Upload paper checkpoint l8k365m_dense_seed47 checkpoint-44000
Browse files- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/config.json +37 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank0.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank1.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank2.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank3.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank4.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/model.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank0.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank1.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank2.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank3.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank4.pt +3 -0
- checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/training_state.pt +3 -0
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vocab_size": 151936,
|
| 3 |
+
"hidden_size": 768,
|
| 4 |
+
"intermediate_size": 3072,
|
| 5 |
+
"num_hidden_layers": 28,
|
| 6 |
+
"num_attention_heads": 6,
|
| 7 |
+
"num_key_value_heads": 2,
|
| 8 |
+
"head_dim": 128,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"max_position_embeddings": 32768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"rms_norm_eps": 1e-06,
|
| 13 |
+
"use_cache": true,
|
| 14 |
+
"tie_word_embeddings": true,
|
| 15 |
+
"rope_theta": 1000000,
|
| 16 |
+
"rope_scaling": null,
|
| 17 |
+
"rope_train_position_scale_set": null,
|
| 18 |
+
"attention_bias": false,
|
| 19 |
+
"attention_dropout": 0.0,
|
| 20 |
+
"bos_token_id": 151643,
|
| 21 |
+
"eos_token_id": 151643,
|
| 22 |
+
"checkpoint_activations": false,
|
| 23 |
+
"attention_type": "sparse_stochastic",
|
| 24 |
+
"sparse_block_size": 128,
|
| 25 |
+
"sparse_num_random_blocks": 31,
|
| 26 |
+
"sparse_mask_pool_size": 1,
|
| 27 |
+
"sparse_max_seq_length": 8192,
|
| 28 |
+
"sparse_layer_indices": null,
|
| 29 |
+
"sparse_mask_cache_dir": null,
|
| 30 |
+
"sparse_mask_seed": 0,
|
| 31 |
+
"sparse_mixed_nrb_set": null,
|
| 32 |
+
"moba_block_size": 128,
|
| 33 |
+
"moba_top_k": 32,
|
| 34 |
+
"asentmax_alpha": 1.5,
|
| 35 |
+
"asentmax_learn_temp": true,
|
| 36 |
+
"asentmax_init_temp": 1.0
|
| 37 |
+
}
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank0.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d29ebcba1740400d02a9731a1b2ce2d06da520e9a5f365770ae3366476cfa6bb
|
| 3 |
+
size 6802
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank1.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6cbcc77de4742713528531e0ef8cdd89a1973c9e431acb1771b42ca22d7bd1f
|
| 3 |
+
size 6802
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank2.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9db5ff6feccc6d1a9e53c22c2c117265883048de542cfdb5f1d72684ce40e97f
|
| 3 |
+
size 6802
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank3.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2448f0fdf688f2e5aca811502f056dcbab182fad72e69178143577736b6f0c8e
|
| 3 |
+
size 6802
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/loader_gen_state_rank4.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e5371f5b80aa7358ef9290ec1864b1c0669c69fb1e0f3f24dfe821940c3a22a
|
| 3 |
+
size 6802
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/model.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fcbc46b085c405a7e6e03207ec616a98329e7104851e140bee1604e516c39ba5
|
| 3 |
+
size 718031899
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank0.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f788c6d59c14d7c16ea7debfe678626698ba4365ccb7298c917ea721473da31d
|
| 3 |
+
size 14757
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank1.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a34d85db33c14d0607eafaf6bd76b96ce7f65d4db9736ce37ee257226b67eb9f
|
| 3 |
+
size 14757
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank2.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e96a39c808efb86651ff4cf83ff1147dfa341b158d21a6defc7f33543970b46
|
| 3 |
+
size 14885
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank3.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:499039a66a3ffcd6e98373202df938050e1153d0c844e9b97c15d2b2e8b021fd
|
| 3 |
+
size 14821
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/rng_state_rank4.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ef8df31e0b5c227a738d3f9369ffa4dda6b0625ef6a72945b60ba9ff28b4047
|
| 3 |
+
size 14821
|
checkpoints/primary_365m/l8k365m_dense_seed47/checkpoint-44000/training_state.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fe64a3649d808af09b395616da85fcf6828bca68a76908519e210cb2c379e2d
|
| 3 |
+
size 1436105299
|