Add ap-sql-v1 LoRA adapter
Browse files- .gitattributes +1 -0
- adapter_config.json +51 -0
- adapter_model.safetensors +3 -0
- chat_template.jinja +54 -0
- run_summary.json +1942 -0
- tokenizer.json +3 -0
- tokenizer_config.json +30 -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
|
adapter_config.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": null,
|
| 6 |
+
"base_model_name_or_path": "Qwen/Qwen2.5-Coder-7B-Instruct",
|
| 7 |
+
"bias": "none",
|
| 8 |
+
"corda_config": null,
|
| 9 |
+
"ensure_weight_tying": false,
|
| 10 |
+
"eva_config": null,
|
| 11 |
+
"exclude_modules": null,
|
| 12 |
+
"fan_in_fan_out": false,
|
| 13 |
+
"inference_mode": true,
|
| 14 |
+
"init_lora_weights": true,
|
| 15 |
+
"kasa_config": null,
|
| 16 |
+
"layer_replication": null,
|
| 17 |
+
"layers_pattern": null,
|
| 18 |
+
"layers_to_transform": null,
|
| 19 |
+
"loftq_config": {},
|
| 20 |
+
"lora_alpha": 32,
|
| 21 |
+
"lora_bias": false,
|
| 22 |
+
"lora_dropout": 0.05,
|
| 23 |
+
"lora_ga_config": null,
|
| 24 |
+
"megatron_config": null,
|
| 25 |
+
"megatron_core": "megatron.core",
|
| 26 |
+
"modules_to_save": null,
|
| 27 |
+
"monteclora_config": null,
|
| 28 |
+
"peft_type": "LORA",
|
| 29 |
+
"peft_version": "0.21.0",
|
| 30 |
+
"qalora_group_size": 16,
|
| 31 |
+
"r": 16,
|
| 32 |
+
"rank_pattern": {},
|
| 33 |
+
"revision": null,
|
| 34 |
+
"target_modules": [
|
| 35 |
+
"up_proj",
|
| 36 |
+
"v_proj",
|
| 37 |
+
"down_proj",
|
| 38 |
+
"k_proj",
|
| 39 |
+
"q_proj",
|
| 40 |
+
"o_proj",
|
| 41 |
+
"gate_proj"
|
| 42 |
+
],
|
| 43 |
+
"target_parameters": null,
|
| 44 |
+
"task_type": "CAUSAL_LM",
|
| 45 |
+
"trainable_token_indices": null,
|
| 46 |
+
"use_bdlora": null,
|
| 47 |
+
"use_dora": false,
|
| 48 |
+
"use_qalora": false,
|
| 49 |
+
"use_rslora": false,
|
| 50 |
+
"velora_config": null
|
| 51 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:46e44fbe7d7144ce7bdc80c6abb4cf9ee1cf73b6bfc53f919c82db3a625b60a3
|
| 3 |
+
size 161533192
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 4 |
+
{{- messages[0]['content'] }}
|
| 5 |
+
{%- else %}
|
| 6 |
+
{{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
|
| 7 |
+
{%- endif %}
|
| 8 |
+
{{- "\n\n# 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>" }}
|
| 9 |
+
{%- for tool in tools %}
|
| 10 |
+
{{- "\n" }}
|
| 11 |
+
{{- tool | tojson }}
|
| 12 |
+
{%- endfor %}
|
| 13 |
+
{{- "\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" }}
|
| 14 |
+
{%- else %}
|
| 15 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 16 |
+
{{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
|
| 17 |
+
{%- else %}
|
| 18 |
+
{{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
|
| 19 |
+
{%- endif %}
|
| 20 |
+
{%- endif %}
|
| 21 |
+
{%- for message in messages %}
|
| 22 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
|
| 23 |
+
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
|
| 24 |
+
{%- elif message.role == "assistant" %}
|
| 25 |
+
{{- '<|im_start|>' + message.role }}
|
| 26 |
+
{%- if message.content %}
|
| 27 |
+
{{- '\n' + message.content }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{%- for tool_call in message.tool_calls %}
|
| 30 |
+
{%- if tool_call.function is defined %}
|
| 31 |
+
{%- set tool_call = tool_call.function %}
|
| 32 |
+
{%- endif %}
|
| 33 |
+
