Hugging Face's logo

LJYAI
/
temp_ss

Safetensors
Model card Files
xet
 Community
LJYAI commited on
Commit
1ee0cb9
·
verified ·
1 Parent(s): fe5bfb4

upload

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +3 -0
  2. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/added_tokens.json +28 -0
  3. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/chat_template.jinja +85 -0
  4. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/config.json +50 -0
  5. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/generation_config.json +6 -0
  6. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_10/checkpoint.pt +3 -0
  7. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_12/checkpoint.pt +3 -0
  8. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_14/checkpoint.pt +3 -0
  9. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_16/checkpoint.pt +3 -0
  10. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_18/checkpoint.pt +3 -0
  11. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_2/checkpoint.pt +3 -0
  12. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_20/checkpoint.pt +3 -0
  13. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_4/checkpoint.pt +3 -0
  14. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_6/checkpoint.pt +3 -0
  15. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_8/checkpoint.pt +3 -0
  16. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_final/checkpoint.pt +3 -0
  17. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/merges.txt +0 -0
  18. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/model.safetensors +3 -0
  19. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/run_args.txt +9 -0
  20. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/special_tokens_map.json +25 -0
  21. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/tokenizer.json +3 -0
  22. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/tokenizer_config.json +239 -0
  23. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/train_loss.txt +188 -0
  24. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/vocab.json +0 -0
  25. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/added_tokens.json +28 -0
  26. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/chat_template.jinja +85 -0
  27. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/config.json +56 -0
  28. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/generation_config.json +6 -0
  29. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_10/checkpoint.pt +3 -0
  30. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_12/checkpoint.pt +3 -0
  31. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_14/checkpoint.pt +3 -0
  32. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_16/checkpoint.pt +3 -0
  33. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_18/checkpoint.pt +3 -0
  34. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_2/checkpoint.pt +3 -0
  35. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_20/checkpoint.pt +3 -0
  36. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_4/checkpoint.pt +3 -0
  37. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_6/checkpoint.pt +3 -0
  38. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_8/checkpoint.pt +3 -0
  39. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_final/checkpoint.pt +3 -0
  40. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/merges.txt +0 -0
  41. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/model.safetensors +3 -0
  42. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/run_args.txt +9 -0
  43. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/special_tokens_map.json +25 -0
  44. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/tokenizer.json +3 -0
  45. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/tokenizer_config.json +239 -0
  46. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/train_loss.txt +188 -0
  47. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/vocab.json +0 -0
  48. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune7_start10/added_tokens.json +28 -0
  49. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune7_start10/chat_template.jinja +85 -0
  50. llmstream_results/qwen3_17b_base_streamline_slimpajama_prune7_start10/config.json +53 -0
.gitattributes CHANGED
@@ -46,3 +46,6 @@ llmstream_results/qwen_8b_inst_streamline_slimorca_prune9_start24/tokenizer.json
46
  llmstream_results/llama31_8b_base_streamline_slimpajama_prune12_start18/tokenizer.json filter=lfs diff=lfs merge=lfs -text
47
  llmstream_results/llama31_8b_base_streamline_slimpajama_prune4_start23/tokenizer.json filter=lfs diff=lfs merge=lfs -text
48
  llmstream_results/llama31_8b_base_streamline_slimpajama_prune8_start22/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
 
 
46
  llmstream_results/llama31_8b_base_streamline_slimpajama_prune12_start18/tokenizer.json filter=lfs diff=lfs merge=lfs -text
47
  llmstream_results/llama31_8b_base_streamline_slimpajama_prune4_start23/tokenizer.json filter=lfs diff=lfs merge=lfs -text
48
  llmstream_results/llama31_8b_base_streamline_slimpajama_prune8_start22/tokenizer.json filter=lfs diff=lfs merge=lfs -text
49
+ llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/tokenizer.json filter=lfs diff=lfs merge=lfs -text
50
+ llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/tokenizer.json filter=lfs diff=lfs merge=lfs -text
51
+ llmstream_results/qwen3_17b_base_streamline_slimpajama_prune7_start10/tokenizer.json filter=lfs diff=lfs merge=lfs -text
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/chat_template.jinja ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
27
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
28
+ {%- elif message.role == "assistant" %}
29
+ {%- set content = message.content %}
30
+ {%- set reasoning_content = '' %}
31
+ {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
32
+ {%- set reasoning_content = message.reasoning_content %}
33
+ {%- else %}
34
+ {%- if '</think>' in message.content %}
