Upload M0_raw_tpfn_s1/config.json with huggingface_hub
Browse files- M0_raw_tpfn_s1/config.json +51 -0
M0_raw_tpfn_s1/config.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_name": "M0_raw_tpfn_s1",
|
| 3 |
+
"tokenizer": "fixed",
|
| 4 |
+
"patch": 32,
|
| 5 |
+
"ctx_span": 4096,
|
| 6 |
+
"fine_span": 1024,
|
| 7 |
+
"enc_hidden": 64,
|
| 8 |
+
"fsq_levels": [
|
| 9 |
+
8,
|
| 10 |
+
6,
|
| 11 |
+
5,
|
| 12 |
+
5,
|
| 13 |
+
5
|
| 14 |
+
],
|
| 15 |
+
"vq_codes": 1024,
|
| 16 |
+
"vq_dim": 64,
|
| 17 |
+
"vq_beta": 0.25,
|
| 18 |
+
"vq_decay": 0.99,
|
| 19 |
+
"vq_entropy_w": 0.0,
|
| 20 |
+
"vq_warm_steps": 2000,
|
| 21 |
+
"vq_warm_decay": 0.9,
|
| 22 |
+
"aux_code_ce": 0.0,
|
| 23 |
+
"d_model": 256,
|
| 24 |
+
"n_layers": 4,
|
| 25 |
+
"n_heads": 4,
|
| 26 |
+
"train_kmin": 4,
|
| 27 |
+
"train_kmax": 32,
|
| 28 |
+
"cpm_cmax": 16,
|
| 29 |
+
"cpm_pmax": 0.4,
|
| 30 |
+
"cpm_span_max": 4,
|
| 31 |
+
"steps": 30000,
|
| 32 |
+
"batch": 64,
|
| 33 |
+
"n_variates": 8,
|
| 34 |
+
"lr": 0.001,
|
| 35 |
+
"normuon_lr": 0.0008,
|
| 36 |
+
"weight_decay": 0.0001,
|
| 37 |
+
"grad_clip": 1.0,
|
| 38 |
+
"warmup_frac": 0.03,
|
| 39 |
+
"decay_frac": 0.1,
|
| 40 |
+
"seed": 1,
|
| 41 |
+
"data_seed": 1,
|
| 42 |
+
"pool": "short",
|
| 43 |
+
"eval_every": 500,
|
| 44 |
+
"ckpt_every": 2500,
|
| 45 |
+
"gift_ckpts": [
|
| 46 |
+
7500,
|
| 47 |
+
15000,
|
| 48 |
+
22500,
|
| 49 |
+
30000
|
| 50 |
+
]
|
| 51 |
+
}
|