JLiangHe commited on
Commit
dc2ee4e
·
verified ·
1 Parent(s): 3316429

Upload agentic RL step 30 model weights

Browse files
Files changed (48) hide show
  1. .gitattributes +1 -0
  2. rl-agentic-new/step-30/README.md +9 -0
  3. rl-agentic-new/step-30/chat_template.jinja +154 -0
  4. rl-agentic-new/step-30/checkpoint-provenance.json +237 -0
  5. rl-agentic-new/step-30/config.json +83 -0
  6. rl-agentic-new/step-30/generation_config.json +4 -0
  7. rl-agentic-new/step-30/model-00001-of-00039.safetensors +3 -0
  8. rl-agentic-new/step-30/model-00002-of-00039.safetensors +3 -0
  9. rl-agentic-new/step-30/model-00003-of-00039.safetensors +3 -0
  10. rl-agentic-new/step-30/model-00004-of-00039.safetensors +3 -0
  11. rl-agentic-new/step-30/model-00005-of-00039.safetensors +3 -0
  12. rl-agentic-new/step-30/model-00006-of-00039.safetensors +3 -0
  13. rl-agentic-new/step-30/model-00007-of-00039.safetensors +3 -0
  14. rl-agentic-new/step-30/model-00008-of-00039.safetensors +3 -0
  15. rl-agentic-new/step-30/model-00009-of-00039.safetensors +3 -0
  16. rl-agentic-new/step-30/model-00010-of-00039.safetensors +3 -0
  17. rl-agentic-new/step-30/model-00011-of-00039.safetensors +3 -0
  18. rl-agentic-new/step-30/model-00012-of-00039.safetensors +3 -0
  19. rl-agentic-new/step-30/model-00013-of-00039.safetensors +3 -0
  20. rl-agentic-new/step-30/model-00014-of-00039.safetensors +3 -0
  21. rl-agentic-new/step-30/model-00015-of-00039.safetensors +3 -0
  22. rl-agentic-new/step-30/model-00016-of-00039.safetensors +3 -0
  23. rl-agentic-new/step-30/model-00017-of-00039.safetensors +3 -0
  24. rl-agentic-new/step-30/model-00018-of-00039.safetensors +3 -0
  25. rl-agentic-new/step-30/model-00019-of-00039.safetensors +3 -0
  26. rl-agentic-new/step-30/model-00020-of-00039.safetensors +3 -0
  27. rl-agentic-new/step-30/model-00021-of-00039.safetensors +3 -0
  28. rl-agentic-new/step-30/model-00022-of-00039.safetensors +3 -0
  29. rl-agentic-new/step-30/model-00023-of-00039.safetensors +3 -0
  30. rl-agentic-new/step-30/model-00024-of-00039.safetensors +3 -0
  31. rl-agentic-new/step-30/model-00025-of-00039.safetensors +3 -0
  32. rl-agentic-new/step-30/model-00026-of-00039.safetensors +3 -0
  33. rl-agentic-new/step-30/model-00027-of-00039.safetensors +3 -0
  34. rl-agentic-new/step-30/model-00028-of-00039.safetensors +3 -0
  35. rl-agentic-new/step-30/model-00029-of-00039.safetensors +3 -0
  36. rl-agentic-new/step-30/model-00030-of-00039.safetensors +3 -0
  37. rl-agentic-new/step-30/model-00031-of-00039.safetensors +3 -0
  38. rl-agentic-new/step-30/model-00032-of-00039.safetensors +3 -0
  39. rl-agentic-new/step-30/model-00033-of-00039.safetensors +3 -0
  40. rl-agentic-new/step-30/model-00034-of-00039.safetensors +3 -0
  41. rl-agentic-new/step-30/model-00035-of-00039.safetensors +3 -0
  42. rl-agentic-new/step-30/model-00036-of-00039.safetensors +3 -0
  43. rl-agentic-new/step-30/model-00037-of-00039.safetensors +3 -0
  44. rl-agentic-new/step-30/model-00038-of-00039.safetensors +3 -0
  45. rl-agentic-new/step-30/model-00039-of-00039.safetensors +3 -0
  46. rl-agentic-new/step-30/model.safetensors.index.json +434 -0
  47. rl-agentic-new/step-30/tokenizer.json +3 -0
  48. rl-agentic-new/step-30/tokenizer_config.json +32 -0
.gitattributes CHANGED
@@ -59,3 +59,4 @@ rl-stage1-candidate/step-180-96k/tokenizer.json filter=lfs diff=lfs merge=lfs -t
59
  rl-stage1-candidate/step-150-96k/tokenizer.json filter=lfs diff=lfs merge=lfs -text
60
  rl-agentic-new/step-10/tokenizer.json filter=lfs diff=lfs merge=lfs -text
61
  rl-agentic-new/step-20/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
59
  rl-stage1-candidate/step-150-96k/tokenizer.json filter=lfs diff=lfs merge=lfs -text
60
  rl-agentic-new/step-10/tokenizer.json filter=lfs diff=lfs merge=lfs -text
61
  rl-agentic-new/step-20/tokenizer.json filter=lfs diff=lfs merge=lfs -text
62
+ rl-agentic-new/step-30/tokenizer.json filter=lfs diff=lfs merge=lfs -text
rl-agentic-new/step-30/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ # RL Stage 1 candidate: step 180, 96K context
2
+
3
+ Final checkpoint after 60 additional RL updates from step 120, restoring the native training state. Training context: 98,304 tokens including prompt and response.
4
+
5
+ Includes all 39 model-weight shards, tokenizer, chat template and model/generation configurations. Optimizer and scheduler states are not included.
6
+
7
+ Training run: https://wandb.ai/jlianghe-yale-university/sci-llm-training/runs/rh3bk2yv
8
+
9
+ See `checkpoint-provenance.json` for file SHA-256 hashes and checkpoint identity.
rl-agentic-new/step-30/chat_template.jinja ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- set image_count = namespace(value=0) %}
2
+ {%- set video_count = namespace(value=0) %}
3
+ {%- macro render_content(content, do_vision_count, is_system_content=false) %}
4
+ {%- if content is string %}
5
+ {{- content }}
6
+ {%- elif content is iterable and content is not mapping %}
7
+ {%- for item in content %}
8
+ {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
9
+ {%- if is_system_content %}
10
+ {{- raise_exception('System message cannot contain images.') }}
11
+ {%- endif %}
12
+ {%- if do_vision_count %}
13
+ {%- set image_count.value = image_count.value + 1 %}
14
+ {%- endif %}
15
+ {%- if add_vision_id %}
16
+ {{- 'Picture ' ~ image_count.value ~ ': ' }}
17
+ {%- endif %}
18
+ {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
19
+ {%- elif 'video' in item or item.type == 'video' %}
20
+ {%- if is_system_content %}
21
+ {{- raise_exception('System message cannot contain videos.') }}
22
+ {%- endif %}
23
+ {%- if do_vision_count %}
24
+ {%- set video_count.value = video_count.value + 1 %}
25
+ {%- endif %}
26
+ {%- if add_vision_id %}
27
+ {{- 'Video ' ~ video_count.value ~ ': ' }}
28
+ {%- endif %}
29
+ {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
30
+ {%- elif 'text' in item %}
31
+ {{- item.text }}
32
+ {%- else %}
33
+ {{- raise_exception('Unexpected item type in content.') }}
34
+ {%- endif %}
35
+ {%- endfor %}
36
+ {%- elif content is none or content is undefined %}
37
+ {{- '' }}
38
+ {%- else %}
39
+ {{- raise_exception('Unexpected content type.') }}
40
+ {%- endif %}
41
+ {%- endmacro %}
42
+ {%- if not messages %}
43
+ {{- raise_exception('No messages provided.') }}
44
+ {%- endif %}
45
+ {%- if tools and tools is iterable and tools is not mapping %}
46
+ {{- '<|im_start|>system\n' }}
47
+ {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
48
+ {%- for tool in tools %}
49
+ {{- "\n" }}
50
+ {{- tool | tojson }}
51
+ {%- endfor %}
52
+ {{- "\n</tools>" }}
53
+ {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
54
+ {%- if messages[0].role == 'system' %}
55
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
56
+ {%- if content %}
57
+ {{- '\n\n' + content }}
58
+ {%- endif %}
59
+ {%- endif %}
60
+ {{- '<|im_end|>\n' }}
61
+ {%- else %}
62
+ {%- if messages[0].role == 'system' %}
63
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
64
+ {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
65
+ {%- endif %}
66
+ {%- endif %}
67
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
68
+ {%- for message in messages[::-1] %}
69
+ {%- set index = (messages|length - 1) - loop.index0 %}
70
+ {%- if ns.multi_step_tool and message.role == "user" %}
71
+ {%- set content = render_content(message.content, false)|trim %}
72
+ {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
73
+ {%- set ns.multi_step_tool = false %}
74
+ {%- set ns.last_query_index = index %}
75
+ {%- endif %}
76
+ {%- endif %}
77
+ {%- endfor %}
78
+ {%- if ns.multi_step_tool %}
79
+ {{- raise_exception('No user query found in messages.') }}
80
+ {%- endif %}
81
+ {%- for message in messages %}
82
+ {%- set content = render_content(message.content, true)|trim %}
83
+ {%- if message.role == "system" %}
84
+ {%- if not loop.first %}
85
+ {{- raise_exception('System message must be at the beginning.') }}
86
+ {%- endif %}
87
+ {%- elif message.role == "user" %}
88
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
89
+ {%- elif message.role == "assistant" %}
90
+ {%- set reasoning_content = '' %}
91
+ {%- if message.reasoning_content is string %}
92
+ {%- set reasoning_content = message.reasoning_content %}
93
+ {%- else %}
94
+ {%- if '</think>' in content %}
95
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
96
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
97