35
+ {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
36
+ {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
37
+ {%- endif %}
38
+ {%- endif %}
39
+ {%- if loop.index0 > ns.last_query_index %}
40
+ {%- if loop.last or (not loop.last and reasoning_content) %}
41
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
42
+ {%- else %}
43
+ {{- '<|im_start|>' + message.role + '\n' + content }}
44
+ {%- endif %}
45
+ {%- else %}
46
+ {{- '<|im_start|>' + message.role + '\n' + content }}
47
+ {%- endif %}
48
+ {%- if message.tool_calls %}
49
+ {%- for tool_call in message.tool_calls %}
50
+ {%- if (loop.first and content) or (not loop.first) %}
51
+ {{- '\n' }}
52
+ {%- endif %}
53
+ {%- if tool_call.function %}
54
+ {%- set tool_call = tool_call.function %}
55
+ {%- endif %}
56
+ {{- '<tool_call>\n{"name": "' }}
57
+ {{- tool_call.name }}
58
+ {{- '", "arguments": ' }}
59
+ {%- if tool_call.arguments is string %}
60
+ {{- tool_call.arguments }}
61
+ {%- else %}
62
+ {{- tool_call.arguments | tojson }}
63
+ {%- endif %}
64
+ {{- '}\n</tool_call>' }}
65
+ {%- endfor %}
66
+ {%- endif %}
67
+ {{- '<|im_end|>\n' }}
68
+ {%- elif message.role == "tool" %}
69
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
70
+ {{- '<|im_start|>user' }}
71
+ {%- endif %}
72
+ {{- '\n<tool_response>\n' }}
73
+ {{- message.content }}
74
+ {{- '\n</tool_response>' }}
75
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
76
+ {{- '<|im_end|>\n' }}
77
+ {%- endif %}
78
+ {%- endif %}
79
+ {%- endfor %}
80
+ {%- if add_generation_prompt %}
81
+ {{- '<|im_start|>assistant\n' }}
82
+ {%- if enable_thinking is defined and enable_thinking is false %}
83
+ {{- '<think>\n\n</think>\n\n' }}
84
+ {%- endif %}
85
+ {%- endif %}
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151643,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 6144,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention"
34
+ ],
35
+ "max_position_embeddings": 32768,
36
+ "max_window_layers": 28,
37
+ "model_type": "qwen3",
38
+ "num_attention_heads": 16,
39
+ "num_hidden_layers": 18,
40
+ "num_key_value_heads": 8,
41
+ "rms_norm_eps": 1e-06,
42
+ "rope_scaling": null,
43
+ "rope_theta": 1000000,
44
+ "sliding_window": null,
45
+ "tie_word_embeddings": true,
46
+ "transformers_version": "4.57.6",
47
+ "use_cache": true,
48
+ "use_sliding_window": false,
49
+ "vocab_size": 151936
50
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 151643,
4
+ "eos_token_id": 151643,
5
+ "transformers_version": "4.57.6"
6
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_10/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66e921f410447dbbfb019eabe22dafb839d7ae508652b8f0bdbfb2a3ea8d95c5
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_12/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0b80872cb5adcc6b1f9cf3238e1105b1778ec69dbb061267c12ca3754c0f1b9
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_14/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ab55e20f2a17d7b1d4142238eb37da9c00b9580055efed5afc1d36607197a10
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_16/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98e4231f774da050f0bfdf9d2a1628a347061ad1eab3070e3e45563cf8952978
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_18/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94688fa78c7f601ed8daebf10701ede958b43c31084c447e0a93d77608e988a0
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_2/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:424b458f92a47d719aae52de4979cefcd7ff9c011c51e8fd3123e398190db086
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_20/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81be076aa61d2331c80298166a5218fe1fbbdee4f7446bdd103012a43cf6e48a
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_4/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac2250cb702711e78516b81f74297d79e5de42dcd63be117193ea380a49ff7b3
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_6/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dd1fc239c001bf653d0f7d3353c4e0ae415788cc9b7031b59e25c34b61edb24
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_epoch_8/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:feaf2f8b15edf095c3705a375bc573f2d5be8e0d9d2aa48d09d5def7efd9b046
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/lightweight_checkpoint_final/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81be076aa61d2331c80298166a5218fe1fbbdee4f7446bdd103012a43cf6e48a
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2678fdeea3aa00791a9ae37144f6157f737288476878d28c3aa05e8e834f5f9e
3
+ size 2434452888
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/run_args.txt ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ git_commit=unknown
2
+ dataset=slimpajama
3
+ dataset_split=train
4
+ collect_batch_size=8
5
+ train_batch_size=32
6
+ gradient_accumulation_step=8
7
+ target_effective_batch=256
8
+ command:
9
+ python /workspace/here/abprune_update/compare_model/LLM-Streamline/qwen_mseloss_entry.py --model_name Qwen/Qwen3-1.7B-Base --output_dir /workspace/here/abprune_update/results/qwen3_17b_base_streamline_slimpajama_prune10_start7 --dataset slimpajama --dataset_split train --layer_intervals 10 --best_layer 7 --cosine_num_data 300 --train_num_data 30000 --epoches 20 --batch_size 8 --collect_batch_size 8 --train_batch_size 32 --gradient_accumulation_step 8 --lr 1e-5 --min_lr 5e-5 --wd 1e-3
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/special_tokens_map.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|endoftext|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": "<|endoftext|>"
25
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/tokenizer_config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|endoftext|>",
232
+ "errors": "replace",
233
+ "extra_special_tokens": {},
234
+ "model_max_length": 131072,
235
+ "pad_token": "<|endoftext|>",
236
+ "split_special_tokens": false,
237
+ "tokenizer_class": "Qwen2Tokenizer",
238
+ "unk_token": null
