Mingke977 commited on
Commit
cc7e1b0
·
verified ·
1 Parent(s): 48d41a0

Add files using upload-large-folder tool

Browse files
Files changed (47) hide show
  1. chat_template.jinja +1 -1
  2. config.json +57 -57
  3. configuration.json +1 -0
  4. model-1-of-40.safetensors +2 -2
  5. model-10-of-40.safetensors +2 -2
  6. model-11-of-40.safetensors +2 -2
  7. model-12-of-40.safetensors +2 -2
  8. model-13-of-40.safetensors +2 -2
  9. model-14-of-40.safetensors +2 -2
  10. model-15-of-40.safetensors +2 -2
  11. model-16-of-40.safetensors +2 -2
  12. model-17-of-40.safetensors +2 -2
  13. model-18-of-40.safetensors +2 -2
  14. model-19-of-40.safetensors +2 -2
  15. model-2-of-40.safetensors +2 -2
  16. model-20-of-40.safetensors +2 -2
  17. model-21-of-40.safetensors +2 -2
  18. model-22-of-40.safetensors +2 -2
  19. model-23-of-40.safetensors +2 -2
  20. model-24-of-40.safetensors +2 -2
  21. model-25-of-40.safetensors +2 -2
  22. model-26-of-40.safetensors +2 -2
  23. model-27-of-40.safetensors +2 -2
  24. model-28-of-40.safetensors +2 -2
  25. model-29-of-40.safetensors +2 -2
  26. model-3-of-40.safetensors +2 -2
  27. model-30-of-40.safetensors +2 -2
  28. model-31-of-40.safetensors +2 -2
  29. model-32-of-40.safetensors +2 -2
  30. model-33-of-40.safetensors +2 -2
  31. model-34-of-40.safetensors +2 -2
  32. model-35-of-40.safetensors +2 -2
  33. model-36-of-40.safetensors +2 -2
  34. model-37-of-40.safetensors +2 -2
  35. model-38-of-40.safetensors +2 -2
  36. model-39-of-40.safetensors +2 -2
  37. model-4-of-40.safetensors +2 -2
  38. model-40-of-40.safetensors +2 -2
  39. model-5-of-40.safetensors +2 -2
  40. model-6-of-40.safetensors +2 -2
  41. model-7-of-40.safetensors +2 -2
  42. model-8-of-40.safetensors +2 -2
  43. model-9-of-40.safetensors +2 -2
  44. model-non-layer.safetensors +2 -2
  45. model.safetensors.index.json +0 -0
  46. modeling_deepseek.py +4 -6
  47. mtp-1-of-1.safetensors +2 -2
chat_template.jinja CHANGED
@@ -100,4 +100,4 @@
100
 
101
  {%- if add_generation_prompt -%}
102
  {{ '<|Assistant|>' }}{{ '<|end_of_thought|>' }}
103
- {%- endif -%}
 
100
 
101
  {%- if add_generation_prompt -%}
102
  {{ '<|Assistant|>' }}{{ '<|end_of_thought|>' }}
103
+ {%- endif -%}
config.json CHANGED
@@ -1,58 +1,58 @@
1
  {
2
- "architectures": [
3
- "DeepseekV3ForCausalLM"
4
- ],
5
- "attention_bias": false,
6
- "attention_dropout": 0.0,
7
- "auto_map": {
8
- "AutoConfig": "configuration_deepseek.DeepseekV3Config",
9
- "AutoModel": "modeling_deepseek.DeepseekV3Model",
10
- "AutoModelForCausalLM": "modeling_deepseek.DeepseekV3ForCausalLM"
11
- },
12
- "bos_token_id": 0,
13
- "eos_token_id": 1,
14
- "ep_size": 1,
15
- "first_k_dense_replace": 1,
16
- "hidden_act": "silu",
17
- "hidden_size": 2048,
18
- "initializer_range": 0.02,
19
- "intermediate_size": 7168,
20
- "kv_lora_rank": 512,
21
- "max_position_embeddings": 131072,
22
- "model_type": "joyai_llm_flash",
23
- "moe_intermediate_size": 768,
24
- "moe_layer_freq": 1,
25
- "n_group": 1,
26
- "n_routed_experts": 256,
27
- "n_shared_experts": 1,
28
- "norm_topk_prob": true,
29
- "num_attention_heads": 32,
30
- "num_experts_per_tok": 8,
31
- "num_hidden_layers": 40,
32
- "num_key_value_heads": 32,
33
- "num_nextn_predict_layers": 1,
34
- "q_lora_rank": 1536,
35
- "qk_nope_head_dim": 128,
36
- "qk_rope_head_dim": 64,
37
- "quantization_config": {
38
- "activation_scheme": "dynamic",
39
- "fmt": "e4m3",
40
- "quant_method": "fp8",
41
- "weight_block_size": [
42
- 128,
43
- 128
44
- ]
45
- },
46
- "rms_norm_eps": 1e-06,
47
- "rope_theta": 32000000,
48
- "routed_scaling_factor": 2.5,
49
- "scoring_func": "sigmoid",
50
- "tie_word_embeddings": false,
51
- "topk_group": 1,
52
- "topk_method": "noaux_tc",
53
- "torch_dtype": "bfloat16",
54
- "transformers_version": "4.44.2",
55
- "use_cache": true,
56
- "v_head_dim": 128,
57
- "vocab_size": 129280
58
- }
 
