Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- Models/50/chat_template.jinja +61 -0
- Models/50/config.json +118 -0
- Models/50/generation_config.json +12 -0
- Models/50/model.safetensors +3 -0
- Models/50/quant_log.csv +253 -0
- Models/50/quantize_config.json +47 -0
- Models/50/tokenizer.json +3 -0
- Models/50/tokenizer_config.json +30 -0
.gitattributes
CHANGED
|
@@ -34,3 +34,4 @@ saved_model/**/* 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 |
Models/70/tokenizer.json 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 |
Models/70/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
Models/50/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
Models/50/chat_template.jinja
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
{%- for message in messages %}
|
| 18 |
+
{%- if message.content is string %}
|
| 19 |
+
{%- set content = message.content %}
|
| 20 |
+
{%- else %}
|
| 21 |
+
{%- set content = '' %}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 24 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 25 |
+
{%- elif message.role == "assistant" %}
|
| 26 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 27 |
+
{%- if message.tool_calls %}
|
| 28 |
+
{%- for tool_call in message.tool_calls %}
|
| 29 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 30 |
+
{{- '\n' }}
|
| 31 |
+
{%- endif %}
|
| 32 |
+
{%- if tool_call.function %}
|
| 33 |
+
{%- set tool_call = tool_call.function %}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 36 |
+
{{- tool_call.name }}
|
| 37 |
+
{{- '", "arguments": ' }}
|
| 38 |
+
{%- if tool_call.arguments is string %}
|
| 39 |
+
{{- tool_call.arguments }}
|
| 40 |
+
{%- else %}
|
| 41 |
+
{{- tool_call.arguments | tojson }}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{{- '}\n</tool_call>' }}
|
| 44 |
+
{%- endfor %}
|
| 45 |
+
{%- endif %}
|
| 46 |
+
{{- '<|im_end|>\n' }}
|
| 47 |
+
{%- elif message.role == "tool" %}
|
| 48 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 49 |
+
{{- '<|im_start|>user' }}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{{- '\n<tool_response>\n' }}
|
| 52 |
+
{{- content }}
|
| 53 |
+
{{- '\n</tool_response>' }}
|
| 54 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 55 |
+
{{- '<|im_end|>\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- endif %}
|
| 58 |
+
{%- endfor %}
|
| 59 |
+
{%- if add_generation_prompt %}
|
| 60 |
+
{{- '<|im_start|>assistant\n' }}
|
| 61 |
+
{%- endif %}
|
Models/50/config.json
ADDED
|
@@ -0,0 +1,118 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 151645,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2560,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 9728,
|
| 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 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"full_attention"