+ {%- endif %}
98
+ {%- endif %}
99
+ {%- set reasoning_content = reasoning_content|trim %}
100
+ {%- if loop.index0 > ns.last_query_index %}
101
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
102
+ {%- else %}
103
+ {{- '<|im_start|>' + message.role + '\n' + content }}
104
+ {%- endif %}
105
+ {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
106
+ {%- for tool_call in message.tool_calls %}
107
+ {%- if tool_call.function is defined %}
108
+ {%- set tool_call = tool_call.function %}
109
+ {%- endif %}
110
+ {%- if loop.first %}
111
+ {%- if content|trim %}
112
+ {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
113
+ {%- else %}
114
+ {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
115
+ {%- endif %}
116
+ {%- else %}
117
+ {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
118
+ {%- endif %}
119
+ {%- if tool_call.arguments is defined %}
120
+ {%- for args_name, args_value in tool_call.arguments|items %}
121
+ {{- '<parameter=' + args_name + '>\n' }}
122
+ {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
123
+ {{- args_value }}
124
+ {{- '\n</parameter>\n' }}
125
+ {%- endfor %}
126
+ {%- endif %}
127
+ {{- '</function>\n</tool_call>' }}
128
+ {%- endfor %}
129
+ {%- endif %}
130
+ {{- '<|im_end|>\n' }}
131
+ {%- elif message.role == "tool" %}
132
+ {%- if loop.previtem and loop.previtem.role != "tool" %}
133
+ {{- '<|im_start|>user' }}
134
+ {%- endif %}
135
+ {{- '\n<tool_response>\n' }}
136
+ {{- content }}
137
+ {{- '\n</tool_response>' }}
138
+ {%- if not loop.last and loop.nextitem.role != "tool" %}
139
+ {{- '<|im_end|>\n' }}
140
+ {%- elif loop.last %}
141
+ {{- '<|im_end|>\n' }}
142
+ {%- endif %}
143
+ {%- else %}
144
+ {{- raise_exception('Unexpected message role.') }}
145
+ {%- endif %}
146
+ {%- endfor %}
147
+ {%- if add_generation_prompt %}
148
+ {{- '<|im_start|>assistant\n' }}
149
+ {%- if enable_thinking is defined and enable_thinking is false %}
150
+ {{- '<think>\n\n</think>\n\n' }}
151
+ {%- else %}
152
+ {{- '<think>\n' }}
153
+ {%- endif %}
154
+ {%- endif %}
rl-agentic-new/step-30/checkpoint-provenance.json ADDED
@@ -0,0 +1,237 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint_step": 180,
3
+ "additional_steps": 60,
4
+ "native_snapshot_rollout_id": 179,
5
+ "parent_model": "JLiangHe/sci-ckpt-final/rl-stage1-candidate/step-120",
6
+ "context_window_tokens": 98304,
7
+ "context_includes_prompt": true,
8
+ "wandb_run": "jlianghe-yale-university/sci-llm-training/rh3bk2yv",
9
+ "contents": "Model weights, tokenizer and model configuration only; no optimizer state.",
10
+ "files": {
11
+ "chat_template.jinja": {
12
+ "size": 7756,
13
+ "sha256": "a4aee8afcf2e0711942cf848899be66016f8d14a889ff9ede07bca099c28f715",
14
+ "git_blob_sha1": "a585dec894e63da457d9440ec6aa7caa16d20860"
15
+ },
16
+ "config.json": {
17
+ "size": 1980,
18
+ "sha256": "7ed31d50f8c0e8dc8219aa686a97c08665b8d9c85321e4c2fb766139fb6d01af",
19
+ "git_blob_sha1": "5afe67ad282e0b32b752f85e1399c997ba81b9b4"
20
+ },
21
+ "generation_config.json": {
22
+ "size": 68,
23
+ "sha256": "58fdcec35458d24f1e411dc30925225f9c1d7877972b443484df0d73f939a7c1",
24
+ "git_blob_sha1": "a78c517faaaebba89ede50060ec1141c51ee566a"
25
+ },
26
+ "model-00001-of-00039.safetensors": {
27
+ "size": 403269512,
28
+ "sha256": "ce2a6ab950100e1b2bc4d1169189ffe039b339de6e92fc371aee81c7e3822c14",
29
+ "git_blob_sha1": "91858a6128ade518f5afe1bad6fe91f7bb2c3234"
30
+ },
31
+ "model-00002-of-00039.safetensors": {
32
+ "size": 403260904,
33
+ "sha256": "f9e5e8799b87785679353425b4eaa9dc3ad0bc9fd5951395e8ba19ea61fd482b",
34
+ "git_blob_sha1": "0a267f3e671da32fa2c64d6926e43e09585ae14b"
35
+ },
36
+ "model-00003-of-00039.safetensors": {
37
+ "size": 369706840,
38
+ "sha256": "0b4a64aff3f7dc82ac34ccf56a1cfac7767ef4129d8d0bc71fd3b0e215f15214",
39
+ "git_blob_sha1": "e7a57210f7fbf72ef0f19262cba381c680f234d8"
40
+ },
41
+ "model-00004-of-00039.safetensors": {
42
+ "size": 335553624,
43
+ "sha256": "a6d093e88489e907d322f5a6e857126f34341b067d32670dbea66d7c91e806ed",
44
+ "git_blob_sha1": "5ed14ce34484063f4ac8f033396b75c8c42d6566"
45
+ },
46
+ "model-00005-of-00039.safetensors": {
47
+ "size": 520711312,
48
+ "sha256": "5558db2d35b3565fabda916bdfe81970b9c1d51f26b384a11d6e4abce827cad1",
49
+ "git_blob_sha1": "4066779a6a5079a62d1ac1211255651c0d509a2a"
50
+ },
51
+ "model-00006-of-00039.safetensors": {
52
+ "size": 335553624,
53
+ "sha256": "3c6981ae01471d7dac4067178bab920edaca521f21afb07e29f1452f91a53874",
54
+ "git_blob_sha1": "3448a54a25be0e7f946a7e1e88537e8f801e1cb0"
55
+ },
56
+ "model-00007-of-00039.safetensors": {
57
+ "size": 235488984,
58
+ "sha256": "73b873b6c86e991e8ff5179e710e76c4865f988cd7b6cfe642f29838c9154a51",
59
+ "git_blob_sha1": "3be96fa18fbe1ebe78a76c072f7763a73a42ab28"
60
+ },
61
+ "model-00008-of-00039.safetensors": {
62
+ "size": 403252936,
63
+ "sha256": "0178f6b007f71cf47e6be3371536fb558d8a243d2a0adcf85fdb797e880e3426",
64
+ "git_blob_sha1": "a82d67c47bc397ca84beaad06c94e01190755909"
65
+ },
66
+ "model-00009-of-00039.safetensors": {
67
+ "size": 453011952,
68
+ "sha256": "ba113353f51d128c1b06341b215d039b9dc9d929b66fa08a89213976f6765205",
69
+ "git_blob_sha1": "a1c3e5820236a1cd395df394913f2abb84cdd517"
70
+ },
71
+ "model-00010-of-00039.safetensors": {
72
+ "size": 403252936,
73
+ "sha256": "420c9fa85efdd6eccebc4162f793f9643a6800e0418782d3f7db6c3056ec48ce",
74
+ "git_blob_sha1": "c078baf3fdb08eb44a1645b1d23d1f9c3751e6c2"
75
+ },
76
+ "model-00011-of-00039.safetensors": {
77
+ "size": 403260920,
78
+ "sha256": "7029c0d3be5d29af5867552f3abcb09ef2e1594f081e43125ae832d141ef4ff6",
79
+ "git_blob_sha1": "e6090dc6700e72fca03eae92e53489d37fd22b9d"
80
+ },
81
+ "model-00012-of-00039.safetensors": {
82
+ "size": 369706848,
83
+ "sha256": "c4825d9467b8611c838fe3d1917faedabb66c01e4fec0f483f66ef43a0ff4b0f",
84
+ "git_blob_sha1": "5ad606182b78a8225d61dde07295122e5c1ba0c0"
85
+ },
86
+ "model-00013-of-00039.safetensors": {
87
+ "size": 335553624,
88
+ "sha256": "f4cab5c4c61ef9737fa80ba75a3ddd0851cdb7694de15babb97813eb4181f6b3",
89
+ "git_blob_sha1": "9ef0a39b7af4049d7663f2368753c9f777529160"
90
+ },
91
+ "model-00014-of-00039.safetensors": {
92
+ "size": 520711304,
93
+ "sha256": "54f9414146ed96dfe9b2471f293a3084153a9bfbcc92a67bcae9e299e1c87d01",
94
+ "git_blob_sha1": "e764d059beb4b0a13583e444e3d52c35500b8644"
95
+ },
96
+ "model-00015-of-00039.safetensors": {
97
+ "size": 335553616,
98
+ "sha256": "98818a209d5239278ad22d45ab07f002cffcf7e820a8280bc11fd3147af7234b",
99
+ "git_blob_sha1": "2a951ed232d7c7bb07d2b6a22484ed619f2bf212"
100
+ },
101
+ "model-00016-of-00039.safetensors": {
102
+ "size": 235488984,
103
+ "sha256": "a1a6809dfbcffcd811ff3ceea1fba1cc8962dbefff1045df368565754a9521f9",
104
+ "git_blob_sha1": "11ed3a8dcb15685e427298ebfecfb6bc15f1e715"
105
+ },
106
+ "model-00017-of-00039.safetensors": {
107
+ "size": 403252936,
108
+ "sha256": "10de80ba24c1da935b081c3f9e6a484b84604d376335a708db7242d745c5743f",
109
+ "git_blob_sha1": "af7aeee058c74de97eaf42f619ba281dce6e3dfc"
110
+ },
111
+ "model-00018-of-00039.safetensors": {
112
+ "size": 403260920,
113
+ "sha256": "4f5c576d09705f10f5a0adcd08823bdcd4e7898234eb7ca44c9971d3ab3a6740",
114
+ "git_blob_sha1": "2e03d567b71d8673717682727c3acf473945bb40"
115
+ },
116
+ "model-00019-of-00039.safetensors": {
117
+ "size": 402671512,
118
+ "sha256": "9ed9ecc99e2604017021340b6ff9ca2ebaaaf76a77b6055ca61a1e1d63484ff5",
119
+ "git_blob_sha1": "145dd0e22d5d4c790630da541bf5d614ebf299eb"
120
+ },
121
+ "model-00020-of-00039.safetensors": {
122
+ "size": 453593408,
123
+ "sha256": "cc7e2b39fe5fa95d7ab57e59a441b84b9db797f2c471a27a40858aba8b44e2f9",
124
+ "git_blob_sha1": "ed6a5135ebc804302827d6ccbb2903db21384c39"
125
+ },
126
+ "model-00021-of-00039.safetensors": {
127
+ "size": 369706848,
128
+ "sha256": "f29728fc6d5082a63b37fc2102219a9dcdc0e25cbbb3a6410ae103c361e1481a",
129
+ "git_blob_sha1": "5431ccec9174834805eac8be84a68edb2ba35ad1"
130
+ },
131
+ "model-00022-of-00039.safetensors": {
132
+ "size": 335553624,
133
+ "sha256": "2cecedd307549b444cd6f0494d0326618389b6152dafbd5d71339548378fcf48",
134
+ "git_blob_sha1": "3ca3926bc5e169a6cf22103bc2ba5fa7adb8e248"
135
+ },
136
+ "model-00023-of-00039.safetensors": {
137
+ "size": 235488984,
138
+ "sha256": "d3d64ac869e3bb591becafc2e0c5d66e5e2ffd3b5d1f7c44d7d0a8c72c605c56",
139
+ "git_blob_sha1": "f5b69ef92af642c682a0d3a25bd3bab368d604e7"
140
+ },
141
+ "model-00024-of-00039.safetensors": {
142
+ "size": 419449088,
143
