Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

chat_template.jinja

@@ -0,0 +1,43 @@
+{%- if not date_string is defined %}
+    {%- set date_string = "26 July 2024" %}
+{%- endif %}
+
+{#- This block extracts the system message, so we can slot it into the right place. #}
+{%- if messages[0]['role'] == 'system' %}
+    {%- set system_message = messages[0]['content'] %}
+    {%- set messages = messages[1:] %}
+{%- else %}
+    {%- set system_message = "" %}
+{%- endif %}
+
+{#- System message + builtin tools #}
+{{- "<|start_header_id|>system<|end_header_id|>
+
+" }}
+{{- "Cutting Knowledge Date: December 2023
+" }}
+{{- "Today Date: " + date_string + "
+
+" }}
+{{- system_message }}
+{{- "<|eot_id|>" }}
+
+
+{%- set first_user_message = True %}
+{%- for message in messages %}
+    {%- if first_user_message and message['role'] == 'user' %}
+		{%- set first_user_message = False %}
+	    {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>
+
+<|reserved_special_token_70|><|reserved_special_token_69|><|reserved_special_token_71|>'+ message['content'].replace('<|reserved_special_token_69|>', '').lstrip() + '<|eot_id|>' }}
+	{%- else %}
+        {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>
+
+'+ message['content'] + '<|eot_id|>' }}
+	{%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|start_header_id|>assistant<|end_header_id|>
+
+' }}
+{%- endif %}

config.json

@@ -2,12 +2,79 @@
   "architectures": [
     "LlavaForConditionalGeneration"
   ],
-  "dtype": "float16",
+  "dtype": "bfloat16",
   "image_seq_length": 729,
   "image_token_index": 128077,
   "model_type": "llava",
   "multimodal_projector_bias": true,
   "projector_hidden_act": "gelu",
+  "quantization_config": {
+    "add_skip_keys": true,
+    "dequantize_fp32": true,
+    "dynamic_loss_threshold": null,
+    "group_size": 0,
+    "is_integer": true,
+    "is_training": false,
+    "modules_dtype_dict": {
+      "int8": [
+        "lm_head"
+      ]
+    },
+    "modules_quant_config": {
+      "embed_tokens_per_layer": {
+        "quantization_device": "cpu"
+      }
+    },
+    "modules_to_not_convert": [
+      "model.language_model.embed_tokens.weight",
+      "correction_coefs",
+      ".txt_in",
+      ".emb_in",
+      ".img_out",
+      ".img_in",
+      ".context_embedder",
+      ".txt_out",
+      ".condition_embedder",
+      "wte",
+      ".emb_out",
+      ".final_layer",
+      "patch_emb",
+      "prediction_coefs",
+      "embedding_projection",
+      ".t_embedder",
+      "lm_head",
+      "patch_embed",
+      ".y_embedder",
+      ".norm_out",
+      ".time_embed",
+      ".vid_out",
+      ".vid_in",
+      ".x_embedder",
+      "multi_modal_projector",
+      "lm_head.weight",
+      "time_text_embed",
+      "patch_embedding",
+      ".proj_out"
+    ],
+    "non_blocking": false,
+    "quant_conv": false,
+    "quant_embedding": false,
+    "quant_method": "sdnq",
+    "quantization_device": "cuda",
+    "quantized_matmul_dtype": null,
+    "return_device": "cpu",
+    "sdnq_version": "0.1.7",
+    "svd_rank": 32,
+    "svd_steps": 8,
+    "use_dynamic_quantization": true,
+    "use_grad_ckpt": true,
+    "use_quantized_matmul": true,
+    "use_quantized_matmul_conv": false,
+    "use_static_quantization": true,
+    "use_stochastic_rounding": false,
+    "use_svd": false,
+    "weights_dtype": "uint4"
+  },
   "text_config": {
     "_name_or_path": "meta-llama/Llama-3.1-8B-Instruct",
     "architectures": [
@@ -16,7 +83,7 @@
     "attention_bias": false,
     "attention_dropout": 0.0,
     "bos_token_id": 128000,
-    "dtype": "float16",
+    "dtype": "bfloat16",
     "eos_token_id": [
       128001,
       128008,
@@ -49,14 +116,14 @@
     "vocab_size": 128256
   },
   "tie_word_embeddings": false,
-  "transformers_version": "5.6.1",
+  "transformers_version": "5.0.0",
   "vision_config": {
     "_name_or_path": "google/siglip2-so400m-patch14-384",
     "architectures": [
       "SiglipVisionModel"
     ],
     "attention_dropout": 0.0,
-    "dtype": "float16",
+    "dtype": "bfloat16",
     "hidden_act": "gelu_pytorch_tanh",
     "hidden_size": 1152,
     "image_size": 384,

generation_config.json

@@ -9,5 +9,5 @@
   ],
   "temperature": 0.6,
   "top_p": 0.9,
-  "transformers_version": "5.6.1"
+  "transformers_version": "5.0.0"
 }

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7798c882971b8c6ab9516c2f7d78bb63fe71138cd7bc8d17a1e61ad0093fc05
+size 1050673296

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0382621bc0fde7706f41cbe5d24c70ddfa91de468e121ca9fa15181a568b5fdb
+size 1975677061

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b108a1225fcca82b741fc0251cb573d954f11c6698bf15b1b2efd7081b80260c
+size 1998676789

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6db11e8283d16df14c207be5107131018052719f9f77d62aa8163a6211f8983
+size 1264525708

processor_config.json

@@ -0,0 +1,31 @@
+{
+  "image_processor": {
+    "data_format": "channels_first",
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "SiglipImageProcessorFast",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "resample": 1,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "height": 384,
+      "width": 384
+    }
+  },
+  "image_token": "<|reserved_special_token_69|>",
+  "num_additional_image_tokens": 1,
+  "patch_size": 14,
+  "processor_class": "LlavaProcessor",
+  "vision_feature_select_strategy": "default"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|begin_of_text|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "is_local": false,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 131072,
+  "model_specific_special_tokens": {},
+  "processor_class": "LlavaProcessor",
+  "tokenizer_class": "TokenizersBackend"
+}