239
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/train_loss.txt ADDED
@@ -0,0 +1,188 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Train loss logged every 100 global steps
2
+ epoch=0 global_step=100 optimizer_step=12 loss=61.8111763000
3
+ epoch=0 global_step=200 optimizer_step=24 loss=61.3172912598
4
+ epoch=0 global_step=300 optimizer_step=37 loss=61.2910614014
5
+ epoch=0 global_step=400 optimizer_step=49 loss=58.9159240723
6
+ epoch=0 global_step=500 optimizer_step=62 loss=61.0487556458
7
+ epoch=0 global_step=600 optimizer_step=74 loss=59.8903465271
8
+ epoch=0 global_step=700 optimizer_step=87 loss=58.9380989075
9
+ epoch=0 global_step=800 optimizer_step=99 loss=56.7384490967
10
+ epoch=0 global_step=900 optimizer_step=112 loss=53.5263824463
11
+ epoch=1 global_step=1000 optimizer_step=124 loss=53.7964935303
12
+ epoch=1 global_step=1100 optimizer_step=137 loss=51.3535690308
13
+ epoch=1 global_step=1200 optimizer_step=149 loss=46.7600708008
14
+ epoch=1 global_step=1300 optimizer_step=162 loss=45.7732582092
15
+ epoch=1 global_step=1400 optimizer_step=174 loss=43.7938156128
16
+ epoch=1 global_step=1500 optimizer_step=187 loss=43.0620880127
17
+ epoch=1 global_step=1600 optimizer_step=199 loss=41.8608360291
18
+ epoch=1 global_step=1700 optimizer_step=212 loss=41.6349334717
19
+ epoch=1 global_step=1800 optimizer_step=224 loss=40.2696914673
20
+ epoch=2 global_step=1900 optimizer_step=236 loss=40.2012939453
21
+ epoch=2 global_step=2000 optimizer_step=249 loss=39.6849060059
22
+ epoch=2 global_step=2100 optimizer_step=261 loss=39.1601028442
23
+ epoch=2 global_step=2200 optimizer_step=274 loss=38.5205917358
24
+ epoch=2 global_step=2300 optimizer_step=286 loss=38.3351364136
25
+ epoch=2 global_step=2400 optimizer_step=299 loss=38.3407173157
26
+ epoch=2 global_step=2500 optimizer_step=311 loss=38.2053909302
27
+ epoch=2 global_step=2600 optimizer_step=324 loss=37.6079406738
28
+ epoch=2 global_step=2700 optimizer_step=336 loss=36.8246154785
29
+ epoch=2 global_step=2800 optimizer_step=349 loss=36.9841232300
30
+ epoch=3 global_step=2900 optimizer_step=361 loss=37.0083427429
31
+ epoch=3 global_step=3000 optimizer_step=374 loss=36.0554275513
32
+ epoch=3 global_step=3100 optimizer_step=386 loss=35.7142372131
33
+ epoch=3 global_step=3200 optimizer_step=399 loss=36.5023193359
34
+ epoch=3 global_step=3300 optimizer_step=411 loss=35.1390457153
35
+ epoch=3 global_step=3400 optimizer_step=424 loss=34.4201660156
36
+ epoch=3 global_step=3500 optimizer_step=436 loss=34.6847610474
37
+ epoch=3 global_step=3600 optimizer_step=449 loss=34.8008117676
38
+ epoch=3 global_step=3700 optimizer_step=461 loss=34.2760162354
39
+ epoch=4 global_step=3800 optimizer_step=473 loss=33.7564620972
40
+ epoch=4 global_step=3900 optimizer_step=486 loss=33.3902893066
41
+ epoch=4 global_step=4000 optimizer_step=498 loss=33.0707626343
42
+ epoch=4 global_step=4100 optimizer_step=511 loss=33.4961090088
43
+ epoch=4 global_step=4200 optimizer_step=523 loss=33.2552566528
44
+ epoch=4 global_step=4300 optimizer_step=536 loss=32.8767166138
45
+ epoch=4 global_step=4400 optimizer_step=548 loss=32.2891387939
46
+ epoch=4 global_step=4500 optimizer_step=561 loss=32.2723655701
47
+ epoch=4 global_step=4600 optimizer_step=573 loss=32.5729331970
48
+ epoch=5 global_step=4700 optimizer_step=586 loss=31.7442245483
49
+ epoch=5 global_step=4800 optimizer_step=598 loss=32.0122909546
50
+ epoch=5 global_step=4900 optimizer_step=611 loss=32.2825584412
51
+ epoch=5 global_step=5000 optimizer_step=623 loss=31.3409729004
52
+ epoch=5 global_step=5100 optimizer_step=636 loss=31.5066032410
53
+ epoch=5 global_step=5200 optimizer_step=648 loss=31.0779724121
54
+ epoch=5 global_step=5300 optimizer_step=661 loss=32.0436706543
55
+ epoch=5 global_step=5400 optimizer_step=673 loss=30.9592647552
56
+ epoch=5 global_step=5500 optimizer_step=686 loss=30.5881576538
57
+ epoch=5 global_step=5600 optimizer_step=698 loss=31.1291732788
58
+ epoch=6 global_step=5700 optimizer_step=710 loss=30.6245613098
59
+ epoch=6 global_step=5800 optimizer_step=723 loss=30.3179073334
60
+ epoch=6 global_step=5900 optimizer_step=735 loss=29.5653247833
61
+ epoch=6 global_step=6000 optimizer_step=748 loss=30.0554027557
62
+ epoch=6 global_step=6100 optimizer_step=760 loss=29.8230056763
63
+ epoch=6 global_step=6200 optimizer_step=773 loss=30.0757102966
64
+ epoch=6 global_step=6300 optimizer_step=785 loss=29.5166015625
65
+ epoch=6 global_step=6400 optimizer_step=798 loss=29.3904190063
66
+ epoch=6 global_step=6500 optimizer_step=810 loss=29.3530120850
67
+ epoch=7 global_step=6600 optimizer_step=823 loss=29.3214378357
68
+ epoch=7 global_step=6700 optimizer_step=835 loss=29.2177028656
69
+ epoch=7 global_step=6800 optimizer_step=848 loss=28.5677661896
70
+ epoch=7 global_step=6900 optimizer_step=860 loss=29.2729091644
71
+ epoch=7 global_step=7000 optimizer_step=873 loss=28.1423110962
72
+ epoch=7 global_step=7100 optimizer_step=885 loss=28.3978919983
73
+ epoch=7 global_step=7200 optimizer_step=898 loss=28.7332496643
74
+ epoch=7 global_step=7300 optimizer_step=910 loss=28.7455711365
75
+ epoch=7 global_step=7400 optimizer_step=923 loss=28.1405372620
76
+ epoch=7 global_step=7500 optimizer_step=935 loss=28.4964752197
77
+ epoch=8 global_step=7600 optimizer_step=947 loss=27.6514358521
78
+ epoch=8 global_step=7700 optimizer_step=960 loss=27.5642356873
79
+ epoch=8 global_step=7800 optimizer_step=972 loss=28.3289890289
80
+ epoch=8 global_step=7900 optimizer_step=985 loss=27.6473426819
81
+ epoch=8 global_step=8000 optimizer_step=997 loss=27.8036746979
82
+ epoch=8 global_step=8100 optimizer_step=1010 loss=27.4297943115
83
+ epoch=8 global_step=8200 optimizer_step=1022 loss=27.5465431213
84
+ epoch=8 global_step=8300 optimizer_step=1035 loss=27.6759014130
85
+ epoch=8 global_step=8400 optimizer_step=1047 loss=27.1972808838
86
+ epoch=9 global_step=8500 optimizer_step=1060 loss=27.8164634705
87
+ epoch=9 global_step=8600 optimizer_step=1072 loss=27.0610504150
