Hugging Face's logo

DatPySci
/
pretrain-sharpen

Safetensors
Model card Files
xet
 Community
DatPySci commited on
Commit
71a5ffc
·
verified ·
1 Parent(s): 88b0e54

upload pretrain 1B

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/extra_state_world_size_4_rank_0.pt +3 -0
  2. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/extra_state_world_size_4_rank_1.pt +3 -0
  3. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/extra_state_world_size_4_rank_2.pt +3 -0
  4. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/extra_state_world_size_4_rank_3.pt +3 -0
  5. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/fsdp_config.json +4 -0
  6. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/config.json +28 -0
  7. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/generation_config.json +6 -0
  8. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/special_tokens_map.json +24 -0
  9. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/tokenizer.json +0 -0
  10. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/tokenizer.model +3 -0
  11. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/tokenizer_config.json +43 -0
  12. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/model_world_size_4_rank_0.pt +3 -0
  13. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/model_world_size_4_rank_1.pt +3 -0
  14. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/model_world_size_4_rank_2.pt +3 -0
  15. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/model_world_size_4_rank_3.pt +3 -0
  16. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/optim_world_size_4_rank_0.pt +3 -0
  17. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/optim_world_size_4_rank_1.pt +3 -0
  18. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/optim_world_size_4_rank_2.pt +3 -0
  19. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/optim_world_size_4_rank_3.pt +3 -0
  20. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/data.pt +3 -0
  21. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/extra_state_world_size_4_rank_0.pt +3 -0
  22. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/extra_state_world_size_4_rank_1.pt +3 -0
  23. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/extra_state_world_size_4_rank_2.pt +3 -0
  24. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/extra_state_world_size_4_rank_3.pt +3 -0
  25. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/fsdp_config.json +4 -0
  26. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/config.json +28 -0
  27. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/generation_config.json +6 -0
  28. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/special_tokens_map.json +24 -0
  29. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/tokenizer.json +0 -0
  30. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/tokenizer.model +3 -0
  31. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/tokenizer_config.json +43 -0
  32. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/model_world_size_4_rank_0.pt +3 -0
  33. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/model_world_size_4_rank_1.pt +3 -0
  34. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/model_world_size_4_rank_2.pt +3 -0
  35. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/model_world_size_4_rank_3.pt +3 -0
  36. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/optim_world_size_4_rank_0.pt +3 -0
  37. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/optim_world_size_4_rank_1.pt +3 -0
  38. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/optim_world_size_4_rank_2.pt +3 -0
  39. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/optim_world_size_4_rank_3.pt +3 -0
  40. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/data.pt +3 -0
  41. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/extra_state_world_size_4_rank_0.pt +3 -0
  42. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/extra_state_world_size_4_rank_1.pt +3 -0
  43. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/extra_state_world_size_4_rank_2.pt +3 -0
  44. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/extra_state_world_size_4_rank_3.pt +3 -0
  45. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/fsdp_config.json +4 -0
  46. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/config.json +28 -0
  47. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/generation_config.json +6 -0
  48. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/special_tokens_map.json +24 -0
  49. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/tokenizer.json +0 -0
  50. OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/tokenizer.model +3 -0
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/extra_state_world_size_4_rank_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db32085e6fcb115e6100ec93db670e5d8b8aacc014f1646e4526a43a81c84b70
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/extra_state_world_size_4_rank_1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aadfe6911ddcf346606b96a82b034b5a145f08b6033cec877bb2f335bcd04ea6
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/extra_state_world_size_4_rank_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85a5b95d18e423d003ed71b40dbcff76b5947c2cecb95b93d1fb685474f585b4
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/extra_state_world_size_4_rank_3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2baf7a3ee7bd205e753b2a5a681de7c9ee4720551e75edb095e27c2d55ce1d88
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/fsdp_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "FSDP_version": 2,
3
+ "world_size": 4
4
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "OlmoForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "clip_qkv": null,
10
+ "eos_token_id": 2,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 8192,
15
+ "max_position_embeddings": 2048,
16
+ "model_type": "olmo",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 16,
19
+ "num_key_value_heads": 16,
20
+ "pad_token_id": 2,
21
+ "rope_scaling": null,
22
+ "rope_theta": 10000.0,
23
+ "tie_word_embeddings": true,
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.50.0",
26
+ "use_cache": true,
27
+ "vocab_size": 32000
28
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "eos_token_id": 2,
4
+ "pad_token_id": 1,
5
+ "transformers_version": "4.50.0"
6
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/huggingface/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "extra_special_tokens": {},
35
+ "legacy": false,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/model_world_size_4_rank_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e7ecf90c65477032204b07d92ee3bce11c1798981428ecef4d5b863befd9d07
3
+ size 1139342746
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/model_world_size_4_rank_1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1183ff6d06ba6d8f25c1945005e8413434df8050f1e83f355f09dc87a2d4737a
3
+ size 1139342746
