Update tiny models for Idefics2ForConditionalGeneration

config.json

@@ -0,0 +1,66 @@
+{
+  "architectures": [
+    "Idefics2ForConditionalGeneration"
+  ],
+  "dtype": "float32",
+  "image_token_id": 99,
+  "model_type": "idefics2",
+  "perceiver_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "silu",
+    "hidden_size": 64,
+    "initializer_range": 0.02,
+    "model_type": "idefics2_perceiver",
+    "num_key_value_heads": 1,
+    "resampler_depth": 2,
+    "resampler_head_dim": 12,
+    "resampler_n_heads": 2,
+    "resampler_n_latents": 2,
+    "rms_norm_eps": 1e-06
+  },
+  "text_config": {
+    "attention_dropout": 0.0,
+    "bos_token_id": 1,
+    "eos_token_id": 2,
+    "head_dim": 32,
+    "hidden_act": "silu",
+    "hidden_size": 64,
+    "image_token_id": 99,
+    "initializer_range": 0.02,
+    "intermediate_size": 56,
+    "max_position_embeddings": 256,
+    "model_type": "mistral",
+    "num_attention_heads": 2,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 2,
+    "pad_token_id": 0,
+    "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": 32003
+  },
+  "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": 12,
+    "initializer_range": 0.02,
+    "intermediate_size": 32,
+    "layer_norm_eps": 1e-06,
+    "model_type": "idefics2_vision",
+    "num_attention_heads": 4,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "patch_size": 12
+  },
+  "vocab_size": 32003
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "pad_token_id": 0,
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d00f8851a089cd375aadde5a59dc7b7bd41968e7e2520bb380fd3ec8a75632d
+size 17182728

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "crop_size": {
+    "longest_edge": 12,
+    "shortest_edge": 12
+  },
+  "do_convert_rgb": true,
+  "do_image_splitting": false,
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "Idefics2ImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 12,
+    "shortest_edge": 12
+  }
+}

tokenizer_config.json

@@ -0,0 +1,20 @@
+{
+  "add_prefix_space": null,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<end_of_utterance>"
+  ],
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<unk>",
+  "processor_class": "Idefics2Processor",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}