88
+ epoch=9 global_step=8700 optimizer_step=1085 loss=26.8193874359
89
+ epoch=9 global_step=8800 optimizer_step=1097 loss=26.5042495728
90
+ epoch=9 global_step=8900 optimizer_step=1110 loss=27.5187721252
91
+ epoch=9 global_step=9000 optimizer_step=1122 loss=26.7285690308
92
+ epoch=9 global_step=9100 optimizer_step=1135 loss=26.7155189514
93
+ epoch=9 global_step=9200 optimizer_step=1147 loss=26.5143203735
94
+ epoch=9 global_step=9300 optimizer_step=1160 loss=26.2485313416
95
+ epoch=10 global_step=9400 optimizer_step=1172 loss=25.9391574860
96
+ epoch=10 global_step=9500 optimizer_step=1184 loss=26.5734863281
97
+ epoch=10 global_step=9600 optimizer_step=1197 loss=26.5991668701
98
+ epoch=10 global_step=9700 optimizer_step=1209 loss=26.2932891846
99
+ epoch=10 global_step=9800 optimizer_step=1222 loss=26.0009422302
100
+ epoch=10 global_step=9900 optimizer_step=1234 loss=25.9818954468
101
+ epoch=10 global_step=10000 optimizer_step=1247 loss=26.2013587952
102
+ epoch=10 global_step=10100 optimizer_step=1259 loss=26.3891105652
103
+ epoch=10 global_step=10200 optimizer_step=1272 loss=25.9828681946
104
+ epoch=10 global_step=10300 optimizer_step=1284 loss=25.5200080872
105
+ epoch=11 global_step=10400 optimizer_step=1297 loss=25.1604328156
106
+ epoch=11 global_step=10500 optimizer_step=1309 loss=25.8807182312
107
+ epoch=11 global_step=10600 optimizer_step=1322 loss=25.4990501404
108
+ epoch=11 global_step=10700 optimizer_step=1334 loss=25.5015258789
109
+ epoch=11 global_step=10800 optimizer_step=1347 loss=25.3566055298
110
+ epoch=11 global_step=10900 optimizer_step=1359 loss=25.3202247620
111
+ epoch=11 global_step=11000 optimizer_step=1372 loss=25.1426544189
112
+ epoch=11 global_step=11100 optimizer_step=1384 loss=25.4124908447
113
+ epoch=11 global_step=11200 optimizer_step=1397 loss=24.7932891846
114
+ epoch=12 global_step=11300 optimizer_step=1409 loss=24.8534622192
115
+ epoch=12 global_step=11400 optimizer_step=1421 loss=24.9645309448
116
+ epoch=12 global_step=11500 optimizer_step=1434 loss=25.0001716614
117
+ epoch=12 global_step=11600 optimizer_step=1446 loss=24.8191261292
118
+ epoch=12 global_step=11700 optimizer_step=1459 loss=24.8512458801
119
+ epoch=12 global_step=11800 optimizer_step=1471 loss=24.5848960876
120
+ epoch=12 global_step=11900 optimizer_step=1484 loss=24.7647628784
121
+ epoch=12 global_step=12000 optimizer_step=1496 loss=24.5570564270
122
+ epoch=12 global_step=12100 optimizer_step=1509 loss=24.2256755829
123
+ epoch=13 global_step=12200 optimizer_step=1521 loss=24.6256027222
124
+ epoch=13 global_step=12300 optimizer_step=1534 loss=24.0865955353
125
+ epoch=13 global_step=12400 optimizer_step=1546 loss=23.7448959351
126
+ epoch=13 global_step=12500 optimizer_step=1559 loss=24.0185546875
127
+ epoch=13 global_step=12600 optimizer_step=1571 loss=24.2570247650
128
+ epoch=13 global_step=12700 optimizer_step=1584 loss=24.1442604065
129
+ epoch=13 global_step=12800 optimizer_step=1596 loss=24.0124988556
130
+ epoch=13 global_step=12900 optimizer_step=1609 loss=24.1430892944
131
+ epoch=13 global_step=13000 optimizer_step=1621 loss=23.6023216248
132
+ epoch=13 global_step=13100 optimizer_step=1634 loss=24.0389404297
133
+ epoch=14 global_step=13200 optimizer_step=1646 loss=24.3021049500
134
+ epoch=14 global_step=13300 optimizer_step=1658 loss=23.8130569458
135
+ epoch=14 global_step=13400 optimizer_step=1671 loss=24.1595306396
136
+ epoch=14 global_step=13500 optimizer_step=1683 loss=24.2647094727
137
+ epoch=14 global_step=13600 optimizer_step=1696 loss=23.3427543640
138
+ epoch=14 global_step=13700 optimizer_step=1708 loss=23.6317939758
139
+ epoch=14 global_step=13800 optimizer_step=1721 loss=23.4779624939
140
+ epoch=14 global_step=13900 optimizer_step=1733 loss=23.7555770874
141
+ epoch=14 global_step=14000 optimizer_step=1746 loss=23.7388343811
142
+ epoch=15 global_step=14100 optimizer_step=1758 loss=23.5254554749
143
+ epoch=15 global_step=14200 optimizer_step=1771 loss=23.2854118347
144
+ epoch=15 global_step=14300 optimizer_step=1783 loss=23.1354923248
145
+ epoch=15 global_step=14400 optimizer_step=1796 loss=22.0143985748
146
+ epoch=15 global_step=14500 optimizer_step=1808 loss=22.9370117188
147
+ epoch=15 global_step=14600 optimizer_step=1821 loss=23.1857337952
148
+ epoch=15 global_step=14700 optimizer_step=1833 loss=22.8030986786
149
+ epoch=15 global_step=14800 optimizer_step=1846 loss=23.1866607666
150
+ epoch=15 global_step=14900 optimizer_step=1858 loss=22.7722396851
151
+ epoch=15 global_step=15000 optimizer_step=1871 loss=23.4702568054
152
+ epoch=16 global_step=15100 optimizer_step=1883 loss=22.7320709229
153
+ epoch=16 global_step=15200 optimizer_step=1895 loss=22.8686485291
154
+ epoch=16 global_step=15300 optimizer_step=1908 loss=22.4258575439
155
+ epoch=16 global_step=15400 optimizer_step=1920 loss=23.1223926544
156
+ epoch=16 global_step=15500 optimizer_step=1933 loss=22.4683704376
157
+ epoch=16 global_step=15600 optimizer_step=1945 loss=22.5086746216
158
+ epoch=16 global_step=15700 optimizer_step=1958 loss=22.8600883484
159
+ epoch=16 global_step=15800 optimizer_step=1970 loss=23.3026580811
160
+ epoch=16 global_step=15900 optimizer_step=1983 loss=22.2357749939
161
+ epoch=17 global_step=16000 optimizer_step=1995 loss=22.4431304932
162
+ epoch=17 global_step=16100 optimizer_step=2008 loss=22.2072048187
163
+ epoch=17 global_step=16200 optimizer_step=2020 loss=22.1248683929
164
+ epoch=17 global_step=16300 optimizer_step=2033 loss=22.3402061462
165
+ epoch=17 global_step=16400 optimizer_step=2045 loss=22.1372089386
166
+ epoch=17 global_step=16500 optimizer_step=2058 loss=22.4117164612
167
+ epoch=17 global_step=16600 optimizer_step=2070 loss=21.7634601593
168
+ epoch=17 global_step=16700 optimizer_step=2083 loss=22.0447540283
169
+ epoch=17 global_step=16800 optimizer_step=2095 loss=21.7455501556
170
+ epoch=18 global_step=16900 optimizer_step=2107 loss=21.7101211548
171
+ epoch=18 global_step=17000 optimizer_step=2120 loss=22.2101898193