1
  {
2
+ "architectures": [
3
+ "DeepseekV3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "auto_map": {
8
+ "AutoConfig": "configuration_deepseek.DeepseekV3Config",
9
+ "AutoModel": "modeling_deepseek.DeepseekV3Model",
10
+ "AutoModelForCausalLM": "modeling_deepseek.DeepseekV3ForCausalLM"
11
+ },
12
+ "bos_token_id": 0,
13
+ "eos_token_id": 1,
14
+ "ep_size": 1,
15
+ "first_k_dense_replace": 1,
16
+ "hidden_act": "silu",
17
+ "hidden_size": 2048,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 7168,
20
+ "kv_lora_rank": 512,
21
+ "max_position_embeddings": 131072,
22
+ "model_type": "joyai_llm_flash",
23
+ "moe_intermediate_size": 768,
24
+ "moe_layer_freq": 1,
25
+ "n_group": 1,
26
+ "n_routed_experts": 256,
27
+ "n_shared_experts": 1,
28
+ "norm_topk_prob": true,
29
+ "num_attention_heads": 32,
30
+ "num_experts_per_tok": 8,
31
+ "num_hidden_layers": 40,
32
+ "num_key_value_heads": 32,
33
+ "num_nextn_predict_layers": 1,
34
+ "q_lora_rank": 1536,
35
+ "qk_nope_head_dim": 128,
36
+ "qk_rope_head_dim": 64,
37
+ "quantization_config": {
38
+ "activation_scheme": "dynamic",
39
+ "fmt": "e4m3",
40
+ "quant_method": "fp8",
41
+ "weight_block_size": [
42
+ 128,
43
+ 128
44
+ ]
45
+ },
46
+ "rms_norm_eps": 1e-06,
47
+ "rope_theta": 32000000,
48
+ "routed_scaling_factor": 2.5,
49
+ "scoring_func": "sigmoid",
50
+ "tie_word_embeddings": false,
51
+ "topk_group": 1,
52
+ "topk_method": "noaux_tc",
53
+ "torch_dtype": "bfloat16",
54
+ "transformers_version": "4.44.2",
55
+ "use_cache": true,
56
+ "v_head_dim": 128,
57
+ "vocab_size": 129280
58
+ }
configuration.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"framework":"Pytorch","task":"text-generation"}
model-1-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d7d2f4fca392ea83b71c94991028ce8c22108082e5853cd8b5d4bac9a6567a5d
3
- size 70417432
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:48497c8832fcf0b1af0fccba0c9eacf948580b18d57dd735021a75330f95b394
3
+ size 70417464
model-10-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3cf0ae859901c7edffb068245fbc6604d94b4d4891f825dc9c39e831bb7d8231
3
- size 1240579000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf9fd692aa8ce99badb836897a561176653f7eeb7693c9e21e1906ea47196ac0
3
+ size 1240579032
model-11-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:adf0ebaa860a6d3540672da76461e417695b22d198ae37f7ce982ea130704225
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb7fa33ede9d972a415d227eb101e605c95bb5ef73b179a3681af2abb9471a57
3
+ size 1240580584
model-12-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dd4c356929aacd52345d948ba514d1959938add01290a1373adaced169a5615a
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bde73a20d75033a5b9f10c27dc6a11d29982f59c484765e5a0abcdbc379dc978
3
+ size 1240580584
model-13-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:46e718a4c1f759e985beb479164851c20f80aab1ae673cb6339fd34601681c20
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee6ff449b07098d305305073a34182b9a9fe06f3e2aa29a3b709bb01c4e2089b
3
+ size 1240580584
model-14-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c8412f9d9bb91e57e32821b7a97c0fff5418d6a9a845dc36041e451661faf15a
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea091e6db39fc79d5c41313dae709355bc3575f2a3f2404501502dd2682f3ec1
3
+ size 1240580584
model-15-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aa0fb469858a28e57990b479096b79108ae13cff06c6ac67c9b7caa235b7e5ca
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e733587d3c0245e7e9ae939a8322ab99a9ee6a57fcc37a050bb7087e4546a44
3
+ size 1240580584
model-16-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d764ca0f5a313145e07b99d029e87bb2cb473d63d431cd9557cd11eebdabc735
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fba336f7bca7eb6c5d4e6741ae2c6414c4cb6b6eee00cc3d83592edcb4c48e21
3
+ size 1240580584
model-17-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3484cdfbbfa11d7dd7fd3c6822ee76de89c15c524b0b3f600b9a8c43dd164971
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:298748f91a66d94a7892a34c3951c7de9d38fa57126ac4d8d8f29a903e0219ea
3
+ size 1240580584
model-18-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:20a4b4fdeea04a49aaf0bffbe1a83e9be2933ddd823132f65578346710c86ee7
