davidquarel commited on
Commit
3c74870
·
verified ·
1 Parent(s): d223e3f

Upload folder using huggingface_hub

Browse files
antban2_al_0.6_g_0.98_id_77_seed_980677/action_logprobs.pth.gz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a39e6010b5b01cab77ae6e67fb79eedfe2dcad3981a2fe5d7622e45ec7309606
3
+ size 221527589
antban2_al_0.6_g_0.98_id_77_seed_980677/action_probs.tar.gz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ded8df77405e5a8aa4857b0fec446b712d9c9ec8553be1907968637249b6bbd4
3
+ size 108219830
antban2_al_0.6_g_0.98_id_77_seed_980677/checkpoints.tar.gz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72d4cc88dfaf80bd1e1a557d3f23a87f5235cd81209e6aa16b718dde4384b9f9
3
+ size 1538358711
antban2_al_0.6_g_0.98_id_77_seed_980677/config.cfg ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ rl_action=train
2
+ model_type=impala
3
+ lr=5e-05
4
+ discount_rate=0.98
5
+ num_rollout_steps=64
6
+ grad_acc_per_chunk=5
7
+ num_rollout_chunks=1
8
+ cheese_loc=any
9
+ env_layout=open
10
+ alpha=0.6
11
+ env_size=13
12
+ num_levels=9600
13
+ compile=True
14
+ use_prev_action=False
15
+ weight_restrictions=None
16
+ weight_restrictions_invert=False
17
+ use_bf16=False
18
+ use_wandb=True
19
+ no_tqdm=True
20
+ seed=980677
21
+ mask_type=first_episode
22
+ ckpt_dir=jaxgmg2_antennaban
23
+ vis_average_state=False
24
+ trim_episodes=False
25
+ num_total_env_steps=9999974400
26
+ eval_every=1
27
+ eff_horizon=None
28
+ optim=adam
29
+ env_rule=ban:c[2,1], m[1,:]; c[1,2], m[:,1]
30
+ env_rule_mixture=None
31
+ hf_user=davidquarel
32
+ hf_collection=davidquarel/jaxgmg
33
+ use_hf=True
34
+ num_hf_uploads=1
35
+ use_log=True
36
+ log_optimizer_state=True
37
+ resume=None
38
+ resume_id=None
39
+ resume_optim=False
40
+ checkpoint=antban2_al_0.6_g_0.98_id_77_seed_980677
41
+ wandb_project=jaxgmg2_antennaban
42
+ eval_schedule=0:1,250:2,500:5,2000:10
43
+ render_sixel=False
44
+ sixel_idx=60
45
+ live_monitor=False
46
+ run_id=77
47
+ seed_formula=
48
+ deterministic=True
49
+ penalize_time=False
50
+ f_str_ckpt=antban2_al_0.6_g_0.98_id_77_seed_980677
51
+ duplication_factor=-1
52
+ smoke=False
53
+ ntfy=david_jaxgmg
54
+ num_chains=6
55
+ num_draws=3000
56
+ num_steps_bw_draws=1
57
+ on_policy=True
58
+ llc_nbeta=3000
59
+ localization=10
60
+ exact_solver_each_draw=False
61
+ llc_optimizer=sgld
62
+ iw_clip_eps=None
63
+ rmsprop_burnin_steps=20
64
+ llc_data_file=llc_scan_open_reinforce.pkl
65
+ llc_checkpoint_index=None
66
+ llc_checkpoint_number=None
67
+ sink=None
68
+ repo_id=davidquarel/jaxgmg_ckpt_zip
69
+ use_shuffled_checkpoints=False
70
+ force_re_download=False
71
+ off_distribution_data=False
72
+ evaluate_every_position=False
73
+ num_prev_actions=1
74
+ eff_acc_steps=5
75
+ chunk_size=9600
76
+ env_steps_per_microbatch=122880
77
+ ckpt_path=jaxgmg2_antennaban/antban2_al_0.6_g_0.98_id_77_seed_980677
78
+ env_steps_per_loop=614400
79
+ total_loops=16276
antban2_al_0.6_g_0.98_id_77_seed_980677/config.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha": 0.6,
3
+ "checkpoint": "antban2_al_0.6_g_0.98_id_77_seed_980677",
4
+ "cheese_loc": "any",
5
+ "chunk_size": 9600,
6
+ "ckpt_dir": "jaxgmg2_antennaban",
7
+ "ckpt_path": "jaxgmg2_antennaban/antban2_al_0.6_g_0.98_id_77_seed_980677",
8
+ "compile": true,
9
+ "deterministic": true,
10
+ "discount_rate": 0.98,
11
+ "duplication_factor": -1,
12
+ "eff_acc_steps": 5,
13
+ "eff_horizon": null,
14
+ "env_layout": "open",
15
+ "env_rule": "ban:c[2,1], m[1,:]; c[1,2], m[:,1]",
