Upload paper checkpoint l8k365m_dense_dropout_p28_seed43 checkpoint-44000
Browse files- checkpoints/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/checkpoint-44000/config.json +37 -0
- checkpoints/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/checkpoint-44000/loader_gen_state_rank0.pt +3 -0
- checkpoints/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/checkpoint-44000/model.pt +3 -0
- checkpoints/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/checkpoint-44000/rng_state_rank0.pt +3 -0
- checkpoints/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/checkpoint-44000/training_state.pt +3 -0
checkpoints/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/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": 0.28,
|
| 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/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/checkpoint-44000/loader_gen_state_rank0.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:584ecbd23f79249da2ff9185b08ee58e2eca66c1008784abeaa2af94963ad6de
|
| 3 |
+
size 6802
|
checkpoints/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/checkpoint-44000/model.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3cbcc7812e037a2f805762e51951dd53a815de1431ec27dc7ac7a04f42e5d61b
|
| 3 |
+
size 718031899
|
checkpoints/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/checkpoint-44000/rng_state_rank0.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94aa87b61d3a073b299abb70876d6500b766c9ed292427e472f96595f50edb46
|
| 3 |
+
size 14757
|
checkpoints/control_dropout_365m/l8k365m_dense_dropout_p28_seed43/checkpoint-44000/training_state.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c17854ca1269dd0ab25649299a709d28d753cec17109bd860d873995db32cf20
|
| 3 |
+
size 1436105299
|