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a66ff92c1b42acc3dd7cddb454c751480467b990694e86e657bb07be55a165f7
3
+ size 1240580584
model-19-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8a016c68e4c75beebe21973b5943fe6cd35496a3c278cfb14589834781357bef
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88fb0f0ce805762bb7d49e479f2d2b2047fa685ff18a937ac36e988030e8fd5f
3
+ size 1240580584
model-2-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2ca44e797ad60b01c1389e28840fab385b787239d2fd4592eb614e9f0605f213
3
- size 1240579000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f70aed8f1f9abd325c36385fb129a97d9e10f6a3ba3a98f1d02eb86d1eb9201
3
+ size 1240579032
model-20-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dd2c7c6ca3658d53f7e3690334c8ceeba73d6e91fe44754e39a42f67eb560b7e
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a47822cfab0deaf9ae7365458d49c3d143744ab988ce9a126782c63888b95b0a
3
+ size 1240580584
model-21-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f7f47f18b836de4cde8e212801a08ff00ce1bb9e728ecd62197b083f33463192
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7465adef3c55ea170471a501b9acf24b88f6de9976d1ac8473c1e337ca75c81f
3
+ size 1240580584
model-22-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c93488f761433d508fc79d57ca12c076497d0d76607b55e9b92a1e00857fdd06
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:253d68a26c0b9057109db575f853d04ba50f12075648c93a4843ba90aa2c3456
3
+ size 1240580584
model-23-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4a141d99d73779c8a41a07a414a040b3eb8ee2da921b3d39057db1268126b4f
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d071fa057c15ada66ab549d9bd854c092789f8803f4a6135668ee87ca4690c2
3
+ size 1240580584
model-24-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5b6cae888ae0203ab72ff226fca2f509a0065c8749149a3be49a894961e2e386
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9d9768e6f052664386193a0b9b8ffdb74b189f806d824ec701bd684cbef252d
3
+ size 1240580584
model-25-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0080dacaa12cca4be6a63399903c46cc17fe81e7362e0b6912d137c5e942197a
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36b3273f48e8005b6d43875a1b8c3ed16711ee9af36799bd44f3918a91dc6c37
3
+ size 1240580584
model-26-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a21eb4f08465c03e0629f5deab8aefcb5041cf65be98c876d3648b61610ade59
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7f5f636d5cd0e81fb1d13284c6981db2f35e0a61bb014d48353c1329245f4f0
3
+ size 1240580584
model-27-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3f77abe8bb7c8d48a46b074eae99bb136eed54b5457d0fd3fb5589ee3fe3cd40
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c02f9e2237cb3af60eae5cf4ca3dcdd581184571019655502274b995017f681a
3
+ size 1240580584
model-28-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fc0b237b6b5284710d2c32d84d516a2fa57511b65ecf815ce104a69f2562bf53
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf2ae06caae01d992a7934975cc0e812bd628e6061c98b04137821bf93cc1e1a
3
+ size 1240580584
model-29-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:452f7f362ca865ef16d152502e30836391f24f9420b72b33123fbec7fb275815
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92ceb810aba74a674768ac5acc5b92754adb734b5dc5a4c03783b1c2d6a40b35
3
+ size 1240580584
model-3-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8f3fd8abe76ba31fa3ee2b7a1bd2083fe3f4c89e10e2aed00e10aacd80c61df9
3
- size 1240579000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d9eb12db9bf0b95b00d4a3c4ae8b80302b8fd7832db02b08edc43e62956b9e8
3
+ size 1240579032
model-30-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dcd4999348779b34b203564c7403965bd6279ad10897a56e4b6acb7f1f7c5baf
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ead2b2e2900e23825a2895aecd6f0e66c9b829ed29ba842f5dae37c39ec041b
3
+ size 1240580584
model-31-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2563e54f2a5536989dbab99d5aa1483488075ec467ee5758f92242b7358de61b
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21fb5a191d41f3fd1608f2ddfcb32bfc3a91e0f639cacde56494006d080067d5
3
+ size 1240580584
model-32-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8b6350967e09c581d1533c0d5bdbd7dafa49b102c438b4739010b5eb6d7f5938
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c3d2a82580d097fed63224daf4f199cc1389e9284f9e1463bf5285577baa147
