Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- chat_template.jinja +112 -0
- config.json +55 -0
- generation_config.json +7 -0
- merge_metadata.json +89 -0
- model-00001-of-00008.safetensors +3 -0
- model-00002-of-00008.safetensors +3 -0
- model-00003-of-00008.safetensors +3 -0
- model-00004-of-00008.safetensors +3 -0
- model-00005-of-00008.safetensors +3 -0
- model-00006-of-00008.safetensors +3 -0
- model-00007-of-00008.safetensors +3 -0
- model-00008-of-00008.safetensors +3 -0
- model.safetensors.index.json +0 -0
- tokenizer.json +3 -0
- tokenizer_config.json +216 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- macro render_content(msg) -%}
|
| 2 |
+
{%- set c = msg.get('content') -%}
|
| 3 |
+
{%- if c is string -%}
|
| 4 |
+
{{ c }}
|
| 5 |
+
{%- elif c is not none -%}
|
| 6 |
+
{% for content in c -%}
|
| 7 |
+
{% if content['type'] == 'image' or content['type'] == 'image_url' -%}
|
| 8 |
+
<|media_begin|>image<|media_content|><|media_pad|><|media_end|>
|
| 9 |
+
{% elif content['type'] == 'video' or content['type']== 'video_url'-%}
|
| 10 |
+
<|kimi_k25_video_placeholder|>
|
| 11 |
+
{% else -%}
|
| 12 |
+
{{ content['text'] }}
|
| 13 |
+
{%- endif -%}
|
| 14 |
+
{%- endfor -%}
|
| 15 |
+
{%- endif -%}
|
| 16 |
+
{%- endmacro -%}
|
| 17 |
+
|
| 18 |
+
{% macro set_roles(message) -%}
|
| 19 |
+
{%- set role_name = message.get('name') or message['role'] -%}
|
| 20 |
+
{%- if message['role'] == 'user' -%}
|
| 21 |
+
<|im_user|>{{role_name}}<|im_middle|>
|
| 22 |
+
{%- elif message['role'] == 'assistant' -%}
|
| 23 |
+
<|im_assistant|>{{role_name}}<|im_middle|>
|
| 24 |
+
{%- else -%}
|
| 25 |
+
<|im_system|>{{role_name}}<|im_middle|>
|
| 26 |
+
{%- endif -%}
|
| 27 |
+
{%- endmacro -%}
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
{%- macro render_toolcalls(message) -%}
|
| 31 |
+
<|tool_calls_section_begin|>
|
| 32 |
+
{%- for tool_call in message['tool_calls'] -%}
|
| 33 |
+
{%- set formatted_id = tool_call['id'] -%}
|
| 34 |
+
<|tool_call_begin|>{{ formatted_id }}<|tool_call_argument_begin|>{% if tool_call['function']['arguments'] is string %}{{ tool_call['function']['arguments'] }}{% else %}{{ tool_call['function']['arguments'] | tojson }}{% endif %}<|tool_call_end|>
|
| 35 |
+
{%- endfor -%}
|
| 36 |
+
<|tool_calls_section_end|>
|
| 37 |
+
{%- endmacro -%}
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
{%- set preserve_thinking = preserve_thinking | default(false) -%}
|
| 41 |
+
{# Find last non-tool-call assistant message. If preserve_thinking, keep -1 so hist is empty and all msgs use suffix (retain reasoning). #}
|
| 42 |
+
{%- set ns = namespace(last_non_tool_call_assistant_msg=-1) -%}
|
| 43 |
+
{%- if not preserve_thinking -%}
|
| 44 |
+
{%- for idx in range(messages|length-1, -1, -1) -%}
|
| 45 |
+
{%- if messages[idx]['role'] == 'assistant' and not messages[idx].get('tool_calls') -%}
|
| 46 |
+
{%- set ns.last_non_tool_call_assistant_msg = idx -%}
|
| 47 |
+
{%- break -%}
|
| 48 |
+
{%- endif -%}
|
| 49 |
+
{%- endfor -%}
|
| 50 |
+
{%- endif -%}
|
| 51 |
+
|
| 52 |
+
{# split all messages into history & suffix, reasoning_content in suffix should be reserved.#}