{{- '\n<tool_call>\n{"name": "' }}
|
| 34 |
+
{{- tool_call.name }}
|
| 35 |
+
{{- '", "arguments": ' }}
|
| 36 |
+
{{- tool_call.arguments | tojson }}
|
| 37 |
+
{{- '}\n</tool_call>' }}
|
| 38 |
+
{%- endfor %}
|
| 39 |
+
{{- '<|im_end|>\n' }}
|
| 40 |
+
{%- elif message.role == "tool" %}
|
| 41 |
+
{%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
|
| 42 |
+
{{- '<|im_start|>user' }}
|
| 43 |
+
{%- endif %}
|
| 44 |
+
{{- '\n<tool_response>\n' }}
|
| 45 |
+
{{- message.content }}
|
| 46 |
+
{{- '\n</tool_response>' }}
|
| 47 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 48 |
+
{{- '<|im_end|>\n' }}
|
| 49 |
+
{%- endif %}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{%- if add_generation_prompt %}
|
| 53 |
+
{{- '<|im_start|>assistant\n' }}
|
| 54 |
+
{%- endif %}
|
run_summary.json
ADDED
|
@@ -0,0 +1,1942 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"base_model": "Qwen/Qwen2.5-Coder-7B-Instruct",
|
| 3 |
+
"rows_train": 1441,
|
| 4 |
+
"rows_val": 95,
|
| 5 |
+
"steps": 273,
|
| 6 |
+
"train_loss": 0.01574120909053851,
|
| 7 |
+
"seconds": 9101.8,
|
| 8 |
+
"peak_gpu_memory_gb": 17.54,
|
| 9 |
+
"lora_r": 16,
|
| 10 |
+
"lora_alpha": 32,
|
| 11 |
+
"target_modules": [
|
| 12 |
+
"q_proj",
|
| 13 |
+
"k_proj",
|
| 14 |
+
"v_proj",
|
| 15 |
+
"o_proj",
|
| 16 |
+
"gate_proj",
|
| 17 |
+
"up_proj",
|
| 18 |
+
"down_proj"
|
| 19 |
+
],
|
| 20 |
+
"log_history": [
|
| 21 |
+
{
|
| 22 |
+
"loss": 0.362233966588974,
|
| 23 |
+
"grad_norm": 0.35578352212905884,
|
| 24 |
+
"learning_rate": 0.0,
|
| 25 |
+
"epoch": 0.011103400416377515,
|
| 26 |
+
"step": 1
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"loss": 0.21971727907657623,
|
| 30 |
+
"grad_norm": 0.2704537808895111,
|
| 31 |
+
"learning_rate": 6.666666666666667e-05,
|
| 32 |
+
"epoch": 0.02220680083275503,
|
| 33 |
+
"step": 2
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"loss": 0.2119903415441513,
|
| 37 |
+
"grad_norm": 0.23446440696716309,
|
| 38 |
+
"learning_rate": 0.00013333333333333334,
|
| 39 |
+
"epoch": 0.033310201249132546,
|
| 40 |
+
"step": 3
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"loss": 0.3059872090816498,
|
| 44 |
+
"grad_norm": 0.2956233620643616,
|
| 45 |
+
"learning_rate": 0.0002,
|
| 46 |
+
"epoch": 0.04441360166551006,
|
| 47 |
+
"step": 4
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"loss": 0.2478485405445099,
|
| 51 |
+
"grad_norm": 0.30406656861305237,
|
| 52 |
+
"learning_rate": 0.00019999323080037624,
|
| 53 |
+
"epoch": 0.055517002081887576,
|
| 54 |
+
"step": 5
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"loss": 0.11736467480659485,
|
| 58 |
+
"grad_norm": 0.19965699315071106,
|
| 59 |
+
"learning_rate": 0.00019997292411794618,
|
| 60 |
+
"epoch": 0.06662040249826509,
|
| 61 |
+
"step": 6
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"loss": 0.1486368626356125,
|
| 65 |
+
"grad_norm": 0.23600247502326965,
|
| 66 |
+
"learning_rate": 0.0001999390827019096,
|
| 67 |
+
"epoch": 0.0777238029146426,
|
| 68 |
+
"step": 7
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"loss": 0.09324199706315994,
|
| 72 |
+
"grad_norm": 0.2227136790752411,
|
| 73 |
+
"learning_rate": 0.0001998917111338525,
|
| 74 |
+
"epoch": 0.08882720333102012,
|
| 75 |
+
"step": 8
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"loss": 0.07419925928115845,