|
| 52 |
+
],
|
| 53 |
+
"max_position_embeddings": 262144,
|
| 54 |
+
"max_window_layers": 36,
|
| 55 |
+
"model_type": "qwen3",
|
| 56 |
+
"num_attention_heads": 32,
|
| 57 |
+
"num_hidden_layers": 36,
|
| 58 |
+
"num_key_value_heads": 8,
|
| 59 |
+
"pad_token_id": 151645,
|
| 60 |
+
"quantization_config": {
|
| 61 |
+
"bits": 4,
|
| 62 |
+
"checkpoint_format": "gptq",
|
| 63 |
+
"desc_act": false,
|
| 64 |
+
"format": "gptq",
|
| 65 |
+
"group_size": 128,
|
| 66 |
+
"lm_head": false,
|
| 67 |
+
"meta": {
|
| 68 |
+
"act_group_aware": true,
|
| 69 |
+
"auto_forward_data_parallel": true,
|
| 70 |
+
"damp_auto_increment": 0.01,
|
| 71 |
+
"damp_percent": 0.05,
|
| 72 |
+
"dense_vram_strategy": "exclusive",
|
| 73 |
+
"dense_vram_strategy_devices": null,
|
| 74 |
+
"fallback": {
|
| 75 |
+
"smooth": null,
|
| 76 |
+
"strategy": "rtn",
|
| 77 |
+
"threshold": "0.5%"
|
| 78 |
+
},
|
| 79 |
+
"foem": null,
|
| 80 |
+
"gc_mode": "interval",
|
| 81 |
+
"gptaq": null,
|
| 82 |
+
"hessian": {
|
| 83 |
+
"chunk_bytes": null,
|
| 84 |
+
"chunk_size": null,
|
| 85 |
+
"staging_dtype": "float32"
|
| 86 |
+
},
|
| 87 |
+
"mock_quantization": false,
|
| 88 |
+
"moe_vram_strategy": "exclusive",
|
| 89 |
+
"moe_vram_strategy_devices": null,
|
| 90 |
+
"mse": 0.0,
|
| 91 |
+
"offload_to_disk": true,
|
| 92 |
+
"offload_to_disk_path": "/tmp/gptqmodel_q5htow1y",
|
| 93 |
+
"pack_impl": "cpu",
|
| 94 |
+
"quantizer": [
|
| 95 |
+
"gptqmodel:7.0.0"
|
| 96 |
+
],
|
| 97 |
+
"static_groups": false,
|
| 98 |
+
"true_sequential": true,
|
| 99 |
+
"uri": "https://github.com/modelcloud/gptqmodel",
|
| 100 |
+
"wait_for_submodule_finalizers": false
|
| 101 |
+
},
|
| 102 |
+
"method": "gptq",
|
| 103 |
+
"pack_dtype": "int32",
|
| 104 |
+
"quant_method": "gptq",
|
| 105 |
+
"sym": true
|
| 106 |
+
},
|
| 107 |
+
"rms_norm_eps": 1e-06,
|
| 108 |
+
"rope_parameters": {
|
| 109 |
+
"rope_theta": 5000000,
|
| 110 |
+
"rope_type": "default"
|
| 111 |
+
},
|
| 112 |
+
"sliding_window": null,
|
| 113 |
+
"tie_word_embeddings": true,
|
| 114 |
+
"transformers_version": "5.7.0",
|
| 115 |
+
"use_cache": true,
|
| 116 |
+
"use_sliding_window": false,
|
| 117 |
+
"vocab_size": 151936
|
| 118 |
+
}
|
Models/50/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"temperature": 0.7,
|
| 9 |
+
"top_k": 20,
|
| 10 |
+
"top_p": 0.8,
|
| 11 |
+
"transformers_version": "5.7.0"
|
| 12 |
+
}
|
Models/50/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f1aa38993a73cba380534ec92fa6b63a51d9c5e00c91d5eae781b8f1d00ce3b
|
| 3 |
+
size 2669888992
|
Models/50/quant_log.csv
ADDED
|
@@ -0,0 +1,253 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.q_proj,0.0000001497,0.05000,4.382
|
| 3 |
+
0,self_attn.k_proj,0.0000000404,0.05000,4.441
