Commit
•
19a5424
1
Parent(s):
5be30ef
Upload 5 files
Browse files- adapter_config.json +26 -0
- adapter_model.bin +3 -0
- config.json +39 -0
- non_lora_trainables.bin +3 -0
- trainer_state.json +0 -0
adapter_config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"auto_mapping": null,
|
3 |
+
"base_model_name_or_path": "/pfss/mlde/workspaces/mlde_wsp_Ramstedt_Mila/hf/vicuna-7b",
|
4 |
+
"bias": "none",
|
5 |
+
"fan_in_fan_out": false,
|
6 |
+
"inference_mode": true,
|
7 |
+
"init_lora_weights": true,
|
8 |
+
"layers_pattern": null,
|
9 |
+
"layers_to_transform": null,
|
10 |
+
"lora_alpha": 256,
|
11 |
+
"lora_dropout": 0.05,
|
12 |
+
"modules_to_save": null,
|
13 |
+
"peft_type": "LORA",
|
14 |
+
"r": 128,
|
15 |
+
"revision": null,
|
16 |
+
"target_modules": [
|
17 |
+
"q_proj",
|
18 |
+
"o_proj",
|
19 |
+
"gate_proj",
|
20 |
+
"up_proj",
|
21 |
+
"down_proj",
|
22 |
+
"k_proj",
|
23 |
+
"v_proj"
|
24 |
+
],
|
25 |
+
"task_type": "CAUSAL_LM"
|
26 |
+
}
|
adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6557ef5f3fb2d691e216faf165c7b6c7211f0ad28eb2fb9c86510034b037415c
|
3 |
+
size 639786637
|
config.json
ADDED
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/pfss/mlde/workspaces/mlde_wsp_Ramstedt_Mila/hf/vicuna-7b",
|
3 |
+
"architectures": [
|
4 |
+
"LlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"bos_token_id": 1,
|
7 |
+
"eos_token_id": 2,
|
8 |
+
"freeze_mm_mlp_adapter": false,
|
9 |
+
"hidden_act": "silu",
|
10 |
+
"hidden_size": 4096,
|
11 |
+
"image_aspect_ratio": "pad",
|
12 |
+
"image_grid_pinpoints": null,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 11008,
|
15 |
+
"max_position_embeddings": 4096,
|
16 |
+
"mm_hidden_size": 1024,
|
17 |
+
"mm_projector_lr": 2e-05,
|
18 |
+
"mm_projector_type": "mlp2x_gelu",
|
19 |
+
"mm_use_im_patch_token": false,
|
20 |
+
"mm_use_im_start_end": false,
|
21 |
+
"mm_vision_select_feature": "patch",
|
22 |
+
"mm_vision_select_layer": -2,
|
23 |
+
"mm_vision_tower": "/pfss/mlde/workspaces/mlde_wsp_Ramstedt_Mila/hf/clip",
|
24 |
+
"model_type": "llava",
|
25 |
+
"num_attention_heads": 32,
|
26 |
+
"num_hidden_layers": 32,
|
27 |
+
"num_key_value_heads": 32,
|
28 |
+
"pad_token_id": 0,
|
29 |
+
"pretraining_tp": 1,
|
30 |
+
"rms_norm_eps": 1e-05,
|
31 |
+
"rope_scaling": null,
|
32 |
+
"tie_word_embeddings": false,
|
33 |
+
"torch_dtype": "float16",
|
34 |
+
"transformers_version": "4.31.0",
|
35 |
+
"tune_mm_mlp_adapter": false,
|
36 |
+
"use_cache": true,
|
37 |
+
"use_mm_proj": true,
|
38 |
+
"vocab_size": 32000
|
39 |
+
}
|
non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5e3eb1f37f123bab2f88d5d15ded864b58579b83b94af26fc4aa2700a2a1d1e8
|
3 |
+
size 649126599
|
trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|