16
+ "env_rule_mixture": null,
17
+ "env_size": 13,
18
+ "env_steps_per_loop": 614400,
19
+ "env_steps_per_microbatch": 122880,
20
+ "eval_every": 1,
21
+ "eval_schedule": "0:1,250:2,500:5,2000:10",
22
+ "evaluate_every_position": false,
23
+ "exact_solver_each_draw": false,
24
+ "f_str_ckpt": "antban2_al_0.6_g_0.98_id_77_seed_980677",
25
+ "force_re_download": false,
26
+ "grad_acc_per_chunk": 5,
27
+ "hf_collection": "davidquarel/jaxgmg",
28
+ "hf_user": "davidquarel",
29
+ "iw_clip_eps": null,
30
+ "live_monitor": false,
31
+ "llc_checkpoint_index": null,
32
+ "llc_checkpoint_number": null,
33
+ "llc_data_file": "llc_scan_open_reinforce.pkl",
34
+ "llc_nbeta": 3000,
35
+ "llc_optimizer": "sgld",
36
+ "localization": 10,
37
+ "log_optimizer_state": true,
38
+ "lr": 5e-05,
39
+ "mask_type": "first_episode",
40
+ "model_type": "impala",
41
+ "no_tqdm": true,
42
+ "ntfy": "david_jaxgmg",
43
+ "num_chains": 6,
44
+ "num_draws": 3000,
45
+ "num_hf_uploads": 1,
46
+ "num_levels": 9600,
47
+ "num_prev_actions": 1,
48
+ "num_rollout_chunks": 1,
49
+ "num_rollout_steps": 64,
50
+ "num_steps_bw_draws": 1,
51
+ "num_total_env_steps": 9999974400,
52
+ "off_distribution_data": false,
53
+ "on_policy": true,
54
+ "optim": "adam",
55
+ "penalize_time": false,
56
+ "render_sixel": false,
57
+ "repo_id": "davidquarel/jaxgmg_ckpt_zip",
58
+ "resume": null,
59
+ "resume_id": null,
60
+ "resume_optim": false,
61
+ "rl_action": "train",
62
+ "rmsprop_burnin_steps": 20,
63
+ "run_id": 77,
64
+ "seed": 980677,
65
+ "seed_formula": "",
66
+ "sink": null,
67
+ "sixel_idx": 60,
68
+ "smoke": false,
69
+ "total_loops": 16276,
70
+ "trim_episodes": false,
71
+ "use_bf16": false,
72
+ "use_hf": true,
73
+ "use_log": true,
74
+ "use_prev_action": false,
75
+ "use_shuffled_checkpoints": false,
76
+ "use_wandb": true,
77
+ "vis_average_state": false,
78
+ "wandb_project": "jaxgmg2_antennaban",
79
+ "weight_restrictions": null,
80
+ "weight_restrictions_invert": false
81
+ }
antban2_al_0.6_g_0.98_id_77_seed_980677/eval.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
antban2_al_0.6_g_0.98_id_77_seed_980677/eval.log ADDED
The diff for this file is too large to render. See raw diff
 
antban2_al_0.6_g_0.98_id_77_seed_980677/latest_train.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "ts": "2026-03-20T16:03:26.589003Z",
3
+ "loop": 16275,
4
+ "env_steps": 9999974399,
5
+ "loss": 0.016029546037316322,
6
+ "avg_return": 0.8638240694999695,
7
+ "regret": 0.0003265374689362943,
8
+ "entropy": 0.08034701645374298,
9
+ "lr": 5e-05,
10
+ "discount_rate": 0.98,
11
+ "num_rollout_steps": 64
12
+ }
antban2_al_0.6_g_0.98_id_77_seed_980677/lvl_rngs.npy ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0664ecfd418afa0e7963b1c54f4b1e49800577e0b6fa98f662f7b242c3d14751
3
+ size 130336
antban2_al_0.6_g_0.98_id_77_seed_980677/optimizer_state.tar.gz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b01357d29eda3a3eece3828ffa27725caa36c751b3522ff077a69a0ac1aba15
3
+ size 2552162555
antban2_al_0.6_g_0.98_id_77_seed_980677/rng_keys.tar.gz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77a27a623aa6b8128e32b58ae4bd39f5ea2f06d8886ca469e9336681f84a54d6
3
+ size 48921
antban2_al_0.6_g_0.98_id_77_seed_980677/setup.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "config": {
3
+ "rl_action": "train",
4
+ "model_type": "impala",
5
+ "lr": 5e-05,
6
+ "discount_rate": 0.98,
7
+ "num_rollout_steps": 64,
8
+ "grad_acc_per_chunk": 5,