|
| 53 |
+
{%- set hist_msgs = messages[:ns.last_non_tool_call_assistant_msg+1] -%}
|
| 54 |
+
{%- set suffix_msgs = messages[ns.last_non_tool_call_assistant_msg+1:] -%}
|
| 55 |
+
|
| 56 |
+
{%- if tools -%}
|
| 57 |
+
{%- if tools_ts_str -%}
|
| 58 |
+
<|im_system|>tool_declare<|im_middle|>{{ tools_ts_str }}<|im_end|>
|
| 59 |
+
{%- else -%}
|
| 60 |
+
<|im_system|>tool_declare<|im_middle|>{{ tools | tojson(separators=(',', ':')) }}<|im_end|>
|
| 61 |
+
{%- endif -%}
|
| 62 |
+
{%- endif -%}
|
| 63 |
+
|
| 64 |
+
|
| 65 |
+
{%- for message in hist_msgs -%}
|
| 66 |
+
{{set_roles(message)}}
|
| 67 |
+
{%- if message['role'] == 'assistant' -%}
|
| 68 |
+
<think></think>{{render_content(message)}}
|
| 69 |
+
{%- if message.get('tool_calls') -%}
|
| 70 |
+
{{render_toolcalls(message)}}
|
| 71 |
+
{%- endif -%}
|
| 72 |
+
{%- elif message['role'] == 'tool' -%}
|
| 73 |
+
{%- set tool_call_id = message.tool_call_id -%}
|
| 74 |
+
## Return of {{ tool_call_id }}
|
| 75 |
+
{{render_content(message)}}
|
| 76 |
+
{%- elif message['content'] is not none -%}
|
| 77 |
+
{{render_content(message)}}
|
| 78 |
+
{%- endif -%}
|
| 79 |
+
<|im_end|>
|
| 80 |
+
{%- endfor -%}
|
| 81 |
+
|
| 82 |
+
{%- for message in suffix_msgs -%}
|
| 83 |
+
{{set_roles(message)}}
|
| 84 |
+
{%- if message['role'] == 'assistant' -%}
|
| 85 |
+
{%- if thinking is defined and thinking is false and preserve_thinking is false -%}
|
| 86 |
+
<think></think>{{render_content(message)}}
|
| 87 |
+
{%- else -%}
|
| 88 |
+
{%- set rc = message.get('reasoning', message.get('reasoning_content', '')) -%}
|
| 89 |
+
<think>{{rc}}</think>{{render_content(message)}}
|
| 90 |
+
{%- endif -%}
|
| 91 |
+
{%- if message.get('tool_calls') -%}
|
| 92 |
+
{{render_toolcalls(message)}}
|
| 93 |
+
{%- endif -%}
|
| 94 |
+
{%- elif message['role'] == 'tool' -%}
|
| 95 |
+
{%- set tool_call_id = message.tool_call_id -%}
|
| 96 |
+
## Return of {{ tool_call_id }}
|
| 97 |
+
{{render_content(message)}}
|
| 98 |
+
{%- elif message['content'] is not none -%}
|
| 99 |
+
{{render_content(message)}}
|
| 100 |
+
{%- endif -%}
|
| 101 |
+
<|im_end|>
|
| 102 |
+
{%- endfor -%}
|
| 103 |
+
|
| 104 |
+
|
| 105 |
+
{%- if add_generation_prompt -%}
|
| 106 |
+
<|im_assistant|>assistant<|im_middle|>
|
| 107 |
+
{%- if thinking is defined and thinking is false -%}
|
| 108 |
+
<think></think>
|
| 109 |
+
{%- else -%}
|
| 110 |
+
<think>
|
| 111 |
+
{%- endif -%}
|
| 112 |
+
{%- endif -%}
|
config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"DeepseekV3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"aux_loss_alpha": 0.001,
|
| 8 |
+
"bos_token_id": 163584,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 163585,
|
| 11 |
+
"ep_size": 1,
|
| 12 |
+
"first_k_dense_replace": 1,
|
| 13 |
+
"head_dim": 64,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 2048,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 11264,
|
| 18 |
+
"kv_lora_rank": 512,
|
| 19 |
+
"max_position_embeddings": 131072,
|
| 20 |
+
"model_type": "deepseek_v3",
|
| 21 |
+
"moe_intermediate_size": 1408,
|
| 22 |
+
"moe_layer_freq": 1,
|
| 23 |
+
"n_group": 1,
|
| 24 |
+
"n_routed_experts": 64,
