Upload 12 files
Browse files- nanochat_huggingface_clean/meta_000483.json +38 -0
- nanochat_huggingface_clean/model_000483.pt +3 -0
- nanochat_huggingface_clean/optim_000483_rank0.pt +3 -0
- nanochat_huggingface_clean/optim_000483_rank1.pt +3 -0
- nanochat_huggingface_clean/optim_000483_rank2.pt +3 -0
- nanochat_huggingface_clean/optim_000483_rank3.pt +3 -0
- nanochat_huggingface_clean/optim_000483_rank4.pt +3 -0
- nanochat_huggingface_clean/optim_000483_rank5.pt +3 -0
- nanochat_huggingface_clean/optim_000483_rank6.pt +3 -0
- nanochat_huggingface_clean/optim_000483_rank7.pt +3 -0
- nanochat_huggingface_clean/token_bytes.pt +3 -0
- nanochat_huggingface_clean/tokenizer.pkl +3 -0
nanochat_huggingface_clean/meta_000483.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"step": 483,
|
| 3 |
+
"val_bpb": 0.2729823272122833,
|
| 4 |
+
"model_config": {
|
| 5 |
+
"sequence_len": 2048,
|
| 6 |
+
"vocab_size": 32768,
|
| 7 |
+
"n_layer": 24,
|
| 8 |
+
"n_head": 12,
|
| 9 |
+
"n_kv_head": 12,
|
| 10 |
+
"n_embd": 1536,
|
| 11 |
+
"window_pattern": "SSSL"
|
| 12 |
+
},
|
| 13 |
+
"user_config": {
|
| 14 |
+
"run": "dummy",
|
| 15 |
+
"device_type": "",
|
| 16 |
+
"model_tag": null,
|
| 17 |
+
"model_step": null,
|
| 18 |
+
"load_optimizer": 1,
|
| 19 |
+
"num_iterations": -1,
|
| 20 |
+
"max_seq_len": null,
|
| 21 |
+
"device_batch_size": 16,
|
| 22 |
+
"total_batch_size": null,
|
| 23 |
+
"embedding_lr": null,
|
| 24 |
+
"unembedding_lr": null,
|
| 25 |
+
"matrix_lr": null,
|
| 26 |
+
"init_lr_frac": 0.8,
|
| 27 |
+
"warmup_ratio": 0.0,
|
| 28 |
+
"warmdown_ratio": 0.5,
|
| 29 |
+
"final_lr_frac": 0.0,
|
| 30 |
+
"eval_every": 200,
|
| 31 |
+
"eval_tokens": 20971520,
|
| 32 |
+
"chatcore_every": 200,
|
| 33 |
+
"chatcore_max_cat": -1,
|
| 34 |
+
"chatcore_max_sample": 24,
|
| 35 |
+
"mmlu_epochs": 3,
|
| 36 |
+
"gsm8k_epochs": 4
|
| 37 |
+
}
|
| 38 |
+
}
|
nanochat_huggingface_clean/model_000483.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f3681e18f514804caaeace90cb733b97f8ae126ed7530b50c54ff071484bf0c9
|
| 3 |
+
size 4227935530
|
nanochat_huggingface_clean/optim_000483_rank0.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebd4531456aeec52f2dccf286050fe7c2fda33b6442098eb83ad756ed4369873
|
| 3 |
+
size 717465109
|
nanochat_huggingface_clean/optim_000483_rank1.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f696da1f7916eced96c51bd0bace338a19fc62c27ac685a29081379fe4f9b889
|
| 3 |
+
size 717465109
|
nanochat_huggingface_clean/optim_000483_rank2.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7d128e6d88b0ee9aeca20ccbf954984d0b85a9b07b7fabb91cdbd819da03816
|
| 3 |
+
size 717465109
|
nanochat_huggingface_clean/optim_000483_rank3.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a9c088d574c3eb0684272255eb303a5870881e6d60659a39c0e71e7b92b7d6e
|
| 3 |
+
size 717465109
|
nanochat_huggingface_clean/optim_000483_rank4.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3078503a56d843a551ce2292453f0f6fa145f48b4fd84c6023a61e4e89692786
|
| 3 |
+
size 717465109
|
nanochat_huggingface_clean/optim_000483_rank5.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8bc3d9822f0f30025c0ec7516b87ea0f0e36d0e39d9f8c70061e1022cb8c093
|
| 3 |
+
size 717465109
|
nanochat_huggingface_clean/optim_000483_rank6.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:610b4385cbe27157b45f1f70e2f5ffec51e088c1a23f6b77b8ceadbcee0f8558
|
| 3 |
+
size 717465109
|
nanochat_huggingface_clean/optim_000483_rank7.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3225fa46d4e3ef26f7705a9d2b0724bb133b746c4d86863b0629ed356fbb6ec0
|
| 3 |
+
size 717465109
|
nanochat_huggingface_clean/token_bytes.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:398732dad177888c6f884c3abb76168a6b54fd7c3f011b2d38371587faca54b7
|
| 3 |
+
size 132649
|
nanochat_huggingface_clean/tokenizer.pkl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:387cfc082b0bee45467774fd6f1310a922ad170886a58ccddcb468f275e06a6c
|
| 3 |
+
size 412105
|