|
| 4 |
+
0,self_attn.v_proj,0.0000000348,0.05000,4.350
|
| 5 |
+
0,self_attn.o_proj,0.0000000850,0.05000,1.280
|
| 6 |
+
0,mlp.gate_proj,0.0000102160,0.05000,1.724
|
| 7 |
+
0,mlp.up_proj,0.0000089166,0.05000,1.743
|
| 8 |
+
0,mlp.down_proj,0.0000010669,0.05000,3.375
|
| 9 |
+
1,self_attn.q_proj,0.0000002582,0.05000,4.558
|
| 10 |
+
1,self_attn.k_proj,0.0000000700,0.05000,4.591
|
| 11 |
+
1,self_attn.v_proj,0.0000000709,0.05000,4.631
|
| 12 |
+
1,self_attn.o_proj,0.0000001153,0.05000,1.476
|
| 13 |
+
1,mlp.gate_proj,0.0003654537,0.05000,2.372
|
| 14 |
+
1,mlp.up_proj,0.0001678912,0.05000,2.411
|
| 15 |
+
1,mlp.down_proj,0.0000015869,0.05000,3.571
|
| 16 |
+
2,self_attn.k_proj,0.0000002100,0.05000,4.218
|
| 17 |
+
2,self_attn.v_proj,0.0000001973,0.05000,4.249
|
| 18 |
+
2,self_attn.q_proj,0.0000007267,0.05000,4.284
|
| 19 |
+
2,self_attn.o_proj,0.0000001442,0.05000,1.321
|
| 20 |
+
2,mlp.gate_proj,0.0006378509,0.05000,1.795
|
| 21 |
+
2,mlp.up_proj,0.0005036621,0.05000,1.815
|
| 22 |
+
2,mlp.down_proj,0.0000010006,0.05000,3.541
|
| 23 |
+
3,self_attn.v_proj,0.0000003527,0.05000,3.926
|
| 24 |
+
3,self_attn.q_proj,0.0000013755,0.05000,4.000
|
| 25 |
+
3,self_attn.k_proj,0.0000003573,0.05000,4.022
|
| 26 |
+
3,self_attn.o_proj,0.0000002609,0.05000,1.337
|
| 27 |
+
3,mlp.gate_proj,0.0005461829,0.05000,1.862
|
| 28 |
+
3,mlp.up_proj,0.0002773691,0.05000,1.880
|
| 29 |
+
3,mlp.down_proj,0.0000025155,0.05000,3.506
|
| 30 |
+
4,self_attn.v_proj,0.0000006756,0.05000,3.921
|
| 31 |
+
4,self_attn.k_proj,0.0000006994,0.05000,3.970
|
| 32 |
+
4,self_attn.q_proj,0.0000025318,0.05000,4.003
|
| 33 |
+
4,self_attn.o_proj,0.0000003630,0.05000,1.330
|
| 34 |
+
4,mlp.up_proj,0.0001704913,0.05000,1.772
|
| 35 |
+
4,mlp.gate_proj,0.0004582719,0.05000,1.796
|
| 36 |
+
4,mlp.down_proj,0.0000030722,0.05000,3.513
|
| 37 |
+
5,self_attn.q_proj,0.0000027408,0.05000,3.876
|
| 38 |
+
5,self_attn.v_proj,0.0000007342,0.05000,3.951
|
| 39 |
+
5,self_attn.k_proj,0.0000007074,0.05000,3.970
|
| 40 |
+
5,self_attn.o_proj,0.0000006741,0.05000,1.321
|
| 41 |
+
5,mlp.gate_proj,0.0001597727,0.05000,1.780
|
| 42 |
+
5,mlp.up_proj,0.0000717189,0.05000,1.806
|
| 43 |
+
5,mlp.down_proj,0.0000043840,0.05000,3.511
|
| 44 |
+
6,self_attn.k_proj,0.0000014229,0.05000,3.949
|
| 45 |
+
6,self_attn.q_proj,0.0000059482,0.05000,4.004
|
| 46 |
+
6,self_attn.v_proj,0.0000015715,0.05000,4.024
|
| 47 |
+
6,self_attn.o_proj,0.0000014801,0.05000,1.334
|
| 48 |
+
6,mlp.up_proj,0.0000913276,0.05000,1.890
|
| 49 |
+
6,mlp.gate_proj,0.0001743651,0.05000,1.909
|
| 50 |
+
6,mlp.down_proj,0.0000798332,0.05000,3.506
|
| 51 |
+
7,self_attn.k_proj,0.0000034002,0.05000,3.909
|
| 52 |
+
7,self_attn.q_proj,0.0000122388,0.05000,3.980
|
| 53 |
+