3
+ size 1240580584
model-33-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:72f04cf97b30ed4f78c621a17a0baba64768daadf834dcbd40c2b34fa9ff35d0
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b093b9e98d02a6dc5aef34989ce766e294e32685d6c16067ddcc1b5d34d2ba75
3
+ size 1240580584
model-34-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5e7c200d90d14deff777ad6173c89812f216465d121ca2ea1321b7eba6ff2bff
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab3f9d0b02a4c0018f84885323d46498ffd0f887adfb514b38ff7cd12b257801
3
+ size 1240580584
model-35-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dc19f9f4a6ab0061b3d1a3745112e18e25ca20596bd4b2062e10eccf562c09f0
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d14d0c3eed182277e98c1c65dacc512ab0378a853915df9977152b0acc02f2d
3
+ size 1240580584
model-36-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e70a79a57258e4fced6967888ceaf461526e6e55dbb4499e03914d28211d7cbc
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9644ccd0bb9c8574233d8c125e7e47d99814daf8982509caf774e88e5e2f1346
3
+ size 1240580584
model-37-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9c46036a5be022ca09c88a83034e60d00620e16ad79c917e4f4309353b639bb5
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1be01191c347a9871c535553c404b03e5e824b00d88ee1692b5f493ad239146
3
+ size 1240580584
model-38-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:90bf3a173dbf30d34e63ee1572130c0b5f32314179cb3673423be8762bd3bf2c
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19778856b18e6f3be4891561ad8cabe5087a716791835e95a5b47928e891103f
3
+ size 1240580584
model-39-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6884d2526b5fbba4b6ac7f395ce57ae94d6db67f352c4612e80ed52dd0c5916e
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:567cecfd87ecb858cb52841b34396ba60c33dc8513a6db54567bd6df4b758aa0
3
+ size 1240580584
model-4-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2b4ce9efb22eb6ee73bb906316495f579ee857e7674b6a5f785e5e1aa2a50427
3
- size 1240579000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72e7477976fe7223bb9d4c5bc36a0fadd9aef97cafe05758764565353e18394c
3
+ size 1240579032
model-40-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:315cb2bfd43a9002670978ba4a864bf541d4bd042e4a22ee7661aa03343bf42d
3
- size 1240580552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9734d1dc7611dd095009ae3e198d03ee4bb525cd0d08ac48929fdfe914bd6165
3
+ size 1240580584
model-5-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b7b45d4a132f074f147b83208193a5b93e99ad3b736cb1febd428ec8dc22f17a
3
- size 1240579000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f25ce7802ce04cc54be0fd059e77b00f4576957d8ed7f7536c3bfc4d599e6583
3
+ size 1240579032
model-6-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:38ce3a586f486ea09aab0dacedf31688fb5aaf7ee5583337219f1d3815fd0585
3
- size 1240579000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b654fcd83e86a401aa8a77eb58aae2470f939e858da553929555c43b1b89e3b9
3
+ size 1240579032
model-7-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1f37d97a025c306b4807d3c9b7b12c9d8d422db6279d63eb2cdee623e432ab0f
3
- size 1240579000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80f8b7d82c2da928acff09102ff78b514a4a9afdb6b63bde6f455354b31d9614
3
+ size 1240579032
model-8-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3ad8fdbaad07cb38452a17a44937f4208ab58b3ace70fabf423976535a6ac962
3
- size 1240579000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:032578714641e56a3f899f02245e3e444e1596b2a9c7bb50aa63c215771a87ae
3
+ size 1240579032
model-9-of-40.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5831a26448dee958644d5d9383d5748726e8931ebbee18e034f285ad1c169596
3
- size 1240579000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d9c00db95ab26608f6aed129da66aad35218d52d56bb71a6b0706d26565c818
3
+ size 1240579032
model-non-layer.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f5b0f8e8ace4f013693d24123ee0db03d731a409cdc5401b8beb87b0c84b7b1f
3
- size 1059066152
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:358810bbba6e784b3be7462f7398fed2ba159a0edcec4f586aa72e877f4187ca
3
+ size 1059066184
model.safetensors.index.json CHANGED
The diff for this file is too large to render. See raw diff
 