|
| 79 |
+
"grad_norm": 0.212725430727005,
|
| 80 |
+
"learning_rate": 0.00019983081582712685,
|
| 81 |
+
"epoch": 0.09993060374739764,
|
| 82 |
+
"step": 9
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"loss": 0.09104721248149872,
|
| 86 |
+
"grad_norm": 0.2687018811702728,
|
| 87 |
+
"learning_rate": 0.00019975640502598244,
|
| 88 |
+
"epoch": 0.11103400416377515,
|
| 89 |
+
"step": 10
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"loss": 0.07975094020366669,
|
| 93 |
+
"grad_norm": 0.2289663851261139,
|
| 94 |
+
"learning_rate": 0.00019966848880445062,
|
| 95 |
+
"epoch": 0.12213740458015267,
|
| 96 |
+
"step": 11
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"loss": 0.07311172783374786,
|
| 100 |
+
"grad_norm": 0.2065844088792801,
|
| 101 |
+
"learning_rate": 0.00019956707906498044,
|
| 102 |
+
"epoch": 0.13324080499653018,
|
| 103 |
+
"step": 12
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"loss": 0.06214028224349022,
|
| 107 |
+
"grad_norm": 0.16729624569416046,
|
| 108 |
+
"learning_rate": 0.00019945218953682734,
|
| 109 |
+
"epoch": 0.1443442054129077,
|
| 110 |
+
"step": 13
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"loss": 0.039421241730451584,
|
| 114 |
+
"grad_norm": 0.14760349690914154,
|
| 115 |
+
"learning_rate": 0.00019932383577419432,
|
| 116 |
+
"epoch": 0.1554476058292852,
|
| 117 |
+
"step": 14
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"loss": 0.04088294133543968,
|
| 121 |
+
"grad_norm": 0.1884099245071411,
|
| 122 |
+
"learning_rate": 0.00019918203515412617,
|
| 123 |
+
"epoch": 0.16655100624566274,
|
| 124 |
+
"step": 15
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"loss": 0.03639013320207596,
|
| 128 |
+
"grad_norm": 0.14482857286930084,
|
| 129 |
+
"learning_rate": 0.00019902680687415705,
|
| 130 |
+
"epoch": 0.17765440666204024,
|
| 131 |
+
"step": 16
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"loss": 0.05018017068505287,
|
| 135 |
+
"grad_norm": 0.20465002954006195,
|
| 136 |
+
"learning_rate": 0.00019885817194971117,
|
| 137 |
+
"epoch": 0.18875780707841777,
|
| 138 |
+
"step": 17
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"loss": 0.09407485276460648,
|
| 142 |
+
"grad_norm": 0.208365797996521,
|
| 143 |
+
"learning_rate": 0.00019867615321125795,
|
| 144 |
+
"epoch": 0.19986120749479527,
|
| 145 |
+
"step": 18
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"loss": 0.0462566502392292,
|
| 149 |
+
"grad_norm": 0.2098541110754013,
|
| 150 |
+
"learning_rate": 0.00019848077530122083,
|
| 151 |
+
"epoch": 0.2109646079111728,
|
| 152 |
+
"step": 19
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"loss": 0.03283077850937843,
|
| 156 |
+
"grad_norm": 0.16920238733291626,
|
| 157 |
+
"learning_rate": 0.00019827206467064133,
|
| 158 |
+
"epoch": 0.2220680083275503,
|
| 159 |
+
"step": 20
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"loss": 0.031955886632204056,
|
| 163 |
+
"grad_norm": 0.12768959999084473,
|
| 164 |
+
"learning_rate": 0.00019805004957559793,
|
| 165 |
+
"epoch": 0.23317140874392783,
|
| 166 |
+
"step": 21
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"loss": 0.034089263528585434,
|
| 170 |
+
"grad_norm": 0.13755914568901062,
|
| 171 |
+
"learning_rate": 0.00019781476007338058,
|
| 172 |
+