+ "sha256": "291597762015e322d0dad94d3c827804a38753d74a80a5b2c368dcce61029da5",
144
+ "git_blob_sha1": "ad6e73388c216cf9884bbeaa4f17ce2a20876e3a"
145
+ },
146
+ "model-00025-of-00039.safetensors": {
147
+ "size": 403252936,
148
+ "sha256": "8b020e3572c9f0bb1e31b623ec2a9ef558bccca674eee605497c44967be258e0",
149
+ "git_blob_sha1": "c738f30dc6dd244c59a1f980288ee8316ed9a6f1"
150
+ },
151
+ "model-00026-of-00039.safetensors": {
152
+ "size": 403260920,
153
+ "sha256": "0255d8f2d3a5a89ed8170de0ad3dde766d5f36d9b8a523e05d50d99b86e85703",
154
+ "git_blob_sha1": "7e72d8be630fd5684a76a2ae6672e72f0cecb946"
155
+ },
156
+ "model-00027-of-00039.safetensors": {
157
+ "size": 402671504,
158
+ "sha256": "653fd99723ee73cd9e534984f9f907b69627d56f9c58a6eb4107f6ff02bba517",
159
+ "git_blob_sha1": "244176812b65db14120175fa9d63a4330d5ceffc"
160
+ },
161
+ "model-00028-of-00039.safetensors": {
162
+ "size": 453593400,
163
+ "sha256": "becbfe28e46d5ddd4584c21809cfba70e27207cd5780e044d25fed94abdc9678",
164
+ "git_blob_sha1": "8f0e7b8bc1954f34da20859c9aaeb9adce127487"
165
+ },
166
+ "model-00029-of-00039.safetensors": {
167
+ "size": 402671512,
168
+ "sha256": "fb4b2071ecd81ae7648b758f280d7a88112bfe9cf5d34432b13b4a23611f5579",
169
+ "git_blob_sha1": "f8e2b5cb349569d4c16b00c091bcb03ddefa632d"
170
+ },
171
+ "model-00030-of-00039.safetensors": {
172
+ "size": 453593392,
173
+ "sha256": "a6ceb2c95326871f53797dd85a4cf6dbdaf75194aa152c1fbf4748ccb499cddb",
174
+ "git_blob_sha1": "3d386a6297e8493c61becb7000a13dc47d64e8eb"
175
+ },
176
+ "model-00031-of-00039.safetensors": {
177
+ "size": 369706832,
178
+ "sha256": "6144600641fb09ab653b93f1c1dfb9a67b1481b57ce396e9eea442024137f5a3",
179
+ "git_blob_sha1": "88560496e9ce57b25518785c551cadf256711992"
180
+ },
181
+ "model-00032-of-00039.safetensors": {
182
+ "size": 335553616,
183
+ "sha256": "71345c0a84e522b12a4d790254677493c3afdb92ef9bfb8b6bb7e498ca77760f",
184
+ "git_blob_sha1": "1c846663e4b2c7682ac3dcb04f6ce995255d136a"
185
+ },
186
+ "model-00033-of-00039.safetensors": {
187
+ "size": 235488968,
188
+ "sha256": "25a51e76fb25120798b046452c84908438254a98b2b2eebd34d1c82b1138b7f7",
189
+ "git_blob_sha1": "348221fd59d69f6d780f56eaf24837ffd471ef89"
190
+ },
191
+ "model-00034-of-00039.safetensors": {
192
+ "size": 419449072,
193
+ "sha256": "e177052fc19a751336917730f200f6dc4b776981637395be0dea618f21be29d4",
194
+ "git_blob_sha1": "75ca02332f025286c83cc3ae83361043ad2cf7b5"
195
+ },
196
+ "model-00035-of-00039.safetensors": {
197
+ "size": 403252928,
198
+ "sha256": "51912607a0364190ed26dc0ca56a9c8211704fe09561741af477aa321c771035",
199
+ "git_blob_sha1": "09e85c872945bded1db281dfbbfe608363295305"
200
+ },
201
+ "model-00036-of-00039.safetensors": {
202
+ "size": 403260904,
203
+ "sha256": "24bf143f54cdc2d2a2a459d301dd0cc531fad48aaf49641edf2508b546d1a57f",
204
+ "git_blob_sha1": "fc7f1258bad68de9f0e66f808a03e09a8283bd7e"
205
+ },
206
+ "model-00037-of-00039.safetensors": {
207
+ "size": 67109496,
208
+ "sha256": "8d1ad0f6c5d0bc30c31ccb7da0d65fc93cf462d7bce42ca164066dec33cf022c",
209
+ "git_blob_sha1": "2eec1b927a690abbd496bf263c1c1507ab79eb1a"
210
+ },
211
+ "model-00038-of-00039.safetensors": {
212
+ "size": 2034237584,
213
+ "sha256": "3701c5312f013ca8c1fdc08911485167f9fe1f6520c88c51735c0b9567394847",
214
+ "git_blob_sha1": "dfb6ffcb45f26cd6d748255d56a68287b8e388e3"
215
+ },
216
+ "model-00039-of-00039.safetensors": {
217
+ "size": 2034237568,
218
+ "sha256": "d4496b8a8ac69b6b24d39b0b8a99cbf1c0e7d559b4815bee24af631ff32f771a",
219
+ "git_blob_sha1": "4aa897a590bf94e5e2ca389785b26c026800b54d"
220
+ },
221
+ "model.safetensors.index.json": {
222
+ "size": 35625,
223
+ "sha256": "cf88d770ef296c535f88249edbe3f7eb5531eb533f5b740012769ac692ed01f7",
224
+ "git_blob_sha1": "077afcfc7caa536c9b6e7d2545cd887893720920"
225
+ },
226
+ "tokenizer.json": {
227
+ "size": 19989325,
228
+ "sha256": "06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523",
229
+ "git_blob_sha1": "d5967dbafffa6c5b250a02b35723f53e702d0d62"
230
+ },
231
+ "tokenizer_config.json": {
232
+ "size": 1123,
233
+ "sha256": "bee8eba30f0eb4af73c0fe2cd06d0f89b657d7819941c438157ec42f7c80ea87",
234
+ "git_blob_sha1": "6913705fa6a05ee4498df98d776803ba8b39f413"
235
+ }
236
+ }
237
+ }
rl-agentic-new/step-30/config.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3_5ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_output_gate": true,
8
+ "bos_token_id": null,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 248044,
11
+ "full_attention_interval": 4,
12
+ "head_dim": 256,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 4096,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 12288,
17
+ "layer_types": [
18
+ "linear_attention",
19
+ "linear_attention",
20
+ "linear_attention",
21
+ "full_attention",
22
+ "linear_attention",
23
+ "linear_attention",
24
+ "linear_attention",
25
+ "full_attention",
26
+ "linear_attention",
27
+ "linear_attention",
28
+ "linear_attention",
29
+ "full_attention",
30
+ "linear_attention",
31
+ "linear_attention",
32
+ "linear_attention",
33
+ "full_attention",
34
+ "linear_attention",
35
+ "linear_attention",
36
+ "linear_attention",
37
+ "full_attention",
38
+ "linear_attention",
39
+ "linear_attention",
40
+ "linear_attention",
41
+ "full_attention",
42
+ "linear_attention",
43
+ "linear_attention",
44
+ "linear_attention",
45
+ "full_attention",
46
+ "linear_attention",
47
+ "linear_attention",
48
+ "linear_attention",
49
+ "full_attention"
50
+ ],
51
+ "linear_conv_kernel_dim": 4,
52
+ "linear_key_head_dim": 128,
53
+ "linear_num_key_heads": 16,
54
+ "linear_num_value_heads": 32,
55
+ "linear_value_head_dim": 128,
56
+ "mamba_ssm_dtype": "float32",
57
+ "max_position_embeddings": 262144,
58
+ "mlp_only_layers": [],
59
+ "model_type": "qwen3_5_text",
60
+ "mtp_num_hidden_layers": 1,
61
+ "mtp_use_dedicated_embeddings": false,
62
+ "num_attention_heads": 16,
63
+ "num_hidden_layers": 32,
64
+ "num_key_value_heads": 4,
65
+ "pad_token_id": null,
66
+ "partial_rotary_factor": 0.25,
67
+ "rms_norm_eps": 1e-06,
68
+ "rope_parameters": {
69
+ "mrope_interleaved": true,
70
+ "mrope_section": [
71
+ 11,
72
+ 11,
73
+ 10
74
+ ],
75
+ "partial_rotary_factor": 0.25,
76
+ "rope_theta": 10000000,
77
+ "rope_type": "default"
78
+ },
79
+ "tie_word_embeddings": false,
80
+ "transformers_version": "5.8.0",
81
+ "use_cache": true,
82
+ "vocab_size": 248320
83
+ }
rl-agentic-new/step-30/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "5.8.0"
4
+ }
rl-agentic-new/step-30/model-00001-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bacd6eedd12fd6a0715845a4aed4cb2384aac81f9e53a953b91ebdd560b1420f
3
+ size 403269512
rl-agentic-new/step-30/model-00002-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bca5b266c73f899edc090849249194c7183f8c28c38dc894527a464d25d6b2de
3
+ size 403260904
rl-agentic-new/step-30/model-00003-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac65870665f44a05550091342c5b606f693a0c6dfb11eca32e9bf18a986e803b
3
+ size 369706840
rl-agentic-new/step-30/model-00004-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b4eb9f87a8b8ab39b13411210a73f0b89f84f7ce825620c0c0d31e099d19b15
3
+ size 335553624
rl-agentic-new/step-30/model-00005-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b61604945b6358d805bc16eada86bac8ee0cbc130e8342c90db6413186e35ebd
3
+ size 520711312
rl-agentic-new/step-30/model-00006-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be05e3a17bb11385bc8be67867fc2b959bfb8ec8e99ef1523106536eb331d3c5
3
+ size 335553624
rl-agentic-new/step-30/model-00007-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9fe54f80af87542434ce93ada6720b849e80b2845ea786cfa603c16b4dc0fd5
3
+ size 235488984
rl-agentic-new/step-30/model-00008-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68546b899b4bc2233caab0ecf29024639270930e8e0003dbdb05868bc75b7e1a
3
+ size 403252936
rl-agentic-new/step-30/model-00009-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72c9b0c510bb49b8448efb3268fd0df36f549058b6a53433110d40f6d3b3c87e
3
+ size 453011952
rl-agentic-new/step-30/model-00010-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80d3b61f33ad4fd44c679a76875722057d73855da0000b95551b2b0b79f04936
3
+ size 403252936
rl-agentic-new/step-30/model-00011-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ef714f420b31fe318bae6859764c6e1d53ea5a6bd6b04590894052ca69c14d1
3
+ size 403260920
rl-agentic-new/step-30/model-00012-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bc4ec70a6c6fa5b55e9dd0d24d5fce30345e84583e6630c1a1237e85685de56
