Upload paper checkpoint l30k1_2b_resampled checkpoint-37465
Browse files- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/config.json +30 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/loader_gen_state_rank0.pt +3 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/loader_gen_state_rank1.pt +3 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/loader_gen_state_rank2.pt +3 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/loader_gen_state_rank3.pt +3 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/model.pt +3 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/rng_state_rank0.pt +3 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/rng_state_rank1.pt +3 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/rng_state_rank2.pt +3 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/rng_state_rank3.pt +3 -0
- checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/training_state.pt +3 -0
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vocab_size": 151936,
|
| 3 |
+
"hidden_size": 1536,
|
| 4 |
+
"intermediate_size": 6144,
|
| 5 |
+
"num_hidden_layers": 28,
|
| 6 |
+
"num_attention_heads": 12,
|
| 7 |
+
"num_key_value_heads": 4,
|
| 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 |
+
"attention_bias": false,
|
| 18 |
+
"attention_dropout": 0.0,
|
| 19 |
+
"bos_token_id": 151643,
|
| 20 |
+
"eos_token_id": 151643,
|
| 21 |
+
"checkpoint_activations": true,
|
| 22 |
+
"attention_type": "sparse_stochastic",
|
| 23 |
+
"sparse_block_size": 128,
|
| 24 |
+
"sparse_num_random_blocks": 120,
|
| 25 |
+
"sparse_mask_pool_size": 1000,
|
| 26 |
+
"sparse_max_seq_length": 30720,
|
| 27 |
+
"sparse_layer_indices": null,
|
| 28 |
+
"sparse_mask_cache_dir": "/mloscratch/homes/navasard/sparse/mask_cache_30k_b128_nrb120",
|
| 29 |
+
"sparse_mask_seed": 0
|
| 30 |
+
}
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/loader_gen_state_rank0.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f59a1416dd062038a7e2dbac8c330163f45326660faeb92a199f62460447539
|
| 3 |
+
size 6802
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/loader_gen_state_rank1.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9907050ee7aa7455a32aae231218aac0666dd0bc7b78294eeb2cb3adc77e63c8
|
| 3 |
+
size 6802
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/loader_gen_state_rank2.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3adc40f4c25919716a49dee06d626601e0abca351145ddf1ed45e0bbfca3ea0c
|
| 3 |
+
size 6802
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/loader_gen_state_rank3.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e32dfbaa9aad434214e9087e9315fae33fa445ebde26c320765a117f68a6465
|
| 3 |
+
size 6802
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/model.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93add5c1a6d279835d1fe44ae9378844173c7ab78932e4f2c24b295f72178d0b
|
| 3 |
+
size 2404819483
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/rng_state_rank0.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f0adc019e458997c98bf4873b0b20294a4bd94a0e976c6320c4ada461144590
|
| 3 |
+
size 14821
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/rng_state_rank1.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e97b18945786b1878a203dd8121dfa44a4a8d60fe3fd56bf95be6300773da02
|
| 3 |
+
size 14821
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/rng_state_rank2.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a6b0c63045ac40369a1a231b9070d899f79195311f0319ab5abc22cd97ec535
|
| 3 |
+
size 14821
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/rng_state_rank3.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:533ec1cebfad604c2a504b6b6e09616f9b97ecf4c01d10444688ef2007299f9a
|
| 3 |
+
size 14885
|
checkpoints/scale_1p2b_factorial/l30k1_2b_resampled/checkpoint-37465/training_state.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fa5c053e10e55cf1945fba271872cb71b7b113edd671807695b3b866c130638
|
| 3 |
+
size 4809681879
|