|
| 25 |
+
"n_shared_experts": 2,
|
| 26 |
+
"norm_topk_prob": true,
|
| 27 |
+
"num_attention_heads": 16,
|
| 28 |
+
"num_experts_per_tok": 6,
|
| 29 |
+
"num_hidden_layers": 27,
|
| 30 |
+
"num_key_value_heads": 16,
|
| 31 |
+
"num_nextn_predict_layers": 1,
|
| 32 |
+
"num_shared_experts": 2,
|
| 33 |
+
"pad_token_id": 163839,
|
| 34 |
+
"pretraining_tp": 1,
|
| 35 |
+
"q_lora_rank": null,
|
| 36 |
+
"qk_head_dim": 192,
|
| 37 |
+
"qk_nope_head_dim": 128,
|
| 38 |
+
"qk_rope_head_dim": 64,
|
| 39 |
+
"rms_norm_eps": 1e-05,
|
| 40 |
+
"rope_interleave": true,
|
| 41 |
+
"rope_parameters": {
|
| 42 |
+
"rope_theta": 800000.0,
|
| 43 |
+
"rope_type": "default"
|
| 44 |
+
},
|
| 45 |
+
"routed_scaling_factor": 2.446,
|
| 46 |
+
"scoring_func": "sigmoid",
|
| 47 |
+
"seq_aux": true,
|
| 48 |
+
"tie_word_embeddings": false,
|
| 49 |
+
"topk_group": 1,
|
| 50 |
+
"topk_method": "noaux_tc",
|
| 51 |
+
"transformers_version": "5.5.3",
|
| 52 |
+
"use_cache": false,
|
| 53 |
+
"v_head_dim": 128,
|
| 54 |
+
"vocab_size": 163840
|
| 55 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 163584,
|
| 4 |
+
"eos_token_id": 163585,
|
| 5 |
+
"pad_token_id": 163839,
|
| 6 |
+
"transformers_version": "5.5.3"
|
| 7 |
+
}
|
merge_metadata.json
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "native",
|
| 3 |
+
"base": "/root/distil_dir/train/model_base",
|
| 4 |
+
"target": "/root/distil_dir/train/model_target",
|
| 5 |
+
"method": "dare_ties",
|
| 6 |
+
"alpha": 0.18,
|
| 7 |
+
"dare_drop": 0.9,
|
| 8 |
+
"ties_keep_top": 0.2,
|
| 9 |
+
"axis_scores": null,
|
| 10 |
+
"axis_plan": {
|
| 11 |
+
"port": {},
|
| 12 |
+
"protect": []
|
| 13 |
+
},
|
| 14 |
+
"precheck": {
|
| 15 |
+
"grade": "PASS",
|
| 16 |
+
"passed": true,
|
| 17 |
+
"mergeable_layer_frac": 1.0,
|
| 18 |
+
"recommendation": "Qualified: homologous layers dominate. Use dare_ties + axis-scores; then fine-tune.",
|
| 19 |
+
"layers": [
|
| 20 |
+
{
|
| 21 |
+
"layer": 0,
|
| 22 |
+
"cosine": 0.999955415725708,
|
| 23 |
+
"rel_delta": 0.03289614524692297,
|
| 24 |
+
"sign_conflict": 0.01445366581901908,
|
| 25 |
+
"tier": "ok",
|
| 26 |
+
"mergeable": true
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"layer": 1,
|
| 30 |
+
"cosine": 0.9999827146530151,
|
| 31 |
+
"rel_delta": 0.03291291184723377,
|
| 32 |
+
"sign_conflict": 0.0117919840849936,
|
| 33 |
+
"tier": "ok",
|
| 34 |
+
"mergeable": true
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"layer": 6,
|
| 38 |
+
"cosine": 0.9999808371067047,
|
| 39 |
+
"rel_delta": 0.03292102366685867,
|
| 40 |
+
"sign_conflict": 0.0118133625946939,
|
| 41 |
+
"tier": "ok",
|
| 42 |
+
"mergeable": true
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"layer": 13,
|
| 46 |
+
"cosine": 0.9999279975891113,
|
| 47 |
+
"rel_delta": 0.032905545085668564,
|
| 48 |
+
"sign_conflict": 0.011811455246061087,
|
| 49 |
+
"tier": "ok",
|
| 50 |
+
"mergeable": true
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"layer": 20,
|
| 54 |
+
"cosine": 0.9999344050884247,
|
| 55 |
+
"rel_delta": 0.0329118799418211,
|
| 56 |
+
"sign_conflict": 0.01176464557647705,
|
| 57 |
+
"tier": "ok",
|
| 58 |
+