3
+ size 369706848
rl-agentic-new/step-30/model-00013-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85801b363a159cfcd8c054158f2595e09b1d6ce2a5d97d9640929dceaf2b8c1d
3
+ size 335553624
rl-agentic-new/step-30/model-00014-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3bbce708a05f17f953ab39b9661798e9d38d9a5e2ff3d607719fefebc0aea841
3
+ size 520711304
rl-agentic-new/step-30/model-00015-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be8353209865cdc41b49d43aa341f0ebb46fedd2bb8dd9514a59cc46acf47d64
3
+ size 335553616
rl-agentic-new/step-30/model-00016-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02c418927ad4e4295256ceecbb4f6866d76ccf088581188793a95f5c332a11d2
3
+ size 235488984
rl-agentic-new/step-30/model-00017-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:630c1c4262b5cfafa9b3500cce2b0a84a9a9a9b097fc9835c7d909254a099959
3
+ size 403252936
rl-agentic-new/step-30/model-00018-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d277e23614aa5b44420b4b174127ffb3d9afe5e47b7ba240ea231c246f9a29ea
3
+ size 403260920
rl-agentic-new/step-30/model-00019-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eba6375d3f8f00430cee6065a09ecd51e690be2640c5f95ee41756c67759f98e
3
+ size 402671512
rl-agentic-new/step-30/model-00020-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afc63578b62a585c811d0522863f8bbb4c1b64bea2430fabfcb1d8d336c9f161
3
+ size 453593408
rl-agentic-new/step-30/model-00021-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f05619da2d47ed9a0a61e9656383887f8ca2613fe6a03925500461a2f40415d6
3
+ size 369706848
rl-agentic-new/step-30/model-00022-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e8c1228dcb3ff2906706d8ba7377b276bb4f2f76bb38f09c0e1df95afd2f8bd
3
+ size 335553624
rl-agentic-new/step-30/model-00023-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65a205130548dddaf2e37c996ec14d126fe939436e84519e435abf92819d90a0
3
+ size 235488984
rl-agentic-new/step-30/model-00024-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7df7b3c02f267ea5869542b74d2239eb3e3bf85ca8372b43fa66a3c5d3b17a6
3
+ size 419449088
rl-agentic-new/step-30/model-00025-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:618e78e5e6801c99d1c288354fde68a8d70230149503b2727a07030db063eeb4
3
+ size 403252936
rl-agentic-new/step-30/model-00026-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68ba33a49fa5df792579cfb34934564ef42a68dfa49165b715493d6ea4abcdb2
3
+ size 403260920
rl-agentic-new/step-30/model-00027-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:319fe71b6b444c938ac8eef7ba0e5813b550a8e01d5a4b2e0e148442c8e2ea04
3
+ size 402671504
rl-agentic-new/step-30/model-00028-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a522bec391377149fd807b605ae7d0c8dee422704ca5ca24cb4299a3fe31c5fc
3
+ size 453593400
rl-agentic-new/step-30/model-00029-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0737d07ca082e5dac8571f283c1113b442b5ca37101ea4f16ef467971624c2a7
3
+ size 402671512
rl-agentic-new/step-30/model-00030-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41bec1ef0a6a2c1711c76ceefff939ad8860d55bb95f4bda88776631dacb0d11
3
+ size 453593392
rl-agentic-new/step-30/model-00031-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b0f769841d965f2c289492681d495d17f29bcd86826fdeba17524f127a218e3
3
+ size 369706832
rl-agentic-new/step-30/model-00032-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd9c0a920568a2a84289b2d94e4efdf0cd37b4959611ae65a807c4c7b2008619
3
+ size 335553616
rl-agentic-new/step-30/model-00033-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:528261986d10360a0674eb1c7733a4b37edc312115d245988c3ce2dc306eecd4
3
+ size 235488968
rl-agentic-new/step-30/model-00034-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:974f135da64e110fa4631092ea28f6e210c42abf6cc68b38c67cf32d374e415e
3
+ size 419449072
rl-agentic-new/step-30/model-00035-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:149cc86400fcfa6484210014621e0352b4c131148d77731b15472daa4321dae3
3
+ size 403252928
rl-agentic-new/step-30/model-00036-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:842b48f5c78f3223e94f0f730926a586ad37c530d0fd10d5f99524fee5e9befa
3
+ size 403260904
rl-agentic-new/step-30/model-00037-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02711b75302e81d158d6c6e7fcde4721c025bc96406c65a4a1f41a339dd71921
3
+ size 67109496
rl-agentic-new/step-30/model-00038-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7da56ba5e3d906db8d400e1c78de3b9cc3c3e2d1dfb4ae8204ce93fcd93c143
3
+ size 2034237584
rl-agentic-new/step-30/model-00039-of-00039.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77409f566d541841fab6f40d1a809ebf83d80e31a1cb8742770e330484d404d1
3
+ size 2034237568
rl-agentic-new/step-30/model.safetensors.index.json ADDED
@@ -0,0 +1,434 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 17907606528
4
+ },
5
+ "weight_map": {
6
+ "model.norm.weight": "model-00001-of-00039.safetensors",
7
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00039.safetensors",
8
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00039.safetensors",
9
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00039.safetensors",
10
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00039.safetensors",
11
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00039.safetensors",
12
+ "model.layers.0.linear_attn.A_log": "model-00001-of-00039.safetensors",
13
+ "model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00039.safetensors",
14
+ "model.layers.0.linear_attn.dt_bias": "model-00001-of-00039.safetensors",
15
+ "model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00039.safetensors",
16
+ "model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00039.safetensors",
17
+ "model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00039.safetensors",
18
+ "model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00039.safetensors",
19
+ "model.layers.0.linear_attn.norm.weight": "model-00001-of-00039.safetensors",
20
+ "model.layers.11.mlp.down_proj.weight": "model-00005-of-00039.safetensors",
21
+ "model.layers.11.self_attn.k_norm.weight": "model-00005-of-00039.safetensors",
22
+ "model.layers.11.self_attn.o_proj.weight": "model-00005-of-00039.safetensors",
23
+ "model.layers.11.input_layernorm.weight": "model-00005-of-00039.safetensors",
24
+ "model.layers.11.self_attn.q_proj.weight": "model-00005-of-00039.safetensors",
25
+ "model.layers.11.self_attn.k_proj.weight": "model-00005-of-00039.safetensors",
26
+ "model.layers.11.self_attn.v_proj.weight": "model-00005-of-00039.safetensors",
27
+ "model.layers.11.self_attn.q_norm.weight": "model-00005-of-00039.safetensors",
28
+ "model.layers.12.post_attention_layernorm.weight": "model-00005-of-00039.safetensors",
29
+ "model.layers.12.mlp.gate_proj.weight": "model-00005-of-00039.safetensors",
30
+ "model.layers.12.mlp.up_proj.weight": "model-00005-of-00039.safetensors",
31
+ "model.layers.12.mlp.down_proj.weight": "model-00005-of-00039.safetensors",
32
+ "model.layers.12.input_layernorm.weight": "model-00005-of-00039.safetensors",
33
+ "model.layers.12.linear_attn.A_log": "model-00005-of-00039.safetensors",
34
+ "model.layers.12.linear_attn.conv1d.weight": "model-00005-of-00039.safetensors",
35
+ "model.layers.12.linear_attn.dt_bias": "model-00005-of-00039.safetensors",
36
+ "model.layers.12.linear_attn.in_proj_a.weight": "model-00005-of-00039.safetensors",
37
+ "model.layers.12.linear_attn.in_proj_b.weight": "model-00005-of-00039.safetensors",
38
+ "model.layers.14.linear_attn.out_proj.weight": "model-00009-of-00039.safetensors",
39
+ "model.layers.15.post_attention_layernorm.weight": "model-00009-of-00039.safetensors",
40
+ "model.layers.15.mlp.gate_proj.weight": "model-00009-of-00039.safetensors",
41
+ "model.layers.15.mlp.up_proj.weight": "model-00009-of-00039.safetensors",
42
+ "model.layers.15.mlp.down_proj.weight": "model-00009-of-00039.safetensors",
43
+ "model.layers.15.self_attn.k_norm.weight": "model-00009-of-00039.safetensors",
44
+ "model.layers.15.self_attn.o_proj.weight": "model-00009-of-00039.safetensors",
45
+ "model.layers.15.input_layernorm.weight": "model-00009-of-00039.safetensors",
46
+ "model.layers.15.self_attn.q_proj.weight": "model-00009-of-00039.safetensors",
47
+ "model.layers.15.self_attn.k_proj.weight": "model-00009-of-00039.safetensors",
48
+ "model.layers.15.self_attn.v_proj.weight": "model-00009-of-00039.safetensors",
49
+ "model.layers.15.self_attn.q_norm.weight": "model-00009-of-00039.safetensors",