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/model_world_size_4_rank_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0c8068de6cf15f92d12cfaf56fc0053b662c25747901c72372fadf67d7e1106
3
+ size 1139342746
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/model_world_size_4_rank_3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56c9d64346fe3bdb20ff88f9eb27d47bf8f2165d584336e47bb5592145316e94
3
+ size 1139342746
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/optim_world_size_4_rank_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89b2cdd273230564747e7c9c2920401a30746992e6659ead88c42fbccc4f7266
3
+ size 2278784506
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/optim_world_size_4_rank_1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27a623955479d7734d92a99a8c98eaf18ac6dc4c38e43221beedbdf6c5f3868a
3
+ size 2278784506
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/optim_world_size_4_rank_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08450721ae154ebf09ef1e6443b28f213fd56e9a8c56f85daf8d4085c52daf57
3
+ size 2278784506
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/actor/optim_world_size_4_rank_3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72272e49edabdf9781f711e4221643cf3ed9a9acb511e79d4784dcabfc1e9bab
3
+ size 2278784506
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_1024/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efaff263c8c7c334aedad69f883b6c9835c234a890d7c5d429bbb339a5d70d5b
3
+ size 1492
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/extra_state_world_size_4_rank_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:662dd8c7a3dbf5c6b301a49bc1e9bf18466ac8a83e9937a87f98461d3c67cc71
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/extra_state_world_size_4_rank_1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b275e96100ffed4cc5db67f0b25bdce798a4089c86248c73fb3346b26f51037
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/extra_state_world_size_4_rank_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8645785dbf9c27ef1a4a15d5c226486649a6044233f343a0054112de26249a9
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/extra_state_world_size_4_rank_3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4d8b740b7941e765c78f4fc82470f32accff42f2218f2684a6ef48397508035
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/fsdp_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "FSDP_version": 2,
3
+ "world_size": 4
4
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "OlmoForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "clip_qkv": null,
10
+ "eos_token_id": 2,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 8192,
15
+ "max_position_embeddings": 2048,
16
+ "model_type": "olmo",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 16,
19
+ "num_key_value_heads": 16,
20
+ "pad_token_id": 2,
21
+ "rope_scaling": null,
22
+ "rope_theta": 10000.0,
23
+ "tie_word_embeddings": true,
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.50.0",
26
+ "use_cache": true,
27
+ "vocab_size": 32000
28
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "eos_token_id": 2,
4
+ "pad_token_id": 1,
5
+ "transformers_version": "4.50.0"
6
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/huggingface/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "extra_special_tokens": {},
35
+ "legacy": false,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/model_world_size_4_rank_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67e5768cef147aeeb2f495fb95a03e3143a4b5e272c65eb19a4c8336c96da21d
3
+ size 1139342746
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/model_world_size_4_rank_1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd229d9969b4486ca4373c740e044684515974321a192a340e368a8e306843b4
3
+ size 1139342746
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/model_world_size_4_rank_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6035b476667091e192c5499c0fe04fc30ea2de95c8c92ff18219a060dd6b3b99
3
+ size 1139342746
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/model_world_size_4_rank_3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:354a3135ed51c82a986aa350b84fd11f3a4106d7fbaa52f427839830224ac79c
3
+ size 1139342746
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/optim_world_size_4_rank_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee7606496fe4fc75558d9516e0e4ac64e406ec50ba4ae5081842bd287358c9c5
3
+ size 2278784506
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/optim_world_size_4_rank_1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:911bd4e05f2dcca25fded331e3ec7329533421c14e34751ccd90604073a23d12
3
+ size 2278784506
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/optim_world_size_4_rank_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0dbe63d3c5de6217bfa9ba0482bae0243d5b7aeea03ba396a5922685ef4673e
3
+ size 2278784506
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/actor/optim_world_size_4_rank_3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93a9cc41857a7059e5053e14278fe914f84159c4874a02ae1d93ef1d6948585d
3
+ size 2278784506
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_128/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f57bf3f56e9a22e4d6744e88e07e2d1b16b5ee29600225017ad0cc66d02b8a8a
3
+ size 1492
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/extra_state_world_size_4_rank_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08d53df345675eacdc40ac527e8d6a07bdcba098f277fd43ad50289426838524
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/extra_state_world_size_4_rank_1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34289df7952a02744d26f13de9c10c8665d452ec3a3750059b149c4cf3659653
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/extra_state_world_size_4_rank_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47f9f5b369212c67a6a0c99464dd6d5d792ed502dabf7c1edafd9cd1117f6898
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/extra_state_world_size_4_rank_3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9195377910368f98959cb59376a2ba45337308435428f996baaa9f1d858d877
3
+ size 14632
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/fsdp_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "FSDP_version": 2,
3
+ "world_size": 4
4
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "OlmoForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "clip_qkv": null,
10
+ "eos_token_id": 2,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 8192,
15
+ "max_position_embeddings": 2048,
16
+ "model_type": "olmo",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 16,
19
+ "num_key_value_heads": 16,
20
+ "pad_token_id": 2,
21
+ "rope_scaling": null,
22
+ "rope_theta": 10000.0,
23
+ "tie_word_embeddings": true,
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.50.0",
26
+ "use_cache": true,
27
+ "vocab_size": 32000
28
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "eos_token_id": 2,
4
+ "pad_token_id": 1,
5
+ "transformers_version": "4.50.0"
6
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
OLMo-1B/OLMo-1B-constant-3e4-mix-step60000-3e4-16/global_step_256/actor/huggingface/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723