"epoch": 0.24427480916030533,
|
| 173 |
+
"step": 22
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"loss": 0.02859678864479065,
|
| 177 |
+
"grad_norm": 0.1348200887441635,
|
| 178 |
+
"learning_rate": 0.00019756622801842143,
|
| 179 |
+
"epoch": 0.25537820957668284,
|
| 180 |
+
"step": 23
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"loss": 0.02967454306781292,
|
| 184 |
+
"grad_norm": 0.17000287771224976,
|
| 185 |
+
"learning_rate": 0.00019730448705798239,
|
| 186 |
+
"epoch": 0.26648160999306036,
|
| 187 |
+
"step": 24
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"loss": 0.03603748977184296,
|
| 191 |
+
"grad_norm": 0.1551555097103119,
|
| 192 |
+
"learning_rate": 0.00019702957262759965,
|
| 193 |
+
"epoch": 0.2775850104094379,
|
| 194 |
+
"step": 25
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"loss": 0.06082950532436371,
|
| 198 |
+
"grad_norm": 0.24252957105636597,
|
| 199 |
+
"learning_rate": 0.00019674152194628638,
|
| 200 |
+
"epoch": 0.2886884108258154,
|
| 201 |
+
"step": 26
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"loss": 0.03331135958433151,
|
| 205 |
+
"grad_norm": 0.16082148253917694,
|
| 206 |
+
"learning_rate": 0.0001964403740114939,
|
| 207 |
+
"epoch": 0.2997918112421929,
|
| 208 |
+
"step": 27
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"loss": 0.016830770298838615,
|
| 212 |
+
"grad_norm": 0.11476831883192062,
|
| 213 |
+
"learning_rate": 0.0001961261695938319,
|
| 214 |
+
"epoch": 0.3108952116585704,
|
| 215 |
+
"step": 28
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"loss": 0.020966561511158943,
|
| 219 |
+
"grad_norm": 0.12785278260707855,
|
| 220 |
+
"learning_rate": 0.0001957989512315489,
|
| 221 |
+
"epoch": 0.32199861207494795,
|
| 222 |
+
"step": 29
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"loss": 0.019154518842697144,
|
| 226 |
+
"grad_norm": 0.18446354568004608,
|
| 227 |
+
"learning_rate": 0.0001954587632247732,
|
| 228 |
+
"epoch": 0.3331020124913255,
|
| 229 |
+
"step": 30
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"loss": 0.015241650864481926,
|
| 233 |
+
"grad_norm": 0.12455496937036514,
|
| 234 |
+
"learning_rate": 0.00019510565162951537,
|
| 235 |
+
"epoch": 0.34420541290770296,
|
| 236 |
+
"step": 31
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"loss": 0.02742624841630459,
|
| 240 |
+
"grad_norm": 0.16151973605155945,
|
| 241 |
+
"learning_rate": 0.00019473966425143292,
|
| 242 |
+
"epoch": 0.3553088133240805,
|
| 243 |
+
"step": 32
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"loss": 0.03055701218545437,
|
| 247 |
+
"grad_norm": 0.1681850254535675,
|
| 248 |
+
"learning_rate": 0.00019436085063935835,
|
| 249 |
+
"epoch": 0.366412213740458,
|
| 250 |
+
"step": 33
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"loss": 0.03377750143408775,
|
| 254 |
+
"grad_norm": 0.18554241955280304,
|
| 255 |
+
"learning_rate": 0.00019396926207859084,
|
| 256 |
+
"epoch": 0.37751561415683554,
|
| 257 |
+
"step": 34
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"loss": 0.011026578955352306,
|
| 261 |
+
"grad_norm": 0.08266780525445938,
|
| 262 |
+
"learning_rate": 0.00019356495158395315,
|
| 263 |
+
"epoch": 0.3886190145732131,
|
| 264 |
+
"step": 35
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"loss": 0.011678414419293404,
|