modeling_deepseek.py CHANGED
@@ -11,15 +11,13 @@ from typing import Callable, Optional, Tuple, Union
11
  import torch
12
  import torch.nn.functional as F
13
  from torch import nn
 
14
  from transformers.activations import ACT2FN
15
  from transformers.cache_utils import Cache, DynamicCache, StaticCache
16
  from transformers.generation import GenerationMixin
17
  from transformers.modeling_attn_mask_utils import AttentionMaskConverter
18
  from transformers.modeling_flash_attention_utils import FlashAttentionKwargs
19
- from transformers.modeling_outputs import (
20
- BaseModelOutputWithPast,
21
- CausalLMOutputWithPast,
22
- )
23
  from transformers.modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update
24
  from transformers.modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
25
  from transformers.processing_utils import Unpack
@@ -33,11 +31,12 @@ from transformers.utils import (
33
  replace_return_docstrings,
34
  )
35
  from transformers.utils.deprecation import deprecate_kwarg
36
-
37
  from .configuration_deepseek import DeepseekV3Config
38
 
 
39
  if is_torch_flex_attn_available():
40
  from torch.nn.attention.flex_attention import BlockMask
 
41
  from transformers.integrations.flex_attention import make_flex_block_causal_mask
42
 
43
 
@@ -1027,4 +1026,3 @@ class DeepseekV3ForCausalLM(DeepseekV3PreTrainedModel, GenerationMixin):
1027
 
1028
 
1029
  __all__ = ["DeepseekV3PreTrainedModel", "DeepseekV3Model", "DeepseekV3ForCausalLM"]
1030
- __all__ = ["DeepseekV3PreTrainedModel", "DeepseekV3Model", "DeepseekV3ForCausalLM"]
 
11
  import torch
12
  import torch.nn.functional as F
13
  from torch import nn
14
+
15
  from transformers.activations import ACT2FN
16
  from transformers.cache_utils import Cache, DynamicCache, StaticCache
17
  from transformers.generation import GenerationMixin
18
  from transformers.modeling_attn_mask_utils import AttentionMaskConverter
19
  from transformers.modeling_flash_attention_utils import FlashAttentionKwargs
20
+ from transformers.modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast
 
 
 
21
  from transformers.modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update
22
  from transformers.modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
23
  from transformers.processing_utils import Unpack
 
31
  replace_return_docstrings,
32
  )
33
  from transformers.utils.deprecation import deprecate_kwarg
 
34
  from .configuration_deepseek import DeepseekV3Config
35
 
36
+
37
  if is_torch_flex_attn_available():
38
  from torch.nn.attention.flex_attention import BlockMask
39
+
40
  from transformers.integrations.flex_attention import make_flex_block_causal_mask
41
 
42
 
 
1026
 
1027
 
1028
  __all__ = ["DeepseekV3PreTrainedModel", "DeepseekV3Model", "DeepseekV3ForCausalLM"]
 
mtp-1-of-1.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2c03c864759878b2837e5aa10a9e0558c6bc0daa7ee0ec6b61b46854f63740f3
3
- size 616738408
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8467e89af978fe43308db5efef180d7a890849385174494bbc6ea0fd2115dc45
3
+ size 616738440