9
+ "num_rollout_chunks": 1,
10
+ "cheese_loc": "any",
11
+ "env_layout": "open",
12
+ "alpha": 0.6,
13
+ "env_size": 13,
14
+ "num_levels": 9600,
15
+ "compile": true,
16
+ "use_prev_action": false,
17
+ "weight_restrictions": null,
18
+ "weight_restrictions_invert": false,
19
+ "use_bf16": false,
20
+ "use_wandb": true,
21
+ "no_tqdm": true,
22
+ "seed": 980677,
23
+ "mask_type": "first_episode",
24
+ "ckpt_dir": "jaxgmg2_antennaban",
25
+ "vis_average_state": false,
26
+ "trim_episodes": false,
27
+ "num_total_env_steps": 9999974400,
28
+ "eval_every": 1,
29
+ "eff_horizon": null,
30
+ "optim": "adam",
31
+ "env_rule": "ban:c[2,1], m[1,:]; c[1,2], m[:,1]",
32
+ "env_rule_mixture": null,
33
+ "hf_user": "davidquarel",
34
+ "hf_collection": "davidquarel/jaxgmg",
35
+ "use_hf": true,
36
+ "num_hf_uploads": 1,
37
+ "use_log": true,
38
+ "log_optimizer_state": true,
39
+ "resume": null,
40
+ "resume_id": null,
41
+ "resume_optim": false,
42
+ "checkpoint": "antban2_al_0.6_g_0.98_id_77_seed_980677",
43
+ "wandb_project": "jaxgmg2_antennaban",
44
+ "eval_schedule": "0:1,250:2,500:5,2000:10",
45
+ "render_sixel": false,
46
+ "sixel_idx": 60,
47
+ "live_monitor": false,
48
+ "run_id": 77,
49
+ "seed_formula": "",
50
+ "deterministic": true,
51
+ "penalize_time": false,
52
+ "f_str_ckpt": "antban2_al_0.6_g_0.98_id_77_seed_980677",
53
+ "duplication_factor": -1,
54
+ "smoke": false,
55
+ "ntfy": "david_jaxgmg",
56
+ "num_chains": 6,
57
+ "num_draws": 3000,
58
+ "num_steps_bw_draws": 1,
59
+ "on_policy": true,
60
+ "llc_nbeta": 3000,
61
+ "localization": 10,
62
+ "exact_solver_each_draw": false,
63
+ "llc_optimizer": "sgld",
64
+ "iw_clip_eps": null,
65
+ "rmsprop_burnin_steps": 20,
66
+ "llc_data_file": "llc_scan_open_reinforce.pkl",
67
+ "llc_checkpoint_index": null,
68
+ "llc_checkpoint_number": null,
69
+ "sink": null,
70
+ "repo_id": "davidquarel/jaxgmg_ckpt_zip",
71
+ "use_shuffled_checkpoints": false,
72
+ "force_re_download": false,
73
+ "off_distribution_data": false,
74
+ "evaluate_every_position": false,
75
+ "num_prev_actions": 1,
76
+ "eff_acc_steps": 5,
77
+ "chunk_size": 9600,
78
+ "env_steps_per_microbatch": 122880,
79
+ "ckpt_path": "jaxgmg2_antennaban/antban2_al_0.6_g_0.98_id_77_seed_980677",
80
+ "env_steps_per_loop": 614400,
81
+ "total_loops": 16276
82
+ },
83
+ "setup": {
84
+ "start_time_utc": "2026-03-20T11:59:06.092301Z",
85
+ "seed": 980677,
86
+ "device": "cuda",
87
+ "python_version": "3.11.11",
88
+ "torch_version": "2.10.0+cu128",
89
+ "jax_version": "0.6.2",
90
+ "hostname": "36605bd7f66d",
91
+ "platform": "Linux-6.5.0-35-generic-x86_64-with-glibc2.39",
92
+ "git_sha": null,
93
+ "git_branch": null,
94
+ "use_wandb": true
95
+ }
96
+ }
antban2_al_0.6_g_0.98_id_77_seed_980677/setup.txt ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ Start: 2026-03-20T11:59:06.092301Z
2
+ Seed: 980677
3
+ Device: cuda
4
+ Python: 3.11.11 | Torch: 2.10.0+cu128 | JAX: 0.6.2
5
+ Host: 36605bd7f66d
6
+ Platform: Linux-6.5.0-35-generic-x86_64-with-glibc2.39
7
+ Git: branch=None sha=None
8
+ use_wandb: True
antban2_al_0.6_g_0.98_id_77_seed_980677/state_counts.npy ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da8efec3c30f2045ac066c05e89923c4ebe602f52de09427ad9312a26585f238
3
+ size 116288
antban2_al_0.6_g_0.98_id_77_seed_980677/train.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
antban2_al_0.6_g_0.98_id_77_seed_980677/train.log ADDED
The diff for this file is too large to render. See raw diff