7,self_attn.v_proj,0.0000031944,0.05000,4.008
|
| 54 |
+
7,self_attn.o_proj,0.0000018697,0.05000,1.365
|
| 55 |
+
7,mlp.gate_proj,0.0001922215,0.05000,1.803
|
| 56 |
+
7,mlp.up_proj,0.0001066429,0.05000,1.820
|
| 57 |
+
7,mlp.down_proj,0.0000077401,0.05000,3.487
|
| 58 |
+
8,self_attn.k_proj,0.0000046752,0.05000,3.969
|
| 59 |
+
8,self_attn.v_proj,0.0000047731,0.05000,4.012
|
| 60 |
+
8,self_attn.q_proj,0.0000178354,0.05000,4.034
|
| 61 |
+
8,self_attn.o_proj,0.0000024056,0.05000,1.357
|
| 62 |
+
8,mlp.gate_proj,0.0001238246,0.05000,1.818
|
| 63 |
+
8,mlp.up_proj,0.0000997321,0.05000,1.841
|
| 64 |
+
8,mlp.down_proj,0.0000107304,0.05000,3.535
|
| 65 |
+
9,self_attn.k_proj,0.0000066289,0.05000,3.922
|
| 66 |
+
9,self_attn.v_proj,0.0000057406,0.05000,3.985
|
| 67 |
+
9,self_attn.q_proj,0.0000228142,0.05000,3.999
|
| 68 |
+
9,self_attn.o_proj,0.0000030817,0.05000,1.339
|
| 69 |
+
9,mlp.gate_proj,0.0001988376,0.05000,1.867
|
| 70 |
+
9,mlp.up_proj,0.0001221854,0.05000,1.890
|
| 71 |
+
9,mlp.down_proj,0.0000100834,0.05000,3.498
|
| 72 |
+
10,self_attn.v_proj,0.0000088171,0.05000,3.950
|
| 73 |
+
10,self_attn.q_proj,0.0000320740,0.05000,3.982
|
| 74 |
+
10,self_attn.k_proj,0.0000083633,0.05000,3.991
|
| 75 |
+
10,self_attn.o_proj,0.0000037997,0.05000,1.328
|
| 76 |
+
10,mlp.gate_proj,0.0001482455,0.05000,1.833
|
| 77 |
+
10,mlp.up_proj,0.0001002380,0.05000,1.847
|
| 78 |
+
10,mlp.down_proj,0.0000081155,0.05000,3.547
|
| 79 |
+
11,self_attn.q_proj,0.0000157204,0.05000,3.920
|
| 80 |
+
11,self_attn.k_proj,0.0000043992,0.05000,3.961
|
| 81 |
+
11,self_attn.v_proj,0.0000040354,0.05000,3.979
|
| 82 |
+
11,self_attn.o_proj,0.0000023009,0.05000,1.361
|
| 83 |
+
11,mlp.gate_proj,0.0001224862,0.05000,1.821
|
| 84 |
+
11,mlp.up_proj,0.0000944792,0.05000,1.843
|
| 85 |
+
11,mlp.down_proj,0.0000084048,0.05000,3.499
|
| 86 |
+
12,self_attn.k_proj,0.0000053237,0.05000,3.876
|
| 87 |
+
12,self_attn.q_proj,0.0000195351,0.05000,3.972
|
| 88 |
+
12,self_attn.v_proj,0.0000052086,0.05000,4.002
|
| 89 |
+
12,self_attn.o_proj,0.0000033271,0.05000,1.326
|
| 90 |
+
12,mlp.up_proj,0.0000952316,0.05000,1.945
|
| 91 |
+
12,mlp.gate_proj,0.0001117326,0.05000,1.964
|
| 92 |
+
12,mlp.down_proj,0.0000084220,0.05000,3.492
|
| 93 |
+
13,self_attn.k_proj,0.0000040553,0.05000,3.982
|
| 94 |
+
13,self_attn.v_proj,0.0000035740,0.05000,4.028
|
| 95 |
+
13,self_attn.q_proj,0.0000149411,0.05000,4.051
|
| 96 |
+
13,self_attn.o_proj,0.0000025440,0.05000,1.361
|
| 97 |
+
13,mlp.up_proj,0.0000980827,0.05000,1.869
|
| 98 |
+
13,mlp.gate_proj,0.0001053022,0.05000,1.887
|
| 99 |
+
13,mlp.down_proj,0.0000091428,0.05000,3.461
|
| 100 |
+
14,self_attn.v_proj,0.0000062012,0.05000,3.923
|
| 101 |
+
14,self_attn.q_proj,0.0000241531,0.05000,4.003
|
| 102 |
+
14,self_attn.k_proj,0.0000063827,0.05000,4.028
