yifanzhang114
commited on
Commit
•
47c4735
1
Parent(s):
e30cdee
Upload 5 files
Browse files- adapter_config.json +33 -0
- adapter_model.safetensors +3 -0
- config.json +82 -0
- non_lora_trainables.bin +3 -0
- trainer_state.json +0 -0
adapter_config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layers_pattern": null,
|
10 |
+
"layers_to_transform": null,
|
11 |
+
"loftq_config": {},
|
12 |
+
"lora_alpha": 256,
|
13 |
+
"lora_dropout": 0.05,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 128,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"v_proj",
|
23 |
+
"gate_proj",
|
24 |
+
"q_proj",
|
25 |
+
"o_proj",
|
26 |
+
"up_proj",
|
27 |
+
"down_proj",
|
28 |
+
"k_proj"
|
29 |
+
],
|
30 |
+
"task_type": "CAUSAL_LM",
|
31 |
+
"use_dora": false,
|
32 |
+
"use_rslora": false
|
33 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:979b7c8121ee33f7f93853f61d6dc235ab28af1ecbd91d9aea82d50fdd99a0b6
|
3 |
+
size 671150064
|
config.json
ADDED
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
|
3 |
+
"architectures": [
|
4 |
+
"LlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 128000,
|
9 |
+
"eos_token_id": 128009,
|
10 |
+
"freeze_mm_mlp_adapter": false,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 4096,
|
13 |
+
"image_aspect_ratio": "anyres",
|
14 |
+
"image_grid_pinpoints": [
|
15 |
+
[
|
16 |
+
672,
|
17 |
+
672
|
18 |
+
],
|
19 |
+
[
|
20 |
+
336,
|
21 |
+
672
|
22 |
+
],
|
23 |
+
[
|
24 |
+
672,
|
25 |
+
336
|
26 |
+
],
|
27 |
+
[
|
28 |
+
336,
|
29 |
+
1008
|
30 |
+
],
|
31 |
+
[
|
32 |
+
1008,
|
33 |
+
336
|
34 |
+
],
|
35 |
+
[
|
36 |
+
336,
|
37 |
+
1344
|
38 |
+
],
|
39 |
+
[
|
40 |
+
1344,
|
41 |
+
336
|
42 |
+
]
|
43 |
+
],
|
44 |
+
"initializer_range": 0.02,
|
45 |
+
"intermediate_size": 14336,
|
46 |
+
"max_position_embeddings": 8192,
|
47 |
+
"mm_hidden_size": 1024,
|
48 |
+
"mm_learnable_gated": -1,
|
49 |
+
"mm_patch_merge_type": "spatial",
|
50 |
+
"mm_projector_lr": 2e-05,
|
51 |
+
"mm_projector_type": "gated",
|
52 |
+
"mm_resampler_dim": 144,
|
53 |
+
"mm_resampler_temp": 1.0,
|
54 |
+
"mm_resampler_topp": 0.95,
|
55 |
+
"mm_resampler_type": "cosine",
|
56 |
+
"mm_use_im_patch_token": false,
|
57 |
+
"mm_use_im_start_end": false,
|
58 |
+
"mm_vision_select_feature": "patch",
|
59 |
+
"mm_vision_select_layer": -2,
|
60 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
61 |
+
"model_type": "llava_llama",
|
62 |
+
"num_attention_heads": 32,
|
63 |
+
"num_hidden_layers": 32,
|
64 |
+
"num_key_value_heads": 8,
|
65 |
+
"pad_token_id": 128001,
|
66 |
+
"pretraining_tp": 1,
|
67 |
+
"rms_norm_eps": 1e-05,
|
68 |
+
"rope_scaling": null,
|
69 |
+
"rope_theta": 500000.0,
|
70 |
+
"seperator": 11,
|
71 |
+
"tie_word_embeddings": false,
|
72 |
+
"tokenizer_model_max_length": 2048,
|
73 |
+
"tokenizer_padding_side": "right",
|
74 |
+
"torch_dtype": "bfloat16",
|
75 |
+
"transformers_version": "4.37.2",
|
76 |
+
"tune_mm_mlp_adapter": false,
|
77 |
+
"use_cache": true,
|
78 |
+
"use_global_only": false,
|
79 |
+
"use_local_only": false,
|
80 |
+
"use_mm_proj": true,
|
81 |
+
"vocab_size": 128257
|
82 |
+
}
|
non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6d47e9c9ee9fe81bec1796f93a549b663425a5b24e1405a7d2ab8c29a68263e4
|
3 |
+
size 60270210
|
trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|