"mergeable": true
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"layer": 25,
|
| 62 |
+
"cosine": 0.9999194443225861,
|
| 63 |
+
"rel_delta": 0.03291044756770134,
|
| 64 |
+
"sign_conflict": 0.012151002883911133,
|
| 65 |
+
"tier": "ok",
|
| 66 |
+
"mergeable": true
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"layer": 26,
|
| 70 |
+
"cosine": 0.9999075531959534,
|
| 71 |
+
"rel_delta": 0.03291253559291363,
|
| 72 |
+
"sign_conflict": 0.01301272725686431,
|
| 73 |
+
"tier": "ok",
|
| 74 |
+
"mergeable": true
|
| 75 |
+
}
|
| 76 |
+
]
|
| 77 |
+
},
|
| 78 |
+
"cross_author": {
|
| 79 |
+
"warnings": [
|
| 80 |
+
"Cross-author merge: base='model_base' target='model_target'. Task vectors are approximate; run --precheck and recovery SFT.",
|
| 81 |
+
"Homologous architecture detected (different training runs). Prefer dare_ties or MergeKit ties; avoid frankenstein."
|
| 82 |
+
],
|
| 83 |
+
"blockers": [],
|
| 84 |
+
"same_architecture": true,
|
| 85 |
+
"same_tokenizer": true
|
| 86 |
+
},
|
| 87 |
+
"seed": 42,
|
| 88 |
+
"references": "MERGE_RESEARCH.md + mergekit (optional)"
|
| 89 |
+
}
|
model-00001-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e020094fda9515b2719439837fe9c0d32301e9a7a375f81d838d22d984b93701
|
| 3 |
+
size 5247984832
|
model-00002-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b89fe1297d6cd152125615a228c57bdeb09c18d9729d1f79b348f99b0ec407c
|
| 3 |
+
size 3814830240
|
model-00003-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd1ba76f30ed6a4e2879fd31ebe7b09c0c85765f1cb68642803ade853027296b
|
| 3 |
+
size 3809062944
|
model-00004-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:457eedef73b0f793f4d284620becf444cf911e68cdb3e90d499bed842c51b05a
|
| 3 |
+
size 3815096384
|
model-00005-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f69d63a937375d47b23402a5be56b74a0284d3636063069f6fa749e1591bac0
|
| 3 |
+
size 3807757424
|
model-00006-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b7981483d30fa95d27908a9897c0077ce329c39fce0e66758dbf40b21f44ee26
|
| 3 |
+
size 3809062848
|
model-00007-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1effdba931edfa6555e6c866886bc9162c2bc82aa1ac06fbc446d9463c9266e
|
| 3 |
+
size 3814829576
|
model-00008-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d204e4edd2ed2a53f78778dd2a622450952df2a0dc76c90de281f4845dcb70ae
|
| 3 |
+
size 3802255808
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:261590fd5234433c558e867ea7d9b8d8a29ebc6bf51d9d43a9d714e4793db0e9
|
| 3 |
+
size 19545255
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,216 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"163584": {
|
| 4 |
+
"content": "[BOS]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"163585": {
|
| 12 |
+
"content": "[EOS]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"163586": {
|
| 20 |
+
"content": "<|im_end|>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"163587": {
|
| 28 |
+
"content": "<|im_user|>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"163588": {
|
| 36 |
+
"content": "<|im_assistant|>",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"163590": {
|
| 44 |
+