172
+ epoch=18 global_step=17100 optimizer_step=2132 loss=21.8840522766
173
+ epoch=18 global_step=17200 optimizer_step=2145 loss=21.9883365631
174
+ epoch=18 global_step=17300 optimizer_step=2157 loss=21.1718349457
175
+ epoch=18 global_step=17400 optimizer_step=2170 loss=21.4985771179
176
+ epoch=18 global_step=17500 optimizer_step=2182 loss=21.5589141846
177
+ epoch=18 global_step=17600 optimizer_step=2195 loss=21.3192443848
178
+ epoch=18 global_step=17700 optimizer_step=2207 loss=21.3583335876
179
+ epoch=18 global_step=17800 optimizer_step=2220 loss=21.3754730225
180
+ epoch=19 global_step=17900 optimizer_step=2232 loss=21.7851066589
181
+ epoch=19 global_step=18000 optimizer_step=2245 loss=21.2104225159
182
+ epoch=19 global_step=18100 optimizer_step=2257 loss=21.5523376465
183
+ epoch=19 global_step=18200 optimizer_step=2270 loss=20.8208522797
184
+ epoch=19 global_step=18300 optimizer_step=2282 loss=21.0332641602
185
+ epoch=19 global_step=18400 optimizer_step=2295 loss=21.8806171417
186
+ epoch=19 global_step=18500 optimizer_step=2307 loss=20.4305744171
187
+ epoch=19 global_step=18600 optimizer_step=2320 loss=20.8918380737
188
+ epoch=19 global_step=18700 optimizer_step=2332 loss=21.4735279083
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune10_start7/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/chat_template.jinja ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
27
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
28
+ {%- elif message.role == "assistant" %}
29
+ {%- set content = message.content %}
30
+ {%- set reasoning_content = '' %}
31
+ {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
32
+ {%- set reasoning_content = message.reasoning_content %}
33
+ {%- else %}
34
+ {%- if '</think>' in message.content %}
35
+ {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
36
+ {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
37
+ {%- endif %}
38
+ {%- endif %}
39
+ {%- if loop.index0 > ns.last_query_index %}
40
+ {%- if loop.last or (not loop.last and reasoning_content) %}
41
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
42
+ {%- else %}
43
+ {{- '<|im_start|>' + message.role + '\n' + content }}
44
+ {%- endif %}
45
+ {%- else %}
46
+ {{- '<|im_start|>' + message.role + '\n' + content }}
47
+ {%- endif %}
48
+ {%- if message.tool_calls %}
49
+ {%- for tool_call in message.tool_calls %}
50
+ {%- if (loop.first and content) or (not loop.first) %}
51
+ {{- '\n' }}
52
+ {%- endif %}
53
+ {%- if tool_call.function %}
54
+ {%- set tool_call = tool_call.function %}
55
+ {%- endif %}
56
+ {{- '<tool_call>\n{"name": "' }}
57
+ {{- tool_call.name }}
58
+ {{- '", "arguments": ' }}
59
+ {%- if tool_call.arguments is string %}
60
+ {{- tool_call.arguments }}
61
+ {%- else %}
62
+ {{- tool_call.arguments | tojson }}
63
+ {%- endif %}
64
+ {{- '}\n</tool_call>' }}
65
+ {%- endfor %}
66
+ {%- endif %}
67
+ {{- '<|im_end|>\n' }}
68
+ {%- elif message.role == "tool" %}
69
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
70
+ {{- '<|im_start|>user' }}
71
+ {%- endif %}
72
+ {{- '\n<tool_response>\n' }}
73
+ {{- message.content }}
74
+ {{- '\n</tool_response>' }}
75
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
76
+ {{- '<|im_end|>\n' }}
77
+ {%- endif %}
78
+ {%- endif %}
79
+ {%- endfor %}
80
+ {%- if add_generation_prompt %}
81
+ {{- '<|im_start|>assistant\n' }}
82
+ {%- if enable_thinking is defined and enable_thinking is false %}
83
+ {{- '<think>\n\n</think>\n\n' }}
84
+ {%- endif %}
85
+ {%- endif %}
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151643,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 6144,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention"
40
+ ],
41
+ "max_position_embeddings": 32768,
42
+ "max_window_layers": 28,
43
+ "model_type": "qwen3",
44
+ "num_attention_heads": 16,
45
+ "num_hidden_layers": 24,
46
+ "num_key_value_heads": 8,
47
+ "rms_norm_eps": 1e-06,
48
+ "rope_scaling": null,
49
+ "rope_theta": 1000000,
50
+ "sliding_window": null,
51
+ "tie_word_embeddings": true,
52
+ "transformers_version": "4.57.6",
53
+ "use_cache": true,
54
+ "use_sliding_window": false,
55
+ "vocab_size": 151936
56
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 151643,
4
+ "eos_token_id": 151643,
5
+ "transformers_version": "4.57.6"
6
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_10/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9917f7196dd6a28c59f015a0c240a162f03ef75db1c01c5bd37f4a56c0e1ab7
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_12/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cde02445b06ea1927ffee7e095d2dd72640bb0b9db66696034543af037a14457
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_14/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2246578f572699b4618a0f2e0ee5fedec64ea650aabfc66c0cf27a49a3751df
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_16/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28f56e1c6207b726b8c9ab447e14f4e11de1108ce839606e6fe112e384f6d5b8
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_18/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f576688d219d5b577923932175a3a1a04b88fad6e98b554d8bcde8ae19ca5189
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_2/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:500d4eab3f09185d6483d23aa5c8a45d7b3279469350d7ba65c8228cd27c0694
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_20/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:913b3939db909490764105b73fc5a05f0e4cf755cc287a617b15ed1d1bd8f10b
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_4/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41c99017de3f7f2fdfb4c189eb1a7094c0c1a3d874364cfdce8d5437cc24342b
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_6/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b13b0c7f3e6e82357e378d46f30180f50a4dc15e94b425ff2ba0f832a8382ceb
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_epoch_8/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45a3b82df2707b6aaf819800cc0c6635266122bad753c37bd0bcdb752dda0a88
