Upload 11 files

.gitattributes

@@ -1,3 +1,7 @@
 version https://git-lfs.github.com/spec/v1
 oid sha256:8f1f58fd32af1d9afe3c8d06ebe19deef3656ab96408bbdfb46cf45e8eeb0bc3
 size 38
+Stage1/llavaphi-2.7b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
+Stage1/llavaphi-2.7b-pretrain/runs/Mar23_12-41-12_r750xa/events.out.tfevents.1711197698.r750xa filter=lfs diff=lfs merge=lfs -text
+Stage1/llavastablelm-1.6b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
+Stage1/llavastablelm-1.6b-pretrain/runs/Apr05_14-33-25_r750xa/events.out.tfevents.1712327628.r750xa filter=lfs diff=lfs merge=lfs -text

Stage1/llavaphi-2.7b-pretrain/config.json

@@ -0,0 +1,53 @@
+{
+  "_name_or_path": "/home/songtaojiang/TinyLLaVABench/phi-2",
+  "architectures": [
+    "PhiForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_phi.PhiConfig",
+    "AutoModelForCausalLM": "modeling_phi.PhiForCausalLM"
+  },
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 50256,
+  "freeze_mm_mlp_adapter": false,
+  "hidden_act": "gelu_new",
+  "hidden_size": 2560,
+  "image_aspect_ratio": "square",
+  "image_projector_type": "mlp2x_gelu",
+  "initializer_range": 0.02,
+  "intermediate_size": 10240,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 2048,
+  "mm_hidden_size": 1024,
+  "mm_image_tower": "/home/songtaojiang/MobileVLM/clip-vit-large-patch14-336",
+  "mm_projector_lr": null,
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_video_tower": null,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -2,
+  "model_type": "llava_phi",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "pad_token_id": 50295,
+  "partial_rotary_factor": 0.4,
+  "qk_layernorm": false,
+  "resid_pdrop": 0.1,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "float16",
+  "transformers_version": "4.37.0",
+  "tune_mm_mlp_adapter": true,
+  "use_cache": true,
+  "use_mm_proj": true,
+  "video_global_proj": false,
+  "video_projector_type": "linear",
+  "video_spatial_proj": false,
+  "video_temproal_proj": false,
+  "vocab_size": 51200
+}

Stage1/llavaphi-2.7b-pretrain/mm_projector.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b175aec5a7168d0477a868d7aae27fea64d072204f0274e55aaa74845cf868d
+size 18362045

Stage1/llavaphi-2.7b-pretrain/runs/Mar23_12-41-12_r750xa/events.out.tfevents.1711197698.r750xa

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c58e5c15cd9e173a9250676fae8fecdfd4daa49807995610617415fa4cd16241
+size 2294058

Stage1/llavastablelm-1.6b-pretrain/config.json

@@ -0,0 +1,47 @@
+{
+  "_name_or_path": "/home/songtaojiang/TinyLLaVABench/stablelm-2-1_6b",
+  "architectures": [
+    "StableLmForCausalLM"
+  ],
+  "bos_token_id": 100257,
+  "eos_token_id": 100257,
+  "freeze_mm_mlp_adapter": false,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "image_aspect_ratio": "square",
+  "image_projector_type": "mlp2x_gelu",
+  "initializer_range": 0.02,
+  "intermediate_size": 5632,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 4096,
+  "mm_hidden_size": 1024,
+  "mm_image_tower": "/home/songtaojiang/MobileVLM/clip-vit-large-patch14-336",
+  "mm_projector_lr": null,
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_video_tower": null,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -2,
+  "model_type": "llava_stablelm",
+  "norm_eps": 1e-05,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 32,
+  "pad_token_id": 100280,
+  "partial_rotary_factor": 0.25,
+  "rope_pct": 0.25,
+  "rope_theta": 10000,
+  "tie_word_embeddings": false,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "float16",
+  "transformers_version": "4.39.2",
+  "tune_mm_mlp_adapter": true,
+  "use_cache": true,
+  "use_mm_proj": true,
+  "use_qkv_bias": true,
+  "video_global_proj": false,
+  "video_projector_type": "linear",
+  "video_spatial_proj": false,
+  "video_temproal_proj": false,
+  "vocab_size": 100352
+}

Stage1/llavastablelm-1.6b-pretrain/mm_projector.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7b5ccec7fc405e95d464ae99c61dda98f0e3e22d1ae421ca383877c4d37a830
+size 12592829

Stage1/llavastablelm-1.6b-pretrain/runs/Apr05_14-33-25_r750xa/events.out.tfevents.1712327628.r750xa

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab281e4bf8a2c980475f899ef0359d6f09c270ffb94f0bf640d9036b2828c1f4
+size 3081128