Robotics
LeRobot
Safetensors
lavla
File size: 2,049 Bytes
96e1819
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
{
    "type": "lavla",
    "n_obs_steps": 1,
    "input_features": {
        "observation.state": {
            "type": "STATE",
            "shape": [
                6
            ]
        },
        "observation.images.camera1": {
            "type": "VISUAL",
            "shape": [
                3,
                480,
                640
            ]
        },
        "observation.images.camera2": {
            "type": "VISUAL",
            "shape": [
                3,
                480,
                640
            ]
        }
    },
    "output_features": {
        "action": {
            "type": "ACTION",
            "shape": [
                6
            ]
        }
    },
    "device": "cuda",
    "use_amp": false,
    "use_peft": false,
    "push_to_hub": true,
    "repo_id": "Alkatt/LAVLA_bfloat16_Test_2026_08_03",
    "private": null,
    "tags": null,
    "license": null,
    "pretrained_path": null,
    "dtype": "bfloat16",
    "s1_only": false,
    "s2_backend_type": "local",
    "s2_remote_server_ip": "0.0.0.0",
    "s2_remote_server_port": "0000",
    "s2_blocking_mode": false,
    "normalization_mapping": {
        "VISUAL": "MEAN_STD",
        "STATE": "MEAN_STD",
        "ACTION": "MEAN_STD"
    },
    "n_action_steps": 50,
    "chunk_size": 50,
    "max_action_dim": 32,
    "s1_mlp_dim": 2048,
    "s1_embedding_dim": 1024,
    "s1_num_attention_heads": 8,
    "s1_num_layers": 8,
    "s1_flow_integration_steps": 10,
    "s1_freeze_vision_encoder": true,
    "s1_efficient_net": "google/efficientnet-b0",
    "s1_add_action_pos_embeddings": true,
    "s2_vlm_model": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct",
    "s2_freeze_vlm": true,
    "optimizer_type": "linearwarmup_cosinedecay",
    "optimizer_lr": 0.0001,
    "optimizer_weight_decay": 1e-10,
    "optimzer_grad_clip_norm": 10,
    "optimizer_decay_lr": 2.5e-06,
    "optimizer_betas": [
        0.9,
        0.95
    ],
    "optimizer_eps": 1e-08,
    "learning_rate_warmup_steps": 6000,
    "learning_rate_decay_steps": 200000
}