Upload folder using huggingface_hub
Browse files- config.json +61 -0
- epoch_1.pt +3 -0
- epoch_2.pt +3 -0
- epoch_3.pt +3 -0
- examine_checkpoints.ipynb +0 -0
- final.pt +3 -0
- step_1000.pt +3 -0
- step_2000.pt +3 -0
config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"hidden_size": 4096,
|
| 3 |
+
"num_head_layers": 1,
|
| 4 |
+
"num_heads": 8,
|
| 5 |
+
"use_adaln": true,
|
| 6 |
+
"use_mask_embeddings": true,
|
| 7 |
+
"use_block_embeddings": false,
|
| 8 |
+
"learning_rate": 1e-05,
|
| 9 |
+
"weight_decay": 0.01,
|
| 10 |
+
"num_epochs": 3,
|
| 11 |
+
"max_grad_norm": 1.0,
|
| 12 |
+
"per_gpu_policy_update_prompt_batch_size": 16,
|
| 13 |
+
"group_size": 16,
|
| 14 |
+
"per_gpu_sampling_mini_batch_size": 8,
|
| 15 |
+
"per_gpu_ref_mini_batch_size": 8,
|
| 16 |
+
"per_gpu_training_mini_batch_size": 2,
|
| 17 |
+
"update_steps_per_rollouts": 2,
|
| 18 |
+
"gradient_accumulation_steps": 32,
|
| 19 |
+
"use_cached_hidden_states": true,
|
| 20 |
+
"use_gradient_checkpointing": false,
|
| 21 |
+
"use_step_probs": true,
|
| 22 |
+
"optimize_denoiser": false,
|
| 23 |
+
"denoiser_learning_rate": null,
|
| 24 |
+
"denoiser_weight_decay": null,
|
| 25 |
+
"kl_coef": 0.1,
|
| 26 |
+
"clip_range": 0.2,
|
| 27 |
+
"advantage_clip_min": null,
|
| 28 |
+
"advantage_clip_max": null,
|
| 29 |
+
"ratio_cap": null,
|
| 30 |
+
"use_ref_kl": true,
|
| 31 |
+
"reward_weights": {
|
| 32 |
+
"judge": 1.0,
|
| 33 |
+
"correctness": 1.0,
|
| 34 |
+
"lookahead": 1.0,
|
| 35 |
+
"format": 0.5
|
| 36 |
+
},
|
| 37 |
+
"alpha_values": [
|
| 38 |
+
1.0
|
| 39 |
+
],
|
| 40 |
+
"steps": 256,
|
| 41 |
+
"gen_length": 512,
|
| 42 |
+
"block_length": 64,
|
| 43 |
+
"temperature": 1.0,
|
| 44 |
+
"temp_explore": 1.0,
|
| 45 |
+
"use_sampling_new": false,
|
| 46 |
+
"mixed_precision": true,
|
| 47 |
+
"reserve_teacher_gpu": true,
|
| 48 |
+
"teacher_device_index": 7,
|
| 49 |
+
"log_interval": 1,
|
| 50 |
+
"eval_interval": 100,
|
| 51 |
+
"save_interval": 1000,
|
| 52 |
+
"use_wandb": true,
|
| 53 |
+
"wandb_project": "planner-rl-8gpu",
|
| 54 |
+
"wandb_run_name": "my-run-8gpu",
|
| 55 |
+
"log_text_samples": 4,
|
| 56 |
+
"debug_logging": false,
|
| 57 |
+
"debug_log_interval": 1,
|
| 58 |
+
"output_dir": "/accounts/projects/songmei/hengyuf/hengyu_scratch/hengyuf/ete-checkpoints/RL",
|
| 59 |
+
"resume_from_checkpoint": null,
|
| 60 |
+
"compute_logprob_tensor": false
|
| 61 |
+
}
|
epoch_1.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c3931a607db1ae7e246039d93f5d1c5a4959c2ba7853d87a7a5e37702521cdb0
|
| 3 |
+
size 20688209557
|
epoch_2.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0b22bece2ff7582168f477c782196ae794fc439477f49f4032f0b19658cdcced
|
| 3 |
+
size 20688209557
|
epoch_3.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fd595e066eb41f57b0838e30268c49b3f52d5c483f9ceb79d63905b668cb68e
|
| 3 |
+
size 20688209557
|
examine_checkpoints.ipynb
ADDED
|
File without changes
|
final.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b4b93e45627a0374aac027097fd8601a75ca121609bff632be13a77711d0cd05
|
| 3 |
+
size 20688204651
|
step_1000.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:699d013f76b3967f47ad1e20a44731f0e4af8671edc9ede92e14b755858690e1
|
| 3 |
+
size 20688211071
|
step_2000.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e5da90b3118dcc4bbc5106c04c7fabbc75ac2bde5f760b026ce2aef2b1009da
|
| 3 |
+
size 20688211071
|