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/lightweight_checkpoint_final/checkpoint.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:913b3939db909490764105b73fc5a05f0e4cf755cc287a617b15ed1d1bd8f10b
3
+ size 201349429
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa47fdb183341c382b7ba179f3c498e48292aa338017c81f9988672fa44f86cd
3
+ size 3038492520
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/run_args.txt ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ git_commit=unknown
2
+ dataset=slimpajama
3
+ dataset_split=train
4
+ collect_batch_size=8
5
+ train_batch_size=32
6
+ gradient_accumulation_step=8
7
+ target_effective_batch=256
8
+ command:
9
+ python /workspace/here/abprune_update/compare_model/LLM-Streamline/qwen_mseloss_entry.py --model_name Qwen/Qwen3-1.7B-Base --output_dir /workspace/here/abprune_update/results/qwen3_17b_base_streamline_slimpajama_prune4_start11 --dataset slimpajama --dataset_split train --layer_intervals 4 --best_layer 11 --cosine_num_data 300 --train_num_data 30000 --epoches 20 --batch_size 8 --collect_batch_size 8 --train_batch_size 32 --gradient_accumulation_step 8 --lr 1e-5 --min_lr 5e-5 --wd 1e-3
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/special_tokens_map.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|endoftext|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": "<|endoftext|>"
25
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/tokenizer_config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|endoftext|>",
232
+ "errors": "replace",
233
+ "extra_special_tokens": {},
234
+ "model_max_length": 131072,
235
+ "pad_token": "<|endoftext|>",
236
+ "split_special_tokens": false,
237
+ "tokenizer_class": "Qwen2Tokenizer",
238
+ "unk_token": null
239
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/train_loss.txt ADDED
@@ -0,0 +1,188 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Train loss logged every 100 global steps
2
+ epoch=0 global_step=100 optimizer_step=12 loss=7.9170217514
3
+ epoch=0 global_step=200 optimizer_step=24 loss=7.8473553658
4
+ epoch=0 global_step=300 optimizer_step=37 loss=7.7116975784
5
+ epoch=0 global_step=400 optimizer_step=49 loss=7.4089689255
6
+ epoch=0 global_step=500 optimizer_step=62 loss=7.1841478348
7
+ epoch=0 global_step=600 optimizer_step=74 loss=6.8926868439
8
+ epoch=0 global_step=700 optimizer_step=87 loss=6.5810146332
9
+ epoch=0 global_step=800 optimizer_step=99 loss=6.2605690956
10
+ epoch=0 global_step=900 optimizer_step=112 loss=6.2256917953
11
+ epoch=1 global_step=1000 optimizer_step=124 loss=6.0888376236
12
+ epoch=1 global_step=1100 optimizer_step=137 loss=5.8560752869
13
+ epoch=1 global_step=1200 optimizer_step=149 loss=5.8659281731
14
+ epoch=1 global_step=1300 optimizer_step=162 loss=5.7942686081
15
+ epoch=1 global_step=1400 optimizer_step=174 loss=5.9467153549
16
+ epoch=1 global_step=1500 optimizer_step=187 loss=5.5050134659
17
+ epoch=1 global_step=1600 optimizer_step=199 loss=5.4962210655
18
+ epoch=1 global_step=1700 optimizer_step=212 loss=5.3773903847
19
+ epoch=1 global_step=1800 optimizer_step=224 loss=5.3068108559
20
+ epoch=2 global_step=1900 optimizer_step=236 loss=5.3665647507
21
+ epoch=2 global_step=2000 optimizer_step=249 loss=5.3980102539
22
+ epoch=2 global_step=2100 optimizer_step=261 loss=5.2461566925
23
+ epoch=2 global_step=2200 optimizer_step=274 loss=5.2681255341
24
+ epoch=2 global_step=2300 optimizer_step=286 loss=5.2004423141
25
+ epoch=2 global_step=2400 optimizer_step=299 loss=5.1885271072
26
+ epoch=2 global_step=2500 optimizer_step=311 loss=5.0665435791
27
+ epoch=2 global_step=2600 optimizer_step=324 loss=4.9362063408
28
+ epoch=2 global_step=2700 optimizer_step=336 loss=5.0362515450
29
+ epoch=2 global_step=2800 optimizer_step=349 loss=5.0134139061
30
+ epoch=3 global_step=2900 optimizer_step=361 loss=5.1420578957
31
+ epoch=3 global_step=3000 optimizer_step=374 loss=4.8358516693
32
+ epoch=3 global_step=3100 optimizer_step=386 loss=4.8853735924
33
+ epoch=3 global_step=3200 optimizer_step=399 loss=4.7420692444
34
+ epoch=3 global_step=3300 optimizer_step=411 loss=4.8255195618
35
+ epoch=3 global_step=3400 optimizer_step=424 loss=4.7528409958
36
+ epoch=3 global_step=3500 optimizer_step=436 loss=4.7702426910
37
+ epoch=3 global_step=3600 optimizer_step=449 loss=4.6359858513
38
+ epoch=3 global_step=3700 optimizer_step=461 loss=4.6865715981
39
+ epoch=4 global_step=3800 optimizer_step=473 loss=4.6681866646
40
+ epoch=4 global_step=3900 optimizer_step=486 loss=4.6668920517
41
+ epoch=4 global_step=4000 optimizer_step=498 loss=4.6514139175
42
+ epoch=4 global_step=4100 optimizer_step=511 loss=4.7157154083
43
+ epoch=4 global_step=4200 optimizer_step=523 loss=4.6793789864
44
+ epoch=4 global_step=4300 optimizer_step=536 loss=4.7080368996
45
+ epoch=4 global_step=4400 optimizer_step=548 loss=4.5494070053
46
+ epoch=4 global_step=4500 optimizer_step=561 loss=4.4851379395
47
+ epoch=4 global_step=4600 optimizer_step=573 loss=4.5187730789
48
+ epoch=5 global_step=4700 optimizer_step=586 loss=4.4088497162
49
+ epoch=5 global_step=4800 optimizer_step=598 loss=4.4902334213
50
+ epoch=5 global_step=4900 optimizer_step=611 loss=4.7872562408
51
+ epoch=5 global_step=5000 optimizer_step=623 loss=4.3599033356
52
+ epoch=5 global_step=5100 optimizer_step=636 loss=4.3905849457
53
+ epoch=5 global_step=5200 optimizer_step=648 loss=4.5187330246
54
+ epoch=5 global_step=5300 optimizer_step=661 loss=4.5045137405
55
+ epoch=5 global_step=5400 optimizer_step=673 loss=4.4139251709
56
+ epoch=5 global_step=5500 optimizer_step=686 loss=4.2963199615
57
+ epoch=5 global_step=5600 optimizer_step=698 loss=4.4732627869
58
+ epoch=6 global_step=5700 optimizer_step=710 loss=4.4239578247
59
+ epoch=6 global_step=5800 optimizer_step=723 loss=4.5372614861
60
+ epoch=6 global_step=5900 optimizer_step=735 loss=4.3654003143
61
+ epoch=6 global_step=6000 optimizer_step=748 loss=4.2875223160
