Update tiny models for Idefics3ForConditionalGeneration

.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 +1,5 @@
+{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>
+
+'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{{ '<|start_header_id|>assistant<|end_header_id|>
+
+' }}

config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "Idefics3ForConditionalGeneration"
+  ],
+  "dtype": "float32",
+  "image_token_id": 57,
+  "model_type": "idefics3",
+  "pad_token_id": 128002,
+  "scale_factor": 2,
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "bos_token_id": 128000,
+    "eos_token_id": 128001,
+    "head_dim": 32,
+    "hidden_act": "silu",
+    "hidden_size": 64,
+    "image_token_id": 57,
+    "initializer_range": 0.02,
+    "intermediate_size": 56,
+    "max_position_embeddings": 256,
+    "mlp_bias": false,
+    "model_type": "llama",
+    "num_attention_heads": 2,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 2,
+    "pad_token_id": 128002,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "rope_theta": 10000.0,
+      "rope_type": "default"
+    },
+    "sliding_window": 32,
+    "tie_word_embeddings": false,
+    "use_cache": true,
+    "vocab_size": 128259
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": false,
+  "vision_config": {
+    "attention_dropout": 0.1,
+    "dropout": 0.1,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 32,
+    "image_size": 16,
+    "initializer_range": 0.02,
+    "intermediate_size": 32,
+    "layer_norm_eps": 1e-06,
+    "model_type": "idefics3_vision",
+    "num_attention_heads": 4,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "patch_size": 4
+  },
+  "vocab_size": 128259
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 128000,
+  "eos_token_id": 128001,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "pad_token_id": 128002,
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a38fc0d1e15d898c8d94e99bcc75a83957892da6096a236a8d0244bb34ca93d3
+size 65987112

preprocessor_config.json

@@ -0,0 +1,32 @@
+{
+  "crop_size": {
+    "longest_edge": 16,
+    "shortest_edge": 16
+  },
+  "do_convert_rgb": true,
+  "do_image_splitting": true,
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "Idefics3ImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "max_image_size": {
+    "longest_edge": 364
+  },
+  "resample": 1,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 16,
+    "shortest_edge": 16
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1b5ea123f25178b4d498f9fd5dafe3e1356c7d179ffeb792659eaee8972b8a6
+size 17210501

tokenizer_config.json

@@ -0,0 +1,24 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|begin_of_text|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|end_of_text|>",
+  "extra_special_tokens": [
+    "<fake_token_around_image>",
+    "<image>",
+    "<end_of_utterance>"
+  ],
+  "is_local": true,
+  "legacy": false,
+  "local_files_only": false,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 131072,
+  "pad_token": "<|reserved_special_token_0|>",
+  "padding_side": "left",
+  "processor_class": "Idefics3Processor",
+  "tokenizer_class": "TokenizersBackend",
+  "truncation_side": "left"
+}