| 268 |
+
"grad_norm": 0.0954771563410759,
|
| 269 |
+
"learning_rate": 0.00019314797389261424,
|
| 270 |
+
"epoch": 0.39972241498959055,
|
| 271 |
+
"step": 36
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"loss": 0.003908986691385508,
|
| 275 |
+
"grad_norm": 0.057436637580394745,
|
| 276 |
+
"learning_rate": 0.00019271838545667876,
|
| 277 |
+
"epoch": 0.4108258154059681,
|
| 278 |
+
"step": 37
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"loss": 0.03971560299396515,
|
| 282 |
+
"grad_norm": 0.1711757332086563,
|
| 283 |
+
"learning_rate": 0.00019227624443554425,
|
| 284 |
+
"epoch": 0.4219292158223456,
|
| 285 |
+
"step": 38
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"loss": 0.026749372482299805,
|
| 289 |
+
"grad_norm": 0.15317557752132416,
|
| 290 |
+
"learning_rate": 0.00019182161068802741,
|
| 291 |
+
"epoch": 0.43303261623872313,
|
| 292 |
+
"step": 39
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"loss": 0.005962643772363663,
|
| 296 |
+
"grad_norm": 0.07149907946586609,
|
| 297 |
+
"learning_rate": 0.0001913545457642601,
|
| 298 |
+
"epoch": 0.4441360166551006,
|
| 299 |
+
"step": 40
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"loss": 0.034220557659864426,
|
| 303 |
+
"grad_norm": 0.13144898414611816,
|
| 304 |
+
"learning_rate": 0.00019087511289735644,
|
| 305 |
+
"epoch": 0.45523941707147814,
|
| 306 |
+
"step": 41
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"loss": 0.013403044082224369,
|
| 310 |
+
"grad_norm": 0.10616762936115265,
|
| 311 |
+
"learning_rate": 0.00019038337699485208,
|
| 312 |
+
"epoch": 0.46634281748785567,
|
| 313 |
+
"step": 42
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"loss": 0.0515558123588562,
|
| 317 |
+
"grad_norm": 0.25350019335746765,
|
| 318 |
+
"learning_rate": 0.0001898794046299167,
|
| 319 |
+
"epoch": 0.4774462179042332,
|
| 320 |
+
"step": 43
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"loss": 0.006211064290255308,
|
| 324 |
+
"grad_norm": 0.07005950063467026,
|
| 325 |
+
"learning_rate": 0.00018936326403234125,
|
| 326 |
+
"epoch": 0.48854961832061067,
|
| 327 |
+
"step": 44
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"loss": 0.03814203664660454,
|
| 331 |
+
"grad_norm": 0.18773649632930756,
|
| 332 |
+
"learning_rate": 0.00018883502507930042,
|
| 333 |
+
"epoch": 0.4996530187369882,
|
| 334 |
+
"step": 45
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"loss": 0.004700696095824242,
|
| 338 |
+
"grad_norm": 0.04619601368904114,
|
| 339 |
+
"learning_rate": 0.00018829475928589271,
|
| 340 |
+
"epoch": 0.5107564191533657,
|
| 341 |
+
"step": 46
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"loss": 0.01920146495103836,
|
| 345 |
+
"grad_norm": 0.21882829070091248,
|
| 346 |
+
"learning_rate": 0.0001877425397954582,
|
| 347 |
+
"epoch": 0.5218598195697433,
|
| 348 |
+
"step": 47
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"loss": 0.015273750759661198,
|
| 352 |
+
"grad_norm": 0.179578959941864,
|
| 353 |
+
"learning_rate": 0.00018717844136967624,
|
| 354 |
+
"epoch": 0.5329632199861207,
|
| 355 |
+
"step": 48
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"loss": 0.02806437574326992,
|
| 359 |
+
"grad_norm": 0.15267197787761688,
|
| 360 |
+
"learning_rate": 0.00018660254037844388,
|
| 361 |
+
"epoch": 0.5440666204024983,