62
+ epoch=6 global_step=6100 optimizer_step=760 loss=4.1896581650
63
+ epoch=6 global_step=6200 optimizer_step=773 loss=4.1644039154
64
+ epoch=6 global_step=6300 optimizer_step=785 loss=4.2756705284
65
+ epoch=6 global_step=6400 optimizer_step=798 loss=4.1578993797
66
+ epoch=6 global_step=6500 optimizer_step=810 loss=4.1570591927
67
+ epoch=7 global_step=6600 optimizer_step=823 loss=4.1190996170
68
+ epoch=7 global_step=6700 optimizer_step=835 loss=4.4135608673
69
+ epoch=7 global_step=6800 optimizer_step=848 loss=4.1999702454
70
+ epoch=7 global_step=6900 optimizer_step=860 loss=4.0975055695
71
+ epoch=7 global_step=7000 optimizer_step=873 loss=4.1644773483
72
+ epoch=7 global_step=7100 optimizer_step=885 loss=4.2144017220
73
+ epoch=7 global_step=7200 optimizer_step=898 loss=4.1876883507
74
+ epoch=7 global_step=7300 optimizer_step=910 loss=4.0543675423
75
+ epoch=7 global_step=7400 optimizer_step=923 loss=4.1312112808
76
+ epoch=7 global_step=7500 optimizer_step=935 loss=4.2607955933
77
+ epoch=8 global_step=7600 optimizer_step=947 loss=4.1456279755
78
+ epoch=8 global_step=7700 optimizer_step=960 loss=4.0029535294
79
+ epoch=8 global_step=7800 optimizer_step=972 loss=4.0145635605
80
+ epoch=8 global_step=7900 optimizer_step=985 loss=4.0330953598
81
+ epoch=8 global_step=8000 optimizer_step=997 loss=4.1285309792
82
+ epoch=8 global_step=8100 optimizer_step=1010 loss=3.9007530212
83
+ epoch=8 global_step=8200 optimizer_step=1022 loss=4.1823410988
84
+ epoch=8 global_step=8300 optimizer_step=1035 loss=4.0233097076
85
+ epoch=8 global_step=8400 optimizer_step=1047 loss=3.8265328407
86
+ epoch=9 global_step=8500 optimizer_step=1060 loss=3.9386122227
87
+ epoch=9 global_step=8600 optimizer_step=1072 loss=4.0075607300
88
+ epoch=9 global_step=8700 optimizer_step=1085 loss=3.9138417244
89
+ epoch=9 global_step=8800 optimizer_step=1097 loss=4.0552701950
90
+ epoch=9 global_step=8900 optimizer_step=1110 loss=3.9374799728
91
+ epoch=9 global_step=9000 optimizer_step=1122 loss=3.9382710457
92
+ epoch=9 global_step=9100 optimizer_step=1135 loss=3.9207730293
93
+ epoch=9 global_step=9200 optimizer_step=1147 loss=3.8809776306
94
+ epoch=9 global_step=9300 optimizer_step=1160 loss=3.9859008789
95
+ epoch=10 global_step=9400 optimizer_step=1172 loss=3.8691575527
96
+ epoch=10 global_step=9500 optimizer_step=1184 loss=3.9256129265
97
+ epoch=10 global_step=9600 optimizer_step=1197 loss=3.8335113525
98
+ epoch=10 global_step=9700 optimizer_step=1209 loss=3.9376654625
99
+ epoch=10 global_step=9800 optimizer_step=1222 loss=3.8290793896
100
+ epoch=10 global_step=9900 optimizer_step=1234 loss=3.8001422882
101
+ epoch=10 global_step=10000 optimizer_step=1247 loss=3.9075949192
102
+ epoch=10 global_step=10100 optimizer_step=1259 loss=3.8084189892
103
+ epoch=10 global_step=10200 optimizer_step=1272 loss=3.8882963657
104
+ epoch=10 global_step=10300 optimizer_step=1284 loss=3.7760980129
105
+ epoch=11 global_step=10400 optimizer_step=1297 loss=3.7978472710
106
+ epoch=11 global_step=10500 optimizer_step=1309 loss=3.8528442383
107
+ epoch=11 global_step=10600 optimizer_step=1322 loss=3.7510962486
108
+ epoch=11 global_step=10700 optimizer_step=1334 loss=3.7886292934
109
+ epoch=11 global_step=10800 optimizer_step=1347 loss=3.8759219646
110
+ epoch=11 global_step=10900 optimizer_step=1359 loss=3.6976299286
111
+ epoch=11 global_step=11000 optimizer_step=1372 loss=3.8747618198
112
+ epoch=11 global_step=11100 optimizer_step=1384 loss=3.7198715210
113
+ epoch=11 global_step=11200 optimizer_step=1397 loss=3.8817169666
114
+ epoch=12 global_step=11300 optimizer_step=1409 loss=3.8562741280
115
+ epoch=12 global_step=11400 optimizer_step=1421 loss=3.7765541077
116
+ epoch=12 global_step=11500 optimizer_step=1434 loss=3.7739019394
117
+ epoch=12 global_step=11600 optimizer_step=1446 loss=3.7178392410
118
+ epoch=12 global_step=11700 optimizer_step=1459 loss=3.8401055336
119
+ epoch=12 global_step=11800 optimizer_step=1471 loss=3.6788630486
120
+ epoch=12 global_step=11900 optimizer_step=1484 loss=3.6729793549
121
+ epoch=12 global_step=12000 optimizer_step=1496 loss=3.6811699867
122
+ epoch=12 global_step=12100 optimizer_step=1509 loss=3.5164837837
123
+ epoch=13 global_step=12200 optimizer_step=1521 loss=3.6303305626
124
+ epoch=13 global_step=12300 optimizer_step=1534 loss=3.5845396519
125
+ epoch=13 global_step=12400 optimizer_step=1546 loss=3.9264063835
126
+ epoch=13 global_step=12500 optimizer_step=1559 loss=3.7047080994
127
+ epoch=13 global_step=12600 optimizer_step=1571 loss=3.6047692299
128
+ epoch=13 global_step=12700 optimizer_step=1584 loss=3.6126563549
129
+ epoch=13 global_step=12800 optimizer_step=1596 loss=3.6897406578
130
+ epoch=13 global_step=12900 optimizer_step=1609 loss=3.6269822121
131
+ epoch=13 global_step=13000 optimizer_step=1621 loss=3.5983562469
132
+ epoch=13 global_step=13100 optimizer_step=1634 loss=3.7300169468
133
+ epoch=14 global_step=13200 optimizer_step=1646 loss=3.6430127621
134
+ epoch=14 global_step=13300 optimizer_step=1658 loss=3.6717987061
135
+ epoch=14 global_step=13400 optimizer_step=1671 loss=3.5239419937
136
+ epoch=14 global_step=13500 optimizer_step=1683 loss=3.7072567940
137
+ epoch=14 global_step=13600 optimizer_step=1696 loss=3.7336971760
138
+ epoch=14 global_step=13700 optimizer_step=1708 loss=3.4893553257
139
+ epoch=14 global_step=13800 optimizer_step=1721 loss=3.6371455193
140
+ epoch=14 global_step=13900 optimizer_step=1733 loss=3.5180556774
141
+ epoch=14 global_step=14000 optimizer_step=1746 loss=3.5890464783
142
+ epoch=15 global_step=14100 optimizer_step=1758 loss=3.7274138927
143
+ epoch=15 global_step=14200 optimizer_step=1771 loss=3.5375967026
144
+ epoch=15 global_step=14300 optimizer_step=1783 loss=3.4809908867
145
+ epoch=15 global_step=14400 optimizer_step=1796 loss=3.4846811295
146
+ epoch=15 global_step=14500 optimizer_step=1808 loss=3.5587661266
147
+ epoch=15 global_step=14600 optimizer_step=1821 loss=3.5376098156
148
