Upload via push_to_hf.py
Browse files- mid/d20/meta_000810.json +29 -0
- mid/d20/model_000810.pt +3 -0
- mid/d20/optim_000810_rank0.pt +3 -0
mid/d20/meta_000810.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"step": 810,
|
| 3 |
+
"val_bpb": 0.41781449284158234,
|
| 4 |
+
"model_config": {
|
| 5 |
+
"sequence_len": 2048,
|
| 6 |
+
"vocab_size": 65536,
|
| 7 |
+
"n_layer": 20,
|
| 8 |
+
"n_head": 10,
|
| 9 |
+
"n_kv_head": 10,
|
| 10 |
+
"n_embd": 1280
|
| 11 |
+
},
|
| 12 |
+
"user_config": {
|
| 13 |
+
"run": "recursive-d20",
|
| 14 |
+
"device_type": "",
|
| 15 |
+
"dtype": "bfloat16",
|
| 16 |
+
"num_iterations": -1,
|
| 17 |
+
"max_seq_len": 2048,
|
| 18 |
+
"device_batch_size": 4,
|
| 19 |
+
"unembedding_lr": 0.004,
|
| 20 |
+
"embedding_lr": 0.2,
|
| 21 |
+
"matrix_lr": 0.02,
|
| 22 |
+
"init_lr_frac": 1.0,
|
| 23 |
+
"weight_decay": 0.0,
|
| 24 |
+
"eval_every": 150,
|
| 25 |
+
"eval_tokens": 10485760,
|
| 26 |
+
"total_batch_size": 524288,
|
| 27 |
+
"dry_run": 0
|
| 28 |
+
}
|
| 29 |
+
}
|
mid/d20/model_000810.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f93b6284edaeead3bcce06f6e40b51b2095ee6fcc20d5389a1e3fba13afa7ad
|
| 3 |
+
size 1145590951
|
mid/d20/optim_000810_rank0.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8de0ca93e2edadfcec104a7ae66a96a668a0a8630926e76a93bf197462e6f1d3
|
| 3 |
+
size 289675885
|