Hugging Face's logo

Fsoft-AIC
/
Phi3-SigLiP-MoE

Image-Text-to-Text
Safetensors
English
Model card Files Community
12
DavidNguyen commited on
Commit
3b40827
•
1 Parent(s): aeb3948

Upload folder using huggingface_hub

Browse files
Files changed (25) hide show
  1. pretrain/checkpoint-1000/config.json +66 -0
  2. pretrain/checkpoint-1000/mm_projector.bin +3 -0
  3. pretrain/checkpoint-1200/config.json +66 -0
  4. pretrain/checkpoint-1200/mm_projector.bin +3 -0
  5. pretrain/checkpoint-1400/config.json +66 -0
  6. pretrain/checkpoint-1400/mm_projector.bin +3 -0
  7. pretrain/checkpoint-1600/config.json +66 -0
  8. pretrain/checkpoint-1600/mm_projector.bin +3 -0
  9. pretrain/checkpoint-1800/config.json +66 -0
  10. pretrain/checkpoint-1800/mm_projector.bin +3 -0
  11. pretrain/checkpoint-200/config.json +66 -0
  12. pretrain/checkpoint-200/mm_projector.bin +3 -0
  13. pretrain/checkpoint-2000/config.json +66 -0
  14. pretrain/checkpoint-2000/mm_projector.bin +3 -0
  15. pretrain/checkpoint-2181/config.json +66 -0
  16. pretrain/checkpoint-2181/mm_projector.bin +3 -0
  17. pretrain/checkpoint-400/config.json +66 -0
  18. pretrain/checkpoint-400/mm_projector.bin +3 -0
  19. pretrain/checkpoint-600/config.json +66 -0
  20. pretrain/checkpoint-600/mm_projector.bin +3 -0
  21. pretrain/checkpoint-800/config.json +66 -0
  22. pretrain/checkpoint-800/mm_projector.bin +3 -0
  23. pretrain/config.json +66 -0
  24. pretrain/mm_projector.bin +3 -0
  25. pretrain/trainer_state.json +0 -0
pretrain/checkpoint-1000/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-1000/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2ccfc95330e3c8b5e91d27971825455541daeb608a7d5a23fee0da29ff97732
3
+ size 33044536
pretrain/checkpoint-1200/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-1200/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:288b587e4af89a5f7bedb1666b7b3ab3474a72953f618b82ba3f690223fc6c9a
3
+ size 33044536
pretrain/checkpoint-1400/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-1400/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e567a40c2e94ef14490d85d8ba7dc620a515dbc6dc46372a0f9612f209730ed
3
+ size 33044536
pretrain/checkpoint-1600/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-1600/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6c0a40574da5a5d31318ab77ac32313c3e4b571ba775fc3c4decd0507f4eed5
3
+ size 33044536
pretrain/checkpoint-1800/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-1800/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e46e869862200a061f9c45ea05e71d35a428498987b72ba688455ec764c04fc
3
+ size 33044536
pretrain/checkpoint-200/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-200/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7d795130be5695049da439fac36063fb12a93c3a12b966aa7faa2241dcae5aa
3
+ size 33044536
pretrain/checkpoint-2000/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-2000/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fda5f1f1d578adbac30ba2fe7191e964aa61ea67454c3ce2f7b5849a84cf9378
3
+ size 33044536
pretrain/checkpoint-2181/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-2181/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ca3a2977d745dd85c30b1cb5f9792e498b764afb2d5bff9204727f760bdbf2c
3
+ size 33044536
pretrain/checkpoint-400/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-400/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88ab5d53438c14dd9bdb5e55d25086e026518fda92930947b2a691591b4d736f
3
+ size 33044536
pretrain/checkpoint-600/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-600/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6accbdae126d6c4aa344e02053c2b5c916c2eb77ff44b3fe39fae426d6fa29a
3
+ size 33044536
pretrain/checkpoint-800/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": false,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/checkpoint-800/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a25f44062df54dc0ff31747d6f6cdca726c6cd1fd6869688af6af11085089eb
3
+ size 33044536
pretrain/config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/cm/archive/thongdt4/toolkitmoe/checkpoints/transformers_checkpoints/phi-3-mini-4k-instruct",
3
+ "architectures": [
4
+ "Phi3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "auto_map": {
9
+ "AutoConfig": "configuration_phi3.Phi3Config",
10
+ "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
11
+ },
12
+ "balance_loss_coef": 0.0,
13
+ "bos_token_id": 1,
14
+ "clip_smoe": false,
15
+ "dropout": false,
16
+ "embd_pdrop": 0.0,
17
+ "eos_token_id": 32000,
18
+ "freeze_mm_mlp_adapter": false,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 3072,
21
+ "image_aspect_ratio": "square",
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 8192,
24
+ "local_rank": 0,
25
+ "max_position_embeddings": 4096,
26
+ "mlp_smoe": false,
27
+ "mm_hidden_size": 1152,
28
+ "mm_patch_merge_type": "flat",
29
+ "mm_projector_lr": null,
30
+ "mm_projector_type": "mlp2x_gelu",
31
+ "mm_use_im_patch_token": false,
32
+ "mm_use_im_start_end": false,
33
+ "mm_vision_select_feature": "patch",
34
+ "mm_vision_select_layer": -2,
35
+ "mm_vision_tower": "google/siglip-so400m-patch14-224",
36
+ "model_type": "llava_phi",
37
+ "moe_name": "smoe",
38
+ "num_attention_heads": 32,
39
+ "num_experts": 1,
40
+ "num_hidden_layers": 32,
41
+ "num_key_value_heads": 32,
42
+ "num_layers": 3,
43
+ "num_selected": 1,
44
+ "original_max_position_embeddings": 4096,
45
+ "pad_token_id": 32000,
46
+ "resid_pdrop": 0.0,
47
+ "rms_norm_eps": 1e-05,
48
+ "rope_scaling": null,
49
+ "rope_theta": 10000.0,
50
+ "router_z_loss_coef": 0.0,
51
+ "scales": [
52
+ 1,
53
+ 3
54
+ ],
55
+ "sliding_window": 2047,
56
+ "tie_word_embeddings": false,
57
+ "tokenizer_model_max_length": 2048,
58
+ "tokenizer_padding_side": "right",
59
+ "torch_dtype": "bfloat16",
60
+ "training": true,
61
+ "transformers_version": "4.43.2",
62
+ "tune_mm_mlp_adapter": true,
63
+ "use_cache": true,
64
+ "use_mm_proj": true,
65
+ "vocab_size": 32064
66
+ }
pretrain/mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ca3a2977d745dd85c30b1cb5f9792e498b764afb2d5bff9204727f760bdbf2c
3
+ size 33044536
pretrain/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff