rubentium commited on
Commit
2005d11
·
verified ·
1 Parent(s): 6d1b435

Upload paper checkpoint l8k365m_dense_ropectl_a1p0_seed44 checkpoint-44000

Browse files
checkpoints/control_rope_365m/l8k365m_dense_ropectl_a1p0_seed44/checkpoint-44000/config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": [
18
+ 1.0,
19
+ 0.9896339488176222,
20
+ 0.9692159692159692,
21
+ 0.9420289855072463,
22
+ 0.8730158730158729
23
+ ],
24
+ "attention_bias": false,
25
+ "attention_dropout": 0.0,
26
+ "bos_token_id": 151643,
27
+ "eos_token_id": 151643,
28
+ "checkpoint_activations": false,
29
+ "attention_type": "sparse_stochastic",
30
+ "sparse_block_size": 128,
31
+ "sparse_num_random_blocks": 31,
32
+ "sparse_mask_pool_size": 1,
33
+ "sparse_max_seq_length": 8192,
34
+ "sparse_layer_indices": null,
35
+ "sparse_mask_cache_dir": null,
36
+ "sparse_mask_seed": 0,
37
+ "sparse_mixed_nrb_set": null,
38
+ "moba_block_size": 128,
39
+ "moba_top_k": 32,
40
+ "asentmax_alpha": 1.5,
41
+ "asentmax_learn_temp": true,
42
+ "asentmax_init_temp": 1.0
43
+ }
checkpoints/control_rope_365m/l8k365m_dense_ropectl_a1p0_seed44/checkpoint-44000/loader_gen_state_rank0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4050e1c23285b5ab02806005252fb5c5d57cc84365951a8d1f8caf328614ee38
3
+ size 6802
checkpoints/control_rope_365m/l8k365m_dense_ropectl_a1p0_seed44/checkpoint-44000/loader_gen_state_rank1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:741d13a2496875c23c79af48aa5ac707c8afbf56f135bf998d00b7373ad25ea9
3
+ size 6802
checkpoints/control_rope_365m/l8k365m_dense_ropectl_a1p0_seed44/checkpoint-44000/model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef7b4b4e8d501cabba3b681db531d0af2fd7bcb236060a0313f05e63c137522e
3
+ size 718031899
checkpoints/control_rope_365m/l8k365m_dense_ropectl_a1p0_seed44/checkpoint-44000/rng_state_rank0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc3d0e8e55e0a0e43312c684dc3b31403c0343063c977db9b943a47c0f1504cb
3
+ size 14821
checkpoints/control_rope_365m/l8k365m_dense_ropectl_a1p0_seed44/checkpoint-44000/rng_state_rank1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fd589f7a24f24bd89effb957ca599772ce85cd8cc5726f7b88604be3c410270
3
+ size 14821
checkpoints/control_rope_365m/l8k365m_dense_ropectl_a1p0_seed44/checkpoint-44000/training_state.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0172ea73f73b7252c9f04f5f046da95eb47209a00b4ce9eeb4ce7c62e934067b
3
+ size 1436105299