|
| 103 |
+
14,self_attn.o_proj,0.0000039574,0.05000,1.324
|
| 104 |
+
14,mlp.gate_proj,0.0001051134,0.05000,1.836
|
| 105 |
+
14,mlp.up_proj,0.0000985452,0.05000,1.853
|
| 106 |
+
14,mlp.down_proj,0.0000083144,0.05000,3.479
|
| 107 |
+
15,self_attn.k_proj,0.0000065399,0.05000,3.894
|
| 108 |
+
15,self_attn.q_proj,0.0000246789,0.05000,3.961
|
| 109 |
+
15,self_attn.v_proj,0.0000057165,0.05000,4.000
|
| 110 |
+
15,self_attn.o_proj,0.0000033902,0.05000,1.349
|
| 111 |
+
15,mlp.gate_proj,0.0000980051,0.05000,1.929
|
| 112 |
+
15,mlp.up_proj,0.0000952363,0.05000,1.953
|
| 113 |
+
15,mlp.down_proj,0.0000081789,0.05000,3.566
|
| 114 |
+
16,self_attn.v_proj,0.0000111526,0.05000,3.887
|
| 115 |
+
16,self_attn.k_proj,0.0000106205,0.05000,3.942
|
| 116 |
+
16,self_attn.q_proj,0.0000424247,0.05000,3.978
|
| 117 |
+
16,self_attn.o_proj,0.0000038802,0.05000,1.318
|
| 118 |
+
16,mlp.up_proj,0.0000966814,0.05000,1.790
|
| 119 |
+
16,mlp.gate_proj,0.0001058810,0.05000,1.820
|
| 120 |
+
16,mlp.down_proj,0.0000451322,0.05000,3.495
|
| 121 |
+
17,self_attn.q_proj,0.0000358253,0.05000,3.836
|
| 122 |
+
17,self_attn.k_proj,0.0000080998,0.05000,3.941
|
| 123 |
+
17,self_attn.v_proj,0.0000082012,0.05000,3.967
|
| 124 |
+
17,self_attn.o_proj,0.0000036289,0.05000,1.328
|
| 125 |
+
17,mlp.gate_proj,0.0000921126,0.05000,1.786
|
| 126 |
+
17,mlp.up_proj,0.0000849465,0.05000,1.782
|
| 127 |
+
17,mlp.down_proj,0.0000071025,0.05000,3.466
|
| 128 |
+
18,self_attn.v_proj,0.0000099717,0.05000,3.973
|
| 129 |
+
18,self_attn.k_proj,0.0000098200,0.05000,3.997
|
| 130 |
+
18,self_attn.q_proj,0.0000393778,0.05000,4.048
|
| 131 |
+
18,self_attn.o_proj,0.0000046202,0.05000,1.349
|
| 132 |
+
18,mlp.up_proj,0.0000947463,0.05000,1.908
|
| 133 |
+
18,mlp.gate_proj,0.0001002435,0.05000,1.925
|
| 134 |
+
18,mlp.down_proj,0.0000094912,0.05000,3.519
|
| 135 |
+
19,self_attn.v_proj,0.0000176808,0.05000,3.878
|
| 136 |
+
19,self_attn.q_proj,0.0000786607,0.05000,3.975
|
| 137 |
+
19,self_attn.k_proj,0.0000189547,0.05000,3.977
|
| 138 |
+
19,self_attn.o_proj,0.0000072878,0.05000,1.341
|
| 139 |
+
19,mlp.gate_proj,0.0001130672,0.05000,1.850
|
| 140 |
+
19,mlp.up_proj,0.0001079505,0.05000,1.874
|
| 141 |
+
19,mlp.down_proj,0.0000130190,0.05000,3.474
|
| 142 |
+
20,self_attn.v_proj,0.0000165799,0.05000,3.919
|
| 143 |
+
20,self_attn.k_proj,0.0000168190,0.05000,3.972
|
| 144 |
+
20,self_attn.q_proj,0.0000756451,0.05000,3.986
|
| 145 |
+
20,self_attn.o_proj,0.0000069506,0.05000,1.357
|
| 146 |
+
20,mlp.gate_proj,0.0001250732,0.05000,1.799
|
| 147 |
+
20,mlp.up_proj,0.0001220248,0.05000,1.819
|
| 148 |
+
20,mlp.down_proj,0.0000152876,0.05000,3.501
|
| 149 |
+
21,self_attn.v_proj,0.0000228874,0.05000,3.940
|
| 150 |
+
21,self_attn.q_proj,0.0000968350,0.05000,3.979