50
+ "model.layers.16.post_attention_layernorm.weight": "model-00009-of-00039.safetensors",
51
+ "model.layers.18.linear_attn.in_proj_qkv.weight": "model-00013-of-00039.safetensors",
52
+ "model.layers.18.linear_attn.in_proj_z.weight": "model-00013-of-00039.safetensors",
53
+ "model.layers.18.linear_attn.norm.weight": "model-00013-of-00039.safetensors",
54
+ "model.layers.18.linear_attn.out_proj.weight": "model-00013-of-00039.safetensors",
55
+ "model.layers.19.post_attention_layernorm.weight": "model-00013-of-00039.safetensors",
56
+ "model.layers.19.mlp.gate_proj.weight": "model-00013-of-00039.safetensors",
57
+ "model.layers.19.mlp.up_proj.weight": "model-00013-of-00039.safetensors",
58
+ "model.layers.21.mlp.gate_proj.weight": "model-00017-of-00039.safetensors",
59
+ "model.layers.21.mlp.up_proj.weight": "model-00017-of-00039.safetensors",
60
+ "model.layers.21.mlp.down_proj.weight": "model-00017-of-00039.safetensors",
61
+ "model.layers.21.input_layernorm.weight": "model-00017-of-00039.safetensors",
62
+ "model.layers.21.linear_attn.A_log": "model-00017-of-00039.safetensors",
63
+ "model.layers.21.linear_attn.conv1d.weight": "model-00017-of-00039.safetensors",
64
+ "model.layers.21.linear_attn.dt_bias": "model-00017-of-00039.safetensors",
65
+ "model.layers.21.linear_attn.in_proj_a.weight": "model-00017-of-00039.safetensors",
66
+ "model.layers.21.linear_attn.in_proj_b.weight": "model-00017-of-00039.safetensors",
67
+ "model.layers.21.linear_attn.in_proj_qkv.weight": "model-00017-of-00039.safetensors",
68
+ "model.layers.21.linear_attn.in_proj_z.weight": "model-00017-of-00039.safetensors",
69
+ "model.layers.21.linear_attn.norm.weight": "model-00017-of-00039.safetensors",
70
+ "model.layers.24.linear_attn.in_proj_z.weight": "model-00021-of-00039.safetensors",
71
+ "model.layers.24.linear_attn.norm.weight": "model-00021-of-00039.safetensors",
72
+ "model.layers.24.linear_attn.out_proj.weight": "model-00021-of-00039.safetensors",
73
+ "model.layers.25.post_attention_layernorm.weight": "model-00021-of-00039.safetensors",
74
+ "model.layers.25.mlp.gate_proj.weight": "model-00021-of-00039.safetensors",
75
+ "model.layers.25.mlp.up_proj.weight": "model-00021-of-00039.safetensors",
76
+ "model.layers.25.mlp.down_proj.weight": "model-00021-of-00039.safetensors",
77
+ "model.layers.25.input_layernorm.weight": "model-00021-of-00039.safetensors",
78
+ "model.layers.25.linear_attn.A_log": "model-00021-of-00039.safetensors",
79
+ "model.layers.25.linear_attn.conv1d.weight": "model-00021-of-00039.safetensors",
80
+ "model.layers.25.linear_attn.dt_bias": "model-00021-of-00039.safetensors",
81
+ "model.layers.25.linear_attn.in_proj_a.weight": "model-00021-of-00039.safetensors",
82
+ "model.layers.25.linear_attn.in_proj_b.weight": "model-00021-of-00039.safetensors",
83
+ "model.layers.28.mlp.gate_proj.weight": "model-00025-of-00039.safetensors",
84
+ "model.layers.28.mlp.up_proj.weight": "model-00025-of-00039.safetensors",
85
+ "model.layers.28.mlp.down_proj.weight": "model-00025-of-00039.safetensors",
86
+ "model.layers.28.input_layernorm.weight": "model-00025-of-00039.safetensors",
87
+ "model.layers.28.linear_attn.A_log": "model-00025-of-00039.safetensors",
88
+ "model.layers.28.linear_attn.conv1d.weight": "model-00025-of-00039.safetensors",
89
+ "model.layers.28.linear_attn.dt_bias": "model-00025-of-00039.safetensors",
90
+ "model.layers.28.linear_attn.in_proj_a.weight": "model-00025-of-00039.safetensors",
91
+ "model.layers.28.linear_attn.in_proj_b.weight": "model-00025-of-00039.safetensors",
92
+ "model.layers.28.linear_attn.in_proj_qkv.weight": "model-00025-of-00039.safetensors",
93
+ "model.layers.28.linear_attn.in_proj_z.weight": "model-00025-of-00039.safetensors",
94
+ "model.layers.28.linear_attn.norm.weight": "model-00025-of-00039.safetensors",
95
+ "model.layers.30.linear_attn.in_proj_z.weight": "model-00029-of-00039.safetensors",
96
+ "model.layers.30.linear_attn.norm.weight": "model-00029-of-00039.safetensors",
97
+ "model.layers.30.linear_attn.out_proj.weight": "model-00029-of-00039.safetensors",
98
+ "model.layers.31.post_attention_layernorm.weight": "model-00029-of-00039.safetensors",
99
+ "model.layers.31.mlp.gate_proj.weight": "model-00029-of-00039.safetensors",
100
+ "model.layers.31.mlp.up_proj.weight": "model-00029-of-00039.safetensors",
101
+ "model.layers.31.mlp.down_proj.weight": "model-00029-of-00039.safetensors",
102
+ "model.layers.31.self_attn.k_norm.weight": "model-00029-of-00039.safetensors",
103
+ "model.layers.31.self_attn.o_proj.weight": "model-00029-of-00039.safetensors",
104
+ "model.layers.31.input_layernorm.weight": "model-00029-of-00039.safetensors",
105
+ "model.layers.6.mlp.down_proj.weight": "model-00033-of-00039.safetensors",
106
+ "model.layers.6.input_layernorm.weight": "model-00033-of-00039.safetensors",
107
+ "model.layers.6.linear_attn.A_log": "model-00033-of-00039.safetensors",
108
+ "model.layers.6.linear_attn.conv1d.weight": "model-00033-of-00039.safetensors",
109
+ "model.layers.6.linear_attn.dt_bias": "model-00033-of-00039.safetensors",
110
+ "model.layers.6.linear_attn.in_proj_a.weight": "model-00033-of-00039.safetensors",
111
+ "model.layers.6.linear_attn.in_proj_b.weight": "model-00033-of-00039.safetensors",
112
+ "model.layers.6.linear_attn.in_proj_qkv.weight": "model-00033-of-00039.safetensors",
113
+ "model.layers.6.linear_attn.in_proj_z.weight": "model-00033-of-00039.safetensors",
114
+ "model.layers.6.linear_attn.norm.weight": "model-00033-of-00039.safetensors",
115
+ "model.layers.6.linear_attn.out_proj.weight": "model-00033-of-00039.safetensors",
116
+ "model.layers.7.post_attention_layernorm.weight": "model-00033-of-00039.safetensors",
117
+ "model.layers.9.linear_attn.in_proj_z.weight": "model-00037-of-00039.safetensors",
118
+ "model.layers.9.linear_attn.norm.weight": "model-00037-of-00039.safetensors",
119
+ "model.layers.9.linear_attn.out_proj.weight": "model-00037-of-00039.safetensors",
120
+ "model.layers.0.linear_attn.out_proj.weight": "model-00002-of-00039.safetensors",
121
+ "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00039.safetensors",
122
+ "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00039.safetensors",
123
+ "model.layers.1.mlp.up_proj.weight": "model-00002-of-00039.safetensors",
124
+ "model.layers.1.mlp.down_proj.weight": "model-00002-of-00039.safetensors",
125
+ "model.layers.1.input_layernorm.weight": "model-00002-of-00039.safetensors",
126
+ "model.layers.1.linear_attn.A_log": "model-00002-of-00039.safetensors",
127
+ "model.layers.1.linear_attn.conv1d.weight": "model-00002-of-00039.safetensors",
128
+ "model.layers.1.linear_attn.dt_bias": "model-00002-of-00039.safetensors",
129
+ "model.layers.1.linear_attn.in_proj_a.weight": "model-00002-of-00039.safetensors",
130
+ "model.layers.1.linear_attn.in_proj_b.weight": "model-00002-of-00039.safetensors",
131
+ "model.layers.1.linear_attn.in_proj_qkv.weight": "model-00002-of-00039.safetensors",
132
+ "model.layers.12.linear_attn.in_proj_qkv.weight": "model-00006-of-00039.safetensors",
133
+ "model.layers.12.linear_attn.in_proj_z.weight": "model-00006-of-00039.safetensors",
134
+ "model.layers.12.linear_attn.norm.weight": "model-00006-of-00039.safetensors",
135
+ "model.layers.12.linear_attn.out_proj.weight": "model-00006-of-00039.safetensors",
136
+ "model.layers.13.post_attention_layernorm.weight": "model-00006-of-00039.safetensors",
137
+ "model.layers.13.mlp.gate_proj.weight": "model-00006-of-00039.safetensors",
138
+ "model.layers.13.mlp.up_proj.weight": "model-00006-of-00039.safetensors",
139
+ "model.layers.16.mlp.gate_proj.weight": "model-00010-of-00039.safetensors",
140
+ "model.layers.16.mlp.up_proj.weight": "model-00010-of-00039.safetensors",
141
+ "model.layers.16.mlp.down_proj.weight": "model-00010-of-00039.safetensors",
142
+ "model.layers.16.input_layernorm.weight": "model-00010-of-00039.safetensors",
143
+ "model.layers.16.linear_attn.A_log": "model-00010-of-00039.safetensors",
144
+ "model.layers.16.linear_attn.conv1d.weight": "model-00010-of-00039.safetensors",
145
+ "model.layers.16.linear_attn.dt_bias": "model-00010-of-00039.safetensors",
146
+ "model.layers.16.linear_attn.in_proj_a.weight": "model-00010-of-00039.safetensors",
147
+ "model.layers.16.linear_attn.in_proj_b.weight": "model-00010-of-00039.safetensors",
148
+ "model.layers.16.linear_attn.in_proj_qkv.weight": "model-00010-of-00039.safetensors",
149
