initial commit

README.md

@@ -1,3 +1,6 @@
 ---
 license: apple-ascl
+pipeline_tag: image-text-to-text
 ---
+
+This repository contains the Elva-OpenELM-270M model presented in [On Efficient Language and Vision Assistants for Visually-Situated Natural Language Understanding: What Matters in Reading and Reasoning](https://huggingface.co/papers/2406.11823).

config.json

@@ -0,0 +1,107 @@
+{
+  "_name_or_path": "apple/OpenELM-270M-Instruct",
+  "activation_fn_name": "swish",
+  "architectures": [
+    "LlavaOpenELMForCausalLM"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_openelm.OpenELMConfig",
+    "AutoModelForCausalLM": "modeling_openelm.OpenELMForCausalLM"
+  },
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "ffn_dim_divisor": 256,
+  "ffn_multipliers": [
+    0.5,
+    0.73,
+    0.97,
+    1.2,
+    1.43,
+    1.67,
+    1.9,
+    2.13,
+    2.37,
+    2.6,
+    2.83,
+    3.07,
+    3.3,
+    3.53,
+    3.77,
+    4.0
+  ],
+  "ffn_with_glu": true,
+  "freeze_mm_mlp_adapter": false,
+  "head_dim": 64,
+  "hidden_size": 1280,
+  "image_aspect_ratio": "anyres",
+  "image_grid_pinpoints": [[224, 224], [224, 448], [448, 224], [448, 448], [448, 672], [672, 448], [672, 672], [672, 896], [896, 672]],
+  "initializer_range": 0.02,
+  "max_context_length": 2048,
+  "mm_hidden_size": 768,
+  "mm_patch_merge_type": "spatial",
+  "mm_projector_lr": null,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -2,
+  "mm_vision_tower": "gwkrsrch/elva-encoder-base-patch32",
+  "mm_vision_tower_lr": null,
+  "model_dim": 1280,
+  "model_type": "llava_openelm",
+  "normalization_layer_name": "rms_norm",
+  "normalize_qk_projections": true,
+  "num_gqa_groups": 4,
+  "num_kv_heads": [
+    3,
+    3,
+    3,
+    3,
+    3,
+    4,
+    4,
+    4,
+    4,
+    4,
+    4,
+    4,
+    5,
+    5,
+    5,
+    5
+  ],
+  "num_query_heads": [
+    12,
+    12,
+    12,
+    12,
+    12,
+    16,
+    16,
+    16,
+    16,
+    16,
+    16,
+    16,
+    20,
+    20,
+    20,
+    20
+  ],
+  "num_transformer_layers": 16,
+  "qkv_multipliers": [
+    0.5,
+    1.0
+  ],
+  "rope_freq_constant": 10000,
+  "rope_max_length": 4096,
+  "share_input_output_layers": true,
+  "tokenizer_model_max_length": 2048,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.42.4",
+  "tune_mm_mlp_adapter": false,
+  "use_cache": true,
+  "use_mm_proj": true,
+  "vocab_size": 32000
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.42.4"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9252271e00e7b019ad5334773e79972bc4c238ac4bd22b7cc0c2dff461455582
+size 723261056

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<unk>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 2048,
+  "pad_token": "<unk>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}