|
| 151 |
+
21,self_attn.k_proj,0.0000223913,0.05000,4.006
|
| 152 |
+
21,self_attn.o_proj,0.0000090330,0.05000,1.340
|
| 153 |
+
21,mlp.gate_proj,0.0001505156,0.05000,1.844
|
| 154 |
+
21,mlp.up_proj,0.0001460762,0.05000,1.855
|
| 155 |
+
21,mlp.down_proj,0.0000205126,0.05000,3.539
|
| 156 |
+
22,self_attn.v_proj,0.0000470790,0.05000,3.991
|
| 157 |
+
22,self_attn.q_proj,0.0002022357,0.05000,4.041
|
| 158 |
+
22,self_attn.k_proj,0.0000466329,0.05000,4.060
|
| 159 |
+
22,self_attn.o_proj,0.0000183918,0.05000,1.325
|
| 160 |
+
22,mlp.gate_proj,0.0001807079,0.05000,1.864
|
| 161 |
+
22,mlp.up_proj,0.0001656352,0.05000,1.887
|
| 162 |
+
22,mlp.down_proj,0.0000324256,0.05000,3.467
|
| 163 |
+
23,self_attn.k_proj,0.0000448676,0.05000,3.970
|
| 164 |
+
23,self_attn.v_proj,0.0000475164,0.05000,4.023
|
| 165 |
+
23,self_attn.q_proj,0.0001911028,0.05000,4.047
|
| 166 |
+
23,self_attn.o_proj,0.0000203201,0.05000,1.321
|
| 167 |
+
23,mlp.up_proj,0.0001968862,0.05000,1.826
|
| 168 |
+
23,mlp.gate_proj,0.0002260412,0.05000,1.850
|
| 169 |
+
23,mlp.down_proj,0.0000465797,0.05000,3.430
|
| 170 |
+
24,self_attn.v_proj,0.0000918311,0.05000,3.915
|
| 171 |
+
24,self_attn.q_proj,0.0003130213,0.05000,3.980
|
| 172 |
+
24,self_attn.k_proj,0.0000751774,0.05000,4.005
|
| 173 |
+
24,self_attn.o_proj,0.0000240284,0.05000,1.329
|
| 174 |
+
24,mlp.up_proj,0.0002207510,0.05000,1.803
|
| 175 |
+
24,mlp.gate_proj,0.0002575789,0.05000,1.812
|
| 176 |
+
24,mlp.down_proj,0.0000600394,0.05000,3.719
|
| 177 |
+
25,self_attn.q_proj,0.0002166719,0.05000,3.930
|
| 178 |
+
25,self_attn.v_proj,0.0000578186,0.05000,3.994
|
| 179 |
+
25,self_attn.k_proj,0.0000542144,0.05000,4.005
|
| 180 |
+
25,self_attn.o_proj,0.0000134359,0.05000,1.332
|
| 181 |
+
25,mlp.up_proj,0.0002558766,0.05000,1.811
|
| 182 |
+
25,mlp.gate_proj,0.0003033456,0.05000,1.834
|
| 183 |
+
25,mlp.down_proj,0.0000730246,0.05000,3.487
|
| 184 |
+
26,self_attn.q_proj,0.0003547980,0.05000,3.916
|
| 185 |
+
26,self_attn.k_proj,0.0000841941,0.05000,3.954
|
| 186 |
+
26,self_attn.v_proj,0.0000925788,0.05000,3.989
|
| 187 |
+
26,self_attn.o_proj,0.0000134945,0.05000,1.351
|
| 188 |
+
26,mlp.gate_proj,0.0003733178,0.05000,1.903
|
| 189 |
+
26,mlp.up_proj,0.0003256705,0.05000,1.927
|
| 190 |
+
26,mlp.down_proj,0.0000952209,0.05000,3.523
|
| 191 |
+
27,self_attn.q_proj,0.0004478611,0.05000,3.904
|
| 192 |
+
27,self_attn.v_proj,0.0001275532,0.05000,3.981
|
| 193 |
+
27,self_attn.k_proj,0.0001037366,0.05000,4.014
|
| 194 |
+
27,self_attn.o_proj,0.0000206886,0.05000,1.319
|
| 195 |
+
27,mlp.gate_proj,0.0004283897,0.05000,1.854
|
| 196 |
+
27,mlp.up_proj,0.0003877850,0.05000,1.870
|
| 197 |
+
27,mlp.down_proj,0.0001451184,0.05000,3.689
|
| 198 |
+
28,self_attn.k_proj,0.0001290215,0.05000,3.955