+ "model.layers.16.linear_attn.in_proj_z.weight": "model-00010-of-00039.safetensors",
150
+ "model.layers.16.linear_attn.norm.weight": "model-00010-of-00039.safetensors",
151
+ "model.layers.19.mlp.down_proj.weight": "model-00014-of-00039.safetensors",
152
+ "model.layers.19.self_attn.k_norm.weight": "model-00014-of-00039.safetensors",
153
+ "model.layers.19.self_attn.o_proj.weight": "model-00014-of-00039.safetensors",
154
+ "model.layers.19.input_layernorm.weight": "model-00014-of-00039.safetensors",
155
+ "model.layers.19.self_attn.q_proj.weight": "model-00014-of-00039.safetensors",
156
+ "model.layers.19.self_attn.k_proj.weight": "model-00014-of-00039.safetensors",
157
+ "model.layers.19.self_attn.v_proj.weight": "model-00014-of-00039.safetensors",
158
+ "model.layers.19.self_attn.q_norm.weight": "model-00014-of-00039.safetensors",
159
+ "model.layers.2.post_attention_layernorm.weight": "model-00014-of-00039.safetensors",
160
+ "model.layers.2.mlp.gate_proj.weight": "model-00014-of-00039.safetensors",
161
+ "model.layers.2.mlp.up_proj.weight": "model-00014-of-00039.safetensors",
162
+ "model.layers.2.mlp.down_proj.weight": "model-00014-of-00039.safetensors",
163
+ "model.layers.2.input_layernorm.weight": "model-00014-of-00039.safetensors",
164
+ "model.layers.2.linear_attn.A_log": "model-00014-of-00039.safetensors",
165
+ "model.layers.2.linear_attn.conv1d.weight": "model-00014-of-00039.safetensors",
166
+ "model.layers.2.linear_attn.dt_bias": "model-00014-of-00039.safetensors",
167
+ "model.layers.2.linear_attn.in_proj_a.weight": "model-00014-of-00039.safetensors",
168
+ "model.layers.2.linear_attn.in_proj_b.weight": "model-00014-of-00039.safetensors",
169
+ "model.layers.21.linear_attn.out_proj.weight": "model-00018-of-00039.safetensors",
170
+ "model.layers.22.post_attention_layernorm.weight": "model-00018-of-00039.safetensors",
171
+ "model.layers.22.mlp.gate_proj.weight": "model-00018-of-00039.safetensors",
172
+ "model.layers.22.mlp.up_proj.weight": "model-00018-of-00039.safetensors",
173
+ "model.layers.22.mlp.down_proj.weight": "model-00018-of-00039.safetensors",
174
+ "model.layers.22.input_layernorm.weight": "model-00018-of-00039.safetensors",
175
+ "model.layers.22.linear_attn.A_log": "model-00018-of-00039.safetensors",
176
+ "model.layers.22.linear_attn.conv1d.weight": "model-00018-of-00039.safetensors",
177
+ "model.layers.22.linear_attn.dt_bias": "model-00018-of-00039.safetensors",
178
+ "model.layers.22.linear_attn.in_proj_a.weight": "model-00018-of-00039.safetensors",
179
+ "model.layers.22.linear_attn.in_proj_b.weight": "model-00018-of-00039.safetensors",
180
+ "model.layers.22.linear_attn.in_proj_qkv.weight": "model-00018-of-00039.safetensors",
181
+ "model.layers.25.linear_attn.in_proj_qkv.weight": "model-00022-of-00039.safetensors",
182
+ "model.layers.25.linear_attn.in_proj_z.weight": "model-00022-of-00039.safetensors",
183
+ "model.layers.25.linear_attn.norm.weight": "model-00022-of-00039.safetensors",
184
+ "model.layers.25.linear_attn.out_proj.weight": "model-00022-of-00039.safetensors",
185
+ "model.layers.26.post_attention_layernorm.weight": "model-00022-of-00039.safetensors",
186
+ "model.layers.26.mlp.gate_proj.weight": "model-00022-of-00039.safetensors",
187
+ "model.layers.26.mlp.up_proj.weight": "model-00022-of-00039.safetensors",
188
+ "model.layers.28.linear_attn.out_proj.weight": "model-00026-of-00039.safetensors",
189
+ "model.layers.29.post_attention_layernorm.weight": "model-00026-of-00039.safetensors",
190
+ "model.layers.29.mlp.gate_proj.weight": "model-00026-of-00039.safetensors",
191
+ "model.layers.29.mlp.up_proj.weight": "model-00026-of-00039.safetensors",
192
+ "model.layers.29.mlp.down_proj.weight": "model-00026-of-00039.safetensors",
193
+ "model.layers.29.input_layernorm.weight": "model-00026-of-00039.safetensors",
194
+ "model.layers.29.linear_attn.A_log": "model-00026-of-00039.safetensors",
195
+ "model.layers.29.linear_attn.conv1d.weight": "model-00026-of-00039.safetensors",
196
+ "model.layers.29.linear_attn.dt_bias": "model-00026-of-00039.safetensors",
197
+ "model.layers.29.linear_attn.in_proj_a.weight": "model-00026-of-00039.safetensors",
198
+ "model.layers.29.linear_attn.in_proj_b.weight": "model-00026-of-00039.safetensors",
199
+ "model.layers.29.linear_attn.in_proj_qkv.weight": "model-00026-of-00039.safetensors",
200
+ "model.layers.31.self_attn.q_proj.weight": "model-00030-of-00039.safetensors",
201
+ "model.layers.31.self_attn.k_proj.weight": "model-00030-of-00039.safetensors",
202
+ "model.layers.31.self_attn.v_proj.weight": "model-00030-of-00039.safetensors",
203
+ "model.layers.31.self_attn.q_norm.weight": "model-00030-of-00039.safetensors",
204
+ "model.layers.4.post_attention_layernorm.weight": "model-00030-of-00039.safetensors",
205
+ "model.layers.4.mlp.gate_proj.weight": "model-00030-of-00039.safetensors",
206
+ "model.layers.4.mlp.up_proj.weight": "model-00030-of-00039.safetensors",
207
+ "model.layers.4.mlp.down_proj.weight": "model-00030-of-00039.safetensors",
208
+ "model.layers.4.input_layernorm.weight": "model-00030-of-00039.safetensors",
209
+ "model.layers.4.linear_attn.A_log": "model-00030-of-00039.safetensors",
210
+ "model.layers.4.linear_attn.conv1d.weight": "model-00030-of-00039.safetensors",
211
+ "model.layers.4.linear_attn.dt_bias": "model-00030-of-00039.safetensors",
212
+ "model.layers.4.linear_attn.in_proj_a.weight": "model-00030-of-00039.safetensors",
213
+ "model.layers.4.linear_attn.in_proj_b.weight": "model-00030-of-00039.safetensors",
214
+ "model.layers.4.linear_attn.in_proj_qkv.weight": "model-00030-of-00039.safetensors",
215
+ "model.layers.7.mlp.gate_proj.weight": "model-00034-of-00039.safetensors",
216
+ "model.layers.7.mlp.up_proj.weight": "model-00034-of-00039.safetensors",
217
+ "model.layers.7.mlp.down_proj.weight": "model-00034-of-00039.safetensors",
218
+ "model.layers.7.self_attn.k_norm.weight": "model-00034-of-00039.safetensors",
219
+ "model.layers.7.self_attn.o_proj.weight": "model-00034-of-00039.safetensors",
220
+ "model.layers.7.input_layernorm.weight": "model-00034-of-00039.safetensors",
221
+ "model.layers.7.self_attn.q_proj.weight": "model-00034-of-00039.safetensors",
222
+ "model.layers.7.self_attn.k_proj.weight": "model-00034-of-00039.safetensors",
223
+ "model.layers.7.self_attn.v_proj.weight": "model-00034-of-00039.safetensors",
224
+ "model.layers.7.self_attn.q_norm.weight": "model-00034-of-00039.safetensors",
225
+ "model.layers.8.post_attention_layernorm.weight": "model-00034-of-00039.safetensors",
226
+ "model.embed_tokens.weight": "model-00038-of-00039.safetensors",
227
+ "model.layers.1.linear_attn.in_proj_z.weight": "model-00003-of-00039.safetensors",
228
+ "model.layers.1.linear_attn.norm.weight": "model-00003-of-00039.safetensors",
229
+ "model.layers.1.linear_attn.out_proj.weight": "model-00003-of-00039.safetensors",
230
+ "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00039.safetensors",
231
+ "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00039.safetensors",
232
+ "model.layers.10.mlp.up_proj.weight": "model-00003-of-00039.safetensors",
233
+ "model.layers.10.mlp.down_proj.weight": "model-00003-of-00039.safetensors",
234
+ "model.layers.10.input_layernorm.weight": "model-00003-of-00039.safetensors",
235
+ "model.layers.10.linear_attn.A_log": "model-00003-of-00039.safetensors",
236
+ "model.layers.10.linear_attn.conv1d.weight": "model-00003-of-00039.safetensors",
237
+ "model.layers.10.linear_attn.dt_bias": "model-00003-of-00039.safetensors",
238
+ "model.layers.10.linear_attn.in_proj_a.weight": "model-00003-of-00039.safetensors",
239
+ "model.layers.10.linear_attn.in_proj_b.weight": "model-00003-of-00039.safetensors",
240
+ "model.layers.13.mlp.down_proj.weight": "model-00007-of-00039.safetensors",
241
+ "model.layers.13.input_layernorm.weight": "model-00007-of-00039.safetensors",
242
+ "model.layers.13.linear_attn.A_log": "model-00007-of-00039.safetensors",
243
+ "model.layers.13.linear_attn.conv1d.weight": "model-00007-of-00039.safetensors",
244
+ "model.layers.13.linear_attn.dt_bias": "model-00007-of-00039.safetensors",
245
+ "model.layers.13.linear_attn.in_proj_a.weight": "model-00007-of-00039.safetensors",
246
+ "model.layers.13.linear_attn.in_proj_b.weight": "model-00007-of-00039.safetensors",
247
+ "model.layers.13.linear_attn.in_proj_qkv.weight": "model-00007-of-00039.safetensors",
248
+ "model.layers.13.linear_attn.in_proj_z.weight": "model-00007-of-00039.safetensors",
249