+ epoch=15 global_step=14700 optimizer_step=1833 loss=3.6073918343
149
+ epoch=15 global_step=14800 optimizer_step=1846 loss=3.4121999741
150
+ epoch=15 global_step=14900 optimizer_step=1858 loss=3.5462059975
151
+ epoch=15 global_step=15000 optimizer_step=1871 loss=3.5630211830
152
+ epoch=16 global_step=15100 optimizer_step=1883 loss=3.5349659920
153
+ epoch=16 global_step=15200 optimizer_step=1895 loss=3.7158668041
154
+ epoch=16 global_step=15300 optimizer_step=1908 loss=3.5764598846
155
+ epoch=16 global_step=15400 optimizer_step=1920 loss=3.3588581085
156
+ epoch=16 global_step=15500 optimizer_step=1933 loss=3.5687055588
157
+ epoch=16 global_step=15600 optimizer_step=1945 loss=3.5069465637
158
+ epoch=16 global_step=15700 optimizer_step=1958 loss=3.5178790092
159
+ epoch=16 global_step=15800 optimizer_step=1970 loss=3.5098664761
160
+ epoch=16 global_step=15900 optimizer_step=1983 loss=3.6788122654
161
+ epoch=17 global_step=16000 optimizer_step=1995 loss=3.5321400166
162
+ epoch=17 global_step=16100 optimizer_step=2008 loss=3.4204969406
163
+ epoch=17 global_step=16200 optimizer_step=2020 loss=3.3072724342
164
+ epoch=17 global_step=16300 optimizer_step=2033 loss=3.4270269871
165
+ epoch=17 global_step=16400 optimizer_step=2045 loss=3.4885556698
166
+ epoch=17 global_step=16500 optimizer_step=2058 loss=3.4741663933
167
+ epoch=17 global_step=16600 optimizer_step=2070 loss=3.7207236290
168
+ epoch=17 global_step=16700 optimizer_step=2083 loss=3.4049930573
169
+ epoch=17 global_step=16800 optimizer_step=2095 loss=3.3907754421
170
+ epoch=18 global_step=16900 optimizer_step=2107 loss=3.3766684532
171
+ epoch=18 global_step=17000 optimizer_step=2120 loss=3.5890717506
172
+ epoch=18 global_step=17100 optimizer_step=2132 loss=3.2862682343
173
+ epoch=18 global_step=17200 optimizer_step=2145 loss=3.4579505920
174
+ epoch=18 global_step=17300 optimizer_step=2157 loss=3.4769377708
175
+ epoch=18 global_step=17400 optimizer_step=2170 loss=3.5202500820
176
+ epoch=18 global_step=17500 optimizer_step=2182 loss=3.3922853470
177
+ epoch=18 global_step=17600 optimizer_step=2195 loss=3.4541213512
178
+ epoch=18 global_step=17700 optimizer_step=2207 loss=3.4019641876
179
+ epoch=18 global_step=17800 optimizer_step=2220 loss=3.3775620461
180
+ epoch=19 global_step=17900 optimizer_step=2232 loss=3.4110507965
181
+ epoch=19 global_step=18000 optimizer_step=2245 loss=3.3769717216
182
+ epoch=19 global_step=18100 optimizer_step=2257 loss=3.4175500870
183
+ epoch=19 global_step=18200 optimizer_step=2270 loss=3.5965318680
184
+ epoch=19 global_step=18300 optimizer_step=2282 loss=3.3860187531
185
+ epoch=19 global_step=18400 optimizer_step=2295 loss=3.2395594120
186
+ epoch=19 global_step=18500 optimizer_step=2307 loss=3.3778638840
187
+ epoch=19 global_step=18600 optimizer_step=2320 loss=3.5496289730
188
+ epoch=19 global_step=18700 optimizer_step=2332 loss=3.4751753807
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune4_start11/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune7_start10/added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune7_start10/chat_template.jinja ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
27
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
28
+ {%- elif message.role == "assistant" %}
29
+ {%- set content = message.content %}
30
+ {%- set reasoning_content = '' %}
31
+ {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
32
+ {%- set reasoning_content = message.reasoning_content %}
33
+ {%- else %}
34
+ {%- if '</think>' in message.content %}
35
+ {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
36
+ {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
37
+ {%- endif %}
38
+ {%- endif %}
39
+ {%- if loop.index0 > ns.last_query_index %}
40
+ {%- if loop.last or (not loop.last and reasoning_content) %}
41
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
42
+ {%- else %}
43
+ {{- '<|im_start|>' + message.role + '\n' + content }}
44
+ {%- endif %}
45
+ {%- else %}
46
+ {{- '<|im_start|>' + message.role + '\n' + content }}
47
+ {%- endif %}
48
+ {%- if message.tool_calls %}
49
+ {%- for tool_call in message.tool_calls %}
50
+ {%- if (loop.first and content) or (not loop.first) %}
51
+ {{- '\n' }}
52
+ {%- endif %}
53
+ {%- if tool_call.function %}
54
+ {%- set tool_call = tool_call.function %}
55
+ {%- endif %}
56
+ {{- '<tool_call>\n{"name": "' }}
57
+ {{- tool_call.name }}
58
+ {{- '", "arguments": ' }}
59
+ {%- if tool_call.arguments is string %}
60
+ {{- tool_call.arguments }}
61
+ {%- else %}
62
+ {{- tool_call.arguments | tojson }}
63
+ {%- endif %}
64
+ {{- '}\n</tool_call>' }}
65
+ {%- endfor %}
66
+ {%- endif %}
67
+ {{- '<|im_end|>\n' }}
68
+ {%- elif message.role == "tool" %}
69
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
70
+ {{- '<|im_start|>user' }}
71
+ {%- endif %}
72
+ {{- '\n<tool_response>\n' }}
73
+ {{- message.content }}
74
+ {{- '\n</tool_response>' }}
75
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
76
+ {{- '<|im_end|>\n' }}
77
+ {%- endif %}
78
+ {%- endif %}
79
+ {%- endfor %}
80
+ {%- if add_generation_prompt %}
81
+ {{- '<|im_start|>assistant\n' }}
82
+ {%- if enable_thinking is defined and enable_thinking is false %}
83
+ {{- '<think>\n\n</think>\n\n' }}
84
+ {%- endif %}
85
+ {%- endif %}
llmstream_results/qwen3_17b_base_streamline_slimpajama_prune7_start10/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151643,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 6144,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention"
37
+ ],
38
+ "max_position_embeddings": 32768,
39
+ "max_window_layers": 28,
40
+ "model_type": "qwen3",
41
+ "num_attention_heads": 16,
42
+ "num_hidden_layers": 21,
43
+ "num_key_value_heads": 8,
44
+ "rms_norm_eps": 1e-06,
45
+ "rope_scaling": null,
46
+ "rope_theta": 1000000,
47
+ "sliding_window": null,
48
+ "tie_word_embeddings": true,
49
+ "transformers_version": "4.57.6",
50
+ "use_cache": true,
51
+ "use_sliding_window": false,
52
+ "vocab_size": 151936
53
+ }