|
| 199 |
+
28,self_attn.v_proj,0.0001380381,0.05000,4.015
|
| 200 |
+
28,self_attn.q_proj,0.0005028249,0.05000,4.041
|
| 201 |
+
28,self_attn.o_proj,0.0000354018,0.05000,1.331
|
| 202 |
+
28,mlp.gate_proj,0.0004989572,0.05000,1.777
|
| 203 |
+
28,mlp.up_proj,0.0004736893,0.05000,1.800
|
| 204 |
+
28,mlp.down_proj,0.0002259568,0.05000,3.485
|
| 205 |
+
29,self_attn.q_proj,0.0012338733,0.05000,3.883
|
| 206 |
+
29,self_attn.k_proj,0.0002937443,0.05000,3.952
|
| 207 |
+
29,self_attn.v_proj,0.0003944995,0.05000,3.980
|
| 208 |
+
29,self_attn.o_proj,0.0000378399,0.05000,1.396
|
| 209 |
+
29,mlp.up_proj,0.0005982406,0.05000,1.797
|
| 210 |
+
29,mlp.gate_proj,0.0005981093,0.05000,1.801
|
| 211 |
+
29,mlp.down_proj,0.0002386291,0.05000,3.496
|
| 212 |
+
30,self_attn.v_proj,0.0004350560,0.05000,3.897
|
| 213 |
+
30,self_attn.q_proj,0.0014054731,0.05000,3.941
|
| 214 |
+
30,self_attn.k_proj,0.0003603588,0.05000,3.986
|
| 215 |
+
30,self_attn.o_proj,0.0000687038,0.05000,1.319
|
| 216 |
+
30,mlp.gate_proj,0.0006366458,0.05000,1.815
|
| 217 |
+
30,mlp.up_proj,0.0006604899,0.05000,1.813
|
| 218 |
+
30,mlp.down_proj,0.0003442335,0.05000,3.540
|
| 219 |
+
31,self_attn.k_proj,0.0005021534,0.05000,4.020
|
| 220 |
+
31,self_attn.q_proj,0.0018248984,0.05000,4.098
|
| 221 |
+
31,self_attn.v_proj,0.0006612717,0.05000,4.119
|
| 222 |
+
31,self_attn.o_proj,0.0000640828,0.05000,1.338
|
| 223 |
+
31,mlp.up_proj,0.0007156640,0.05000,1.771
|
| 224 |
+
31,mlp.gate_proj,0.0006541759,0.05000,1.792
|
| 225 |
+
31,mlp.down_proj,0.0004586547,0.05000,3.540
|
| 226 |
+
32,self_attn.q_proj,0.0027697828,0.05000,3.898
|
| 227 |
+
32,self_attn.k_proj,0.0006917736,0.05000,3.967
|
| 228 |
+
32,self_attn.v_proj,0.0010175645,0.05000,4.006
|
| 229 |
+
32,self_attn.o_proj,0.0001164538,0.05000,1.352
|
| 230 |
+
32,mlp.up_proj,0.0007774354,0.05000,1.754
|
| 231 |
+
32,mlp.gate_proj,0.0006933136,0.05000,1.779
|
| 232 |
+
32,mlp.down_proj,0.0005796508,0.05000,3.510
|
| 233 |
+
33,self_attn.q_proj,0.0060845944,0.05000,3.802
|
| 234 |
+
33,self_attn.k_proj,0.0012694908,0.05000,3.924
|
| 235 |
+
33,self_attn.v_proj,0.0023390502,0.05000,3.946
|
| 236 |
+
33,self_attn.o_proj,0.0001119355,0.05000,1.334
|
| 237 |
+
33,mlp.gate_proj,0.0007650941,0.05000,1.843
|
| 238 |
+
33,mlp.up_proj,0.0008669923,0.05000,1.857
|
| 239 |
+
33,mlp.down_proj,0.0007560028,0.05000,3.585
|
| 240 |
+
34,self_attn.v_proj,0.0016800856,0.05000,3.933
|
| 241 |
+
34,self_attn.q_proj,0.0047758973,0.05000,3.971
|
| 242 |
+
34,self_attn.k_proj,0.0010792233,0.05000,4.010
|
| 243 |
+
34,self_attn.o_proj,0.0002744614,0.05000,1.330
|
| 244 |
+
34,mlp.gate_proj,0.0009299143,0.05000,1.851
|
| 245 |
+
34,mlp.up_proj,0.0009779911,0.05000,1.870
|
| 246 |
+
34,mlp.down_proj,0.0009601268,0.05000,3.484
|
| 247 |