+ "model.layers.13.linear_attn.norm.weight": "model-00007-of-00039.safetensors",
250
+ "model.layers.13.linear_attn.out_proj.weight": "model-00007-of-00039.safetensors",
251
+ "model.layers.14.post_attention_layernorm.weight": "model-00007-of-00039.safetensors",
252
+ "model.layers.16.linear_attn.out_proj.weight": "model-00011-of-00039.safetensors",
253
+ "model.layers.17.post_attention_layernorm.weight": "model-00011-of-00039.safetensors",
254
+ "model.layers.17.mlp.gate_proj.weight": "model-00011-of-00039.safetensors",
255
+ "model.layers.17.mlp.up_proj.weight": "model-00011-of-00039.safetensors",
256
+ "model.layers.17.mlp.down_proj.weight": "model-00011-of-00039.safetensors",
257
+ "model.layers.17.input_layernorm.weight": "model-00011-of-00039.safetensors",
258
+ "model.layers.17.linear_attn.A_log": "model-00011-of-00039.safetensors",
259
+ "model.layers.17.linear_attn.conv1d.weight": "model-00011-of-00039.safetensors",
260
+ "model.layers.17.linear_attn.dt_bias": "model-00011-of-00039.safetensors",
261
+ "model.layers.17.linear_attn.in_proj_a.weight": "model-00011-of-00039.safetensors",
262
+ "model.layers.17.linear_attn.in_proj_b.weight": "model-00011-of-00039.safetensors",
263
+ "model.layers.17.linear_attn.in_proj_qkv.weight": "model-00011-of-00039.safetensors",
264
+ "model.layers.2.linear_attn.in_proj_qkv.weight": "model-00015-of-00039.safetensors",
265
+ "model.layers.2.linear_attn.in_proj_z.weight": "model-00015-of-00039.safetensors",
266
+ "model.layers.2.linear_attn.norm.weight": "model-00015-of-00039.safetensors",
267
+ "model.layers.2.linear_attn.out_proj.weight": "model-00015-of-00039.safetensors",
268
+ "model.layers.20.post_attention_layernorm.weight": "model-00015-of-00039.safetensors",
269
+ "model.layers.20.mlp.gate_proj.weight": "model-00015-of-00039.safetensors",
270
+ "model.layers.20.mlp.up_proj.weight": "model-00015-of-00039.safetensors",
271
+ "model.layers.22.linear_attn.in_proj_z.weight": "model-00019-of-00039.safetensors",
272
+ "model.layers.22.linear_attn.norm.weight": "model-00019-of-00039.safetensors",
273
+ "model.layers.22.linear_attn.out_proj.weight": "model-00019-of-00039.safetensors",
274
+ "model.layers.23.post_attention_layernorm.weight": "model-00019-of-00039.safetensors",
275
+ "model.layers.23.mlp.gate_proj.weight": "model-00019-of-00039.safetensors",
276
+ "model.layers.23.mlp.up_proj.weight": "model-00019-of-00039.safetensors",
277
+ "model.layers.23.mlp.down_proj.weight": "model-00019-of-00039.safetensors",
278
+ "model.layers.23.self_attn.k_norm.weight": "model-00019-of-00039.safetensors",
279
+ "model.layers.23.self_attn.o_proj.weight": "model-00019-of-00039.safetensors",
280
+ "model.layers.23.input_layernorm.weight": "model-00019-of-00039.safetensors",
281
+ "model.layers.26.mlp.down_proj.weight": "model-00023-of-00039.safetensors",
282
+ "model.layers.26.input_layernorm.weight": "model-00023-of-00039.safetensors",
283
+ "model.layers.26.linear_attn.A_log": "model-00023-of-00039.safetensors",
284
+ "model.layers.26.linear_attn.conv1d.weight": "model-00023-of-00039.safetensors",
285
+ "model.layers.26.linear_attn.dt_bias": "model-00023-of-00039.safetensors",
286
+ "model.layers.26.linear_attn.in_proj_a.weight": "model-00023-of-00039.safetensors",
287
+ "model.layers.26.linear_attn.in_proj_b.weight": "model-00023-of-00039.safetensors",
288
+ "model.layers.26.linear_attn.in_proj_qkv.weight": "model-00023-of-00039.safetensors",
289
+ "model.layers.26.linear_attn.in_proj_z.weight": "model-00023-of-00039.safetensors",
290
+ "model.layers.26.linear_attn.norm.weight": "model-00023-of-00039.safetensors",
291
+ "model.layers.26.linear_attn.out_proj.weight": "model-00023-of-00039.safetensors",
292
+ "model.layers.27.post_attention_layernorm.weight": "model-00023-of-00039.safetensors",
293
+ "model.layers.29.linear_attn.in_proj_z.weight": "model-00027-of-00039.safetensors",
294
+ "model.layers.29.linear_attn.norm.weight": "model-00027-of-00039.safetensors",
295
+ "model.layers.29.linear_attn.out_proj.weight": "model-00027-of-00039.safetensors",
296
+ "model.layers.3.post_attention_layernorm.weight": "model-00027-of-00039.safetensors",
297
+ "model.layers.3.mlp.gate_proj.weight": "model-00027-of-00039.safetensors",
298
+ "model.layers.3.mlp.up_proj.weight": "model-00027-of-00039.safetensors",
299
+ "model.layers.3.mlp.down_proj.weight": "model-00027-of-00039.safetensors",
300
+ "model.layers.3.self_attn.k_norm.weight": "model-00027-of-00039.safetensors",
301
+ "model.layers.3.self_attn.o_proj.weight": "model-00027-of-00039.safetensors",
302
+ "model.layers.3.input_layernorm.weight": "model-00027-of-00039.safetensors",
303
+ "model.layers.4.linear_attn.in_proj_z.weight": "model-00031-of-00039.safetensors",
304
+ "model.layers.4.linear_attn.norm.weight": "model-00031-of-00039.safetensors",
305
+ "model.layers.4.linear_attn.out_proj.weight": "model-00031-of-00039.safetensors",
306
+ "model.layers.5.post_attention_layernorm.weight": "model-00031-of-00039.safetensors",
307
+ "model.layers.5.mlp.gate_proj.weight": "model-00031-of-00039.safetensors",
308
+ "model.layers.5.mlp.up_proj.weight": "model-00031-of-00039.safetensors",
309
+ "model.layers.5.mlp.down_proj.weight": "model-00031-of-00039.safetensors",
310
+ "model.layers.5.input_layernorm.weight": "model-00031-of-00039.safetensors",
311
+ "model.layers.5.linear_attn.A_log": "model-00031-of-00039.safetensors",
312
+ "model.layers.5.linear_attn.conv1d.weight": "model-00031-of-00039.safetensors",
313
+ "model.layers.5.linear_attn.dt_bias": "model-00031-of-00039.safetensors",
314
+ "model.layers.5.linear_attn.in_proj_a.weight": "model-00031-of-00039.safetensors",
315
+ "model.layers.5.linear_attn.in_proj_b.weight": "model-00031-of-00039.safetensors",
316
+ "model.layers.8.mlp.gate_proj.weight": "model-00035-of-00039.safetensors",
317
+ "model.layers.8.mlp.up_proj.weight": "model-00035-of-00039.safetensors",
318
+ "model.layers.8.mlp.down_proj.weight": "model-00035-of-00039.safetensors",
319
+ "model.layers.8.input_layernorm.weight": "model-00035-of-00039.safetensors",
320
+ "model.layers.8.linear_attn.A_log": "model-00035-of-00039.safetensors",
321
+ "model.layers.8.linear_attn.conv1d.weight": "model-00035-of-00039.safetensors",
322
+ "model.layers.8.linear_attn.dt_bias": "model-00035-of-00039.safetensors",
323
+ "model.layers.8.linear_attn.in_proj_a.weight": "model-00035-of-00039.safetensors",
324
+ "model.layers.8.linear_attn.in_proj_b.weight": "model-00035-of-00039.safetensors",
325
+ "model.layers.8.linear_attn.in_proj_qkv.weight": "model-00035-of-00039.safetensors",
326
+ "model.layers.8.linear_attn.in_proj_z.weight": "model-00035-of-00039.safetensors",
327
+ "model.layers.8.linear_attn.norm.weight": "model-00035-of-00039.safetensors",
328
+ "lm_head.weight": "model-00039-of-00039.safetensors",
329
+ "model.layers.10.linear_attn.in_proj_qkv.weight": "model-00004-of-00039.safetensors",
330
+ "model.layers.10.linear_attn.in_proj_z.weight": "model-00004-of-00039.safetensors",
331
+ "model.layers.10.linear_attn.norm.weight": "model-00004-of-00039.safetensors",
332
+ "model.layers.10.linear_attn.out_proj.weight": "model-00004-of-00039.safetensors",
333
+ "model.layers.11.post_attention_layernorm.weight": "model-00004-of-00039.safetensors",
334
+ "model.layers.11.mlp.gate_proj.weight": "model-00004-of-00039.safetensors",
335
+ "model.layers.11.mlp.up_proj.weight": "model-00004-of-00039.safetensors",
336
+ "model.layers.14.mlp.gate_proj.weight": "model-00008-of-00039.safetensors",
337
+ "model.layers.14.mlp.up_proj.weight": "model-00008-of-00039.safetensors",
338
+ "model.layers.14.mlp.down_proj.weight": "model-00008-of-00039.safetensors",
339
+ "model.layers.14.input_layernorm.weight": "model-00008-of-00039.safetensors",
340
+ "model.layers.14.linear_attn.A_log": "model-00008-of-00039.safetensors",
341
+ "model.layers.14.linear_attn.conv1d.weight": "model-00008-of-00039.safetensors",
342
+ "model.layers.14.linear_attn.dt_bias": "model-00008-of-00039.safetensors",
343
+ "model.layers.14.linear_attn.in_proj_a.weight": "model-00008-of-00039.safetensors",
344
+ "model.layers.14.linear_attn.in_proj_b.weight": "model-00008-of-00039.safetensors",
345
+ "model.layers.14.linear_attn.in_proj_qkv.weight": "model-00008-of-00039.safetensors",
346
+ "model.layers.14.linear_attn.in_proj_z.weight": "model-00008-of-00039.safetensors",
347
+ "model.layers.14.linear_attn.norm.weight": "model-00008-of-00039.safetensors",
348