"content": "<|start_header_id|>",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"163591": {
|
| 52 |
+
"content": "<|end_header_id|>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"163593": {
|
| 60 |
+
"content": "[EOT]",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"163594": {
|
| 68 |
+
"content": "<|im_system|>",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"163595": {
|
| 76 |
+
"content": "<|tool_calls_section_begin|>",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": false
|
| 82 |
+
},
|
| 83 |
+
"163596": {
|
| 84 |
+
"content": "<|tool_calls_section_end|>",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": false
|
| 90 |
+
},
|
| 91 |
+
"163597": {
|
| 92 |
+
"content": "<|tool_call_begin|>",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": false
|
| 98 |
+
},
|
| 99 |
+
"163598": {
|
| 100 |
+
"content": "<|tool_call_argument_begin|>",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": false
|
| 106 |
+
},
|
| 107 |
+
"163599": {
|
| 108 |
+
"content": "<|tool_call_end|>",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": false
|
| 114 |
+
},
|
| 115 |
+
"163601": {
|
| 116 |
+
"content": "<|im_middle|>",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"163602": {
|
| 124 |
+
"content": "<|media_begin|>",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"163603": {
|
| 132 |
+
"content": "<|media_content|>",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"163604": {
|
| 140 |
+
"content": "<|media_end|>",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"163605": {
|
| 148 |
+
"content": "<|media_pad|>",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"163606": {
|
| 156 |
+
"content": "<think>",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": false
|
| 162 |
+
},
|
| 163 |
+
"163607": {
|
| 164 |
+
"content": "</think>",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": false
|
| 170 |
+
},
|
| 171 |
+
"163838": {
|
| 172 |
+
"content": "[UNK]",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
},
|
| 179 |
+
"163839": {
|
| 180 |
+
"content": "[PAD]",
|
| 181 |
+
"lstrip": false,
|
| 182 |
+
"normalized": false,
|
| 183 |
+
"rstrip": false,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": true
|
| 186 |
+
}
|
| 187 |
+
},
|
| 188 |
+
"additional_special_tokens": [
|
| 189 |
+
"<|im_end|>",
|
| 190 |
+
"<|im_user|>",
|
| 191 |
+
"<|im_assistant|>",
|
| 192 |
+
"<|start_header_id|>",
|
| 193 |
+
"<|end_header_id|>",
|
| 194 |
+
"[EOT]",
|
| 195 |
+
"<|im_system|>",
|
| 196 |
+
"<|im_middle|>",
|
| 197 |
+
"<|media_begin|>",
|
| 198 |
+
"<|media_content|>",
|
| 199 |
+
"<|media_end|>",
|
| 200 |
+
"<|media_pad|>"
|
| 201 |
+
],
|
| 202 |
+
"bos_token": "[BOS]",
|
| 203 |
+
"clean_up_tokenization_spaces": false,
|
| 204 |
+
"eos_token": "[EOS]",
|
| 205 |
+
"extra_special_tokens": {},
|
| 206 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 207 |
+
"pad_token": "[PAD]",
|
| 208 |
+
"tokenizer_class": "TikTokenTokenizer",
|
| 209 |
+
"unk_token": "[UNK]",
|
| 210 |
+
"auto_map": {
|
| 211 |
+
"AutoTokenizer": [
|
| 212 |
+
"tokenization_kimi.TikTokenTokenizer",
|
| 213 |
+
null
|
| 214 |
+
]
|
| 215 |
+
}
|
| 216 |
+
}
|