+
35,self_attn.q_proj,0.0022361877,0.05000,3.879
|
| 248 |
+
35,self_attn.k_proj,0.0005822161,0.05000,3.916
|
| 249 |
+
35,self_attn.v_proj,0.0007247474,0.05000,3.974
|
| 250 |
+
35,self_attn.o_proj,0.0003144698,0.05000,1.357
|
| 251 |
+
35,mlp.gate_proj,0.0015252715,0.05000,1.929
|
| 252 |
+
35,mlp.up_proj,0.0015549625,0.05000,1.927
|
| 253 |
+
35,mlp.down_proj,0.0016945056,0.05000,3.497
|
Models/50/quantize_config.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bits": 4,
|
| 3 |
+
"group_size": 128,
|
| 4 |
+
"desc_act": false,
|
| 5 |
+
"lm_head": false,
|
| 6 |
+
"method": "gptq",
|
| 7 |
+
"quant_method": "gptq",
|
| 8 |
+
"format": "gptq",
|
| 9 |
+
"checkpoint_format": "gptq",
|
| 10 |
+
"pack_dtype": "int32",
|
| 11 |
+
"meta": {
|
| 12 |
+
"quantizer": [
|
| 13 |
+
"gptqmodel:7.0.0"
|
| 14 |
+
],
|
| 15 |
+
"uri": "https://github.com/modelcloud/gptqmodel",
|
| 16 |
+
"damp_percent": 0.05,
|
| 17 |
+
"damp_auto_increment": 0.01,
|
| 18 |
+
"static_groups": false,
|
| 19 |
+
"true_sequential": true,
|
| 20 |
+
"mse": 0.0,
|
| 21 |
+
"gptaq": null,
|
| 22 |
+
"foem": null,
|
| 23 |
+
"act_group_aware": true,
|
| 24 |
+
"fallback": {
|
| 25 |
+
"strategy": "rtn",
|
| 26 |
+
"threshold": "0.5%",
|
| 27 |
+
"smooth": null
|
| 28 |
+
},
|
| 29 |
+
"offload_to_disk": true,
|
| 30 |
+
"offload_to_disk_path": "/tmp/gptqmodel_q5htow1y",
|
| 31 |
+
"pack_impl": "cpu",
|
| 32 |
+
"gc_mode": "interval",
|
| 33 |
+
"wait_for_submodule_finalizers": false,
|
| 34 |
+
"auto_forward_data_parallel": true,
|
| 35 |
+
"dense_vram_strategy": "exclusive",
|
| 36 |
+
"dense_vram_strategy_devices": null,
|
| 37 |
+
"moe_vram_strategy": "exclusive",
|
| 38 |
+
"moe_vram_strategy_devices": null,
|
| 39 |
+
"mock_quantization": false,
|
| 40 |
+
"hessian": {
|
| 41 |
+
"chunk_size": null,
|
| 42 |
+
"chunk_bytes": null,
|
| 43 |
+
"staging_dtype": "float32"
|
| 44 |
+
}
|
| 45 |
+
},
|
| 46 |
+
"sym": true
|
| 47 |
+
}
|
Models/50/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:749876d361036526939013c5243386793a5ccf69b68004faff0bab4b9c156991
|
| 3 |
+
size 11422925
|
Models/50/tokenizer_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|im_start|>",
|
| 10 |
+
"<|im_end|>",
|
| 11 |
+
"<|object_ref_start|>",
|
| 12 |
+
"<|object_ref_end|>",
|
| 13 |
+
"<|box_start|>",
|
| 14 |
+
"<|box_end|>",
|
| 15 |
+
"<|quad_start|>",
|
| 16 |
+
"<|quad_end|>",
|
| 17 |
+
"<|vision_start|>",
|
| 18 |
+
"<|vision_end|>",
|
| 19 |
+
"<|vision_pad|>",
|
| 20 |
+
"<|image_pad|>",
|
| 21 |
+
"<|video_pad|>"
|
| 22 |
+
],
|
| 23 |
+
"is_local": false,
|
| 24 |
+
"local_files_only": false,
|
| 25 |
+
"model_max_length": 1010000,
|
| 26 |
+
"pad_token": "<|im_end|>",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null
|
| 30 |
+
}
|