+ "model.layers.17.linear_attn.in_proj_z.weight": "model-00012-of-00039.safetensors",
349
+ "model.layers.17.linear_attn.norm.weight": "model-00012-of-00039.safetensors",
350
+ "model.layers.17.linear_attn.out_proj.weight": "model-00012-of-00039.safetensors",
351
+ "model.layers.18.post_attention_layernorm.weight": "model-00012-of-00039.safetensors",
352
+ "model.layers.18.mlp.gate_proj.weight": "model-00012-of-00039.safetensors",
353
+ "model.layers.18.mlp.up_proj.weight": "model-00012-of-00039.safetensors",
354
+ "model.layers.18.mlp.down_proj.weight": "model-00012-of-00039.safetensors",
355
+ "model.layers.18.input_layernorm.weight": "model-00012-of-00039.safetensors",
356
+ "model.layers.18.linear_attn.A_log": "model-00012-of-00039.safetensors",
357
+ "model.layers.18.linear_attn.conv1d.weight": "model-00012-of-00039.safetensors",
358
+ "model.layers.18.linear_attn.dt_bias": "model-00012-of-00039.safetensors",
359
+ "model.layers.18.linear_attn.in_proj_a.weight": "model-00012-of-00039.safetensors",
360
+ "model.layers.18.linear_attn.in_proj_b.weight": "model-00012-of-00039.safetensors",
361
+ "model.layers.20.mlp.down_proj.weight": "model-00016-of-00039.safetensors",
362
+ "model.layers.20.input_layernorm.weight": "model-00016-of-00039.safetensors",
363
+ "model.layers.20.linear_attn.A_log": "model-00016-of-00039.safetensors",
364
+ "model.layers.20.linear_attn.conv1d.weight": "model-00016-of-00039.safetensors",
365
+ "model.layers.20.linear_attn.dt_bias": "model-00016-of-00039.safetensors",
366
+ "model.layers.20.linear_attn.in_proj_a.weight": "model-00016-of-00039.safetensors",
367
+ "model.layers.20.linear_attn.in_proj_b.weight": "model-00016-of-00039.safetensors",
368
+ "model.layers.20.linear_attn.in_proj_qkv.weight": "model-00016-of-00039.safetensors",
369
+ "model.layers.20.linear_attn.in_proj_z.weight": "model-00016-of-00039.safetensors",
370
+ "model.layers.20.linear_attn.norm.weight": "model-00016-of-00039.safetensors",
371
+ "model.layers.20.linear_attn.out_proj.weight": "model-00016-of-00039.safetensors",
372
+ "model.layers.21.post_attention_layernorm.weight": "model-00016-of-00039.safetensors",
373
+ "model.layers.23.self_attn.q_proj.weight": "model-00020-of-00039.safetensors",
374
+ "model.layers.23.self_attn.k_proj.weight": "model-00020-of-00039.safetensors",
375
+ "model.layers.23.self_attn.v_proj.weight": "model-00020-of-00039.safetensors",
376
+ "model.layers.23.self_attn.q_norm.weight": "model-00020-of-00039.safetensors",
377
+ "model.layers.24.post_attention_layernorm.weight": "model-00020-of-00039.safetensors",
378
+ "model.layers.24.mlp.gate_proj.weight": "model-00020-of-00039.safetensors",
379
+ "model.layers.24.mlp.up_proj.weight": "model-00020-of-00039.safetensors",
380
+ "model.layers.24.mlp.down_proj.weight": "model-00020-of-00039.safetensors",
381
+ "model.layers.24.input_layernorm.weight": "model-00020-of-00039.safetensors",
382
+ "model.layers.24.linear_attn.A_log": "model-00020-of-00039.safetensors",
383
+ "model.layers.24.linear_attn.conv1d.weight": "model-00020-of-00039.safetensors",
384
+ "model.layers.24.linear_attn.dt_bias": "model-00020-of-00039.safetensors",
385
+ "model.layers.24.linear_attn.in_proj_a.weight": "model-00020-of-00039.safetensors",
386
+ "model.layers.24.linear_attn.in_proj_b.weight": "model-00020-of-00039.safetensors",
387
+ "model.layers.24.linear_attn.in_proj_qkv.weight": "model-00020-of-00039.safetensors",
388
+ "model.layers.27.mlp.gate_proj.weight": "model-00024-of-00039.safetensors",
389
+ "model.layers.27.mlp.up_proj.weight": "model-00024-of-00039.safetensors",
390
+ "model.layers.27.mlp.down_proj.weight": "model-00024-of-00039.safetensors",
391
+ "model.layers.27.self_attn.k_norm.weight": "model-00024-of-00039.safetensors",
392
+ "model.layers.27.self_attn.o_proj.weight": "model-00024-of-00039.safetensors",
393
+ "model.layers.27.input_layernorm.weight": "model-00024-of-00039.safetensors",
394
+ "model.layers.27.self_attn.q_proj.weight": "model-00024-of-00039.safetensors",
395
+ "model.layers.27.self_attn.k_proj.weight": "model-00024-of-00039.safetensors",
396
+ "model.layers.27.self_attn.v_proj.weight": "model-00024-of-00039.safetensors",
397
+ "model.layers.27.self_attn.q_norm.weight": "model-00024-of-00039.safetensors",
398
+ "model.layers.28.post_attention_layernorm.weight": "model-00024-of-00039.safetensors",
399
+ "model.layers.3.self_attn.q_proj.weight": "model-00028-of-00039.safetensors",
400
+ "model.layers.3.self_attn.k_proj.weight": "model-00028-of-00039.safetensors",
401
+ "model.layers.3.self_attn.v_proj.weight": "model-00028-of-00039.safetensors",
402
+ "model.layers.3.self_attn.q_norm.weight": "model-00028-of-00039.safetensors",
403
+ "model.layers.30.post_attention_layernorm.weight": "model-00028-of-00039.safetensors",
404
+ "model.layers.30.mlp.gate_proj.weight": "model-00028-of-00039.safetensors",
405
+ "model.layers.30.mlp.up_proj.weight": "model-00028-of-00039.safetensors",
406
+ "model.layers.30.mlp.down_proj.weight": "model-00028-of-00039.safetensors",
407
+ "model.layers.30.input_layernorm.weight": "model-00028-of-00039.safetensors",
408
+ "model.layers.30.linear_attn.A_log": "model-00028-of-00039.safetensors",
409
+ "model.layers.30.linear_attn.conv1d.weight": "model-00028-of-00039.safetensors",
410
+ "model.layers.30.linear_attn.dt_bias": "model-00028-of-00039.safetensors",
411
+ "model.layers.30.linear_attn.in_proj_a.weight": "model-00028-of-00039.safetensors",
412
+ "model.layers.30.linear_attn.in_proj_b.weight": "model-00028-of-00039.safetensors",
413
+ "model.layers.30.linear_attn.in_proj_qkv.weight": "model-00028-of-00039.safetensors",
414
+ "model.layers.5.linear_attn.in_proj_qkv.weight": "model-00032-of-00039.safetensors",
415
+ "model.layers.5.linear_attn.in_proj_z.weight": "model-00032-of-00039.safetensors",
416
+ "model.layers.5.linear_attn.norm.weight": "model-00032-of-00039.safetensors",
417
+ "model.layers.5.linear_attn.out_proj.weight": "model-00032-of-00039.safetensors",
418
+ "model.layers.6.post_attention_layernorm.weight": "model-00032-of-00039.safetensors",
419
+ "model.layers.6.mlp.gate_proj.weight": "model-00032-of-00039.safetensors",
420
+ "model.layers.6.mlp.up_proj.weight": "model-00032-of-00039.safetensors",
421
+ "model.layers.8.linear_attn.out_proj.weight": "model-00036-of-00039.safetensors",
422
+ "model.layers.9.post_attention_layernorm.weight": "model-00036-of-00039.safetensors",
423
+ "model.layers.9.mlp.gate_proj.weight": "model-00036-of-00039.safetensors",
424
+ "model.layers.9.mlp.up_proj.weight": "model-00036-of-00039.safetensors",
425
+ "model.layers.9.mlp.down_proj.weight": "model-00036-of-00039.safetensors",
426
+ "model.layers.9.input_layernorm.weight": "model-00036-of-00039.safetensors",
427
+ "model.layers.9.linear_attn.A_log": "model-00036-of-00039.safetensors",
428
+ "model.layers.9.linear_attn.conv1d.weight": "model-00036-of-00039.safetensors",
429
+ "model.layers.9.linear_attn.dt_bias": "model-00036-of-00039.safetensors",
430
+ "model.layers.9.linear_attn.in_proj_a.weight": "model-00036-of-00039.safetensors",
431
+ "model.layers.9.linear_attn.in_proj_b.weight": "model-00036-of-00039.safetensors",
432
+ "model.layers.9.linear_attn.in_proj_qkv.weight": "model-00036-of-00039.safetensors"
433
+ }
434
+ }
rl-agentic-new/step-30/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
3
+ size 19989325
rl-agentic-new/step-30/tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "audio_bos_token": "<|audio_start|>",
4
+ "audio_eos_token": "<|audio_end|>",
5
+ "audio_token": "<|audio_pad|>",
6
+ "backend": "tokenizers",
7
+ "bos_token": null,
8
+ "clean_up_tokenization_spaces": false,
9
+ "eos_token": "<|im_end|>",
10
+ "errors": "replace",
11
+ "image_token": "<|image_pad|>",
12
+ "is_local": true,
13
+ "local_files_only": true,
14
+ "model_max_length": 262144,
15
+ "model_specific_special_tokens": {
16
+ "audio_bos_token": "<|audio_start|>",
17
+ "audio_eos_token": "<|audio_end|>",
18
+ "audio_token": "<|audio_pad|>",
19
+ "image_token": "<|image_pad|>",
20
+ "video_token": "<|video_pad|>",
21
+ "vision_bos_token": "<|vision_start|>",
22
+ "vision_eos_token": "<|vision_end|>"
23
+ },
24
+ "pad_token": "<|endoftext|>",
25
+ "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
26
+ "split_special_tokens": false,
27
+ "tokenizer_class": "Qwen2Tokenizer",
28
+ "unk_token": null,
29
+ "video_token": "<|video_pad|>",
30
+ "vision_bos_token": "<|vision_start|>",
31
+ "vision_eos_token": "<|vision_end|>"
32
+ }