_long_factor": null
+            },
+            "registered_name": "keras_nlp>Phi3Backbone"
+        },
+        "preprocessor": {
+            "module": "keras_nlp.src.models.phi3.phi3_causal_lm_preprocessor",
+            "class_name": "Phi3CausalLMPreprocessor",
+            "config": {
+                "name": "phi3_causal_lm_preprocessor",
+                "trainable": true,
+                "dtype": {
+                    "module": "keras",
+                    "class_name": "DTypePolicy",
+                    "config": {
+                        "name": "mixed_bfloat16"
+                    },
+                    "registered_name": null
+                },
+                "tokenizer": {
+                    "module": "keras_nlp.src.models.phi3.phi3_tokenizer",
+                    "class_name": "Phi3Tokenizer",
+                    "config": {
+                        "name": "phi3_tokenizer",
+                        "trainable": true,
+                        "dtype": {
+                            "module": "keras",
+                            "class_name": "DTypePolicy",
+                            "config": {
+                                "name": "int32"
+                            },
+                            "registered_name": null
+                        },
+                        "proto": null,
+                        "sequence_length": null
+                    },
+                    "registered_name": "keras_nlp>Phi3Tokenizer"
+                },
+                "sequence_length": 1024,
+                "add_start_token": true,
+                "add_end_token": false
+            },
+            "registered_name": "keras_nlp>Phi3CausalLMPreprocessor"
+        },
+        "name": "phi3_causal_lm"
+    },
+    "registered_name": "keras_nlp>Phi3CausalLM"
+}

tokenizer.json

@@ -0,0 +1,19 @@
+{
+    "module": "keras_nlp.src.models.phi3.phi3_tokenizer",
+    "class_name": "Phi3Tokenizer",
+    "config": {
+        "name": "phi3_tokenizer",
+        "trainable": true,
+        "dtype": {
+            "module": "keras",
+            "class_name": "DTypePolicy",
+            "config": {
+                "name": "int32"
+            },
+            "registered_name": null
+        },
+        "proto": null,
+        "sequence_length": null
+    },
+    "registered_name": "keras_nlp>Phi3Tokenizer"
+}