|
| 362 |
+
"step": 49
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"loss": 0.02385113015770912,
|
| 366 |
+
"grad_norm": 0.20542870461940765,
|
| 367 |
+
"learning_rate": 0.00018601491478953657,
|
| 368 |
+
"epoch": 0.5551700208188758,
|
| 369 |
+
"step": 50
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"loss": 0.028099708259105682,
|
| 373 |
+
"grad_norm": 0.1347873955965042,
|
| 374 |
+
"learning_rate": 0.00018541564415805258,
|
| 375 |
+
"epoch": 0.5662734212352533,
|
| 376 |
+
"step": 51
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"loss": 0.019417954608798027,
|
| 380 |
+
"grad_norm": 0.09850087761878967,
|
| 381 |
+
"learning_rate": 0.0001848048096156426,
|
| 382 |
+
"epoch": 0.5773768216516308,
|
| 383 |
+
"step": 52
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"loss": 0.004227581899613142,
|
| 387 |
+
"grad_norm": 0.06576558947563171,
|
| 388 |
+
"learning_rate": 0.00018418249385952575,
|
| 389 |
+
"epoch": 0.5884802220680083,
|
| 390 |
+
"step": 53
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"loss": 0.005577925592660904,
|
| 394 |
+
"grad_norm": 0.06200479716062546,
|
| 395 |
+
"learning_rate": 0.00018354878114129367,
|
| 396 |
+
"epoch": 0.5995836224843858,
|
| 397 |
+
"step": 54
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"loss": 0.049879446625709534,
|
| 401 |
+
"grad_norm": 0.3281773328781128,
|
| 402 |
+
"learning_rate": 0.00018290375725550417,
|
| 403 |
+
"epoch": 0.6106870229007634,
|
| 404 |
+
"step": 55
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"loss": 0.006307754199951887,
|
| 408 |
+
"grad_norm": 0.06183058023452759,
|
| 409 |
+
"learning_rate": 0.00018224750952806624,
|
| 410 |
+
"epoch": 0.6217904233171409,
|
| 411 |
+
"step": 56
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"loss": 0.04659178853034973,
|
| 415 |
+
"grad_norm": 0.20692244172096252,
|
| 416 |
+
"learning_rate": 0.00018158012680441723,
|
| 417 |
+
"epoch": 0.6328938237335184,
|
| 418 |
+
"step": 57
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"loss": 0.022197065874934196,
|
| 422 |
+
"grad_norm": 0.1442933827638626,
|
| 423 |
+
"learning_rate": 0.00018090169943749476,
|
| 424 |
+
"epoch": 0.6439972241498959,
|
| 425 |
+
"step": 58
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"loss": 0.011002053506672382,
|
| 429 |
+
"grad_norm": 0.11611606925725937,
|
| 430 |
+
"learning_rate": 0.0001802123192755044,
|
| 431 |
+
"epoch": 0.6551006245662734,
|
| 432 |
+
"step": 59
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"loss": 0.005552457645535469,
|
| 436 |
+
"grad_norm": 0.06183777004480362,
|
| 437 |
+
"learning_rate": 0.0001795120796494848,
|
| 438 |
+
"epoch": 0.666204024982651,
|
| 439 |
+
"step": 60
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"loss": 0.00588051276281476,
|
| 443 |
+
"grad_norm": 0.07396812736988068,
|
| 444 |
+
"learning_rate": 0.00017880107536067218,
|
| 445 |
+
"epoch": 0.6773074253990284,
|
| 446 |
+
"step": 61
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"loss": 0.005040598101913929,
|
| 450 |
+
"grad_norm": 0.06021937355399132,
|
| 451 |
+
"learning_rate": 0.00017807940266766593,
|
| 452 |
+
"epoch": 0.6884108258154059,
|
| 453 |
+
"step": 62
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"loss": 0.006408208981156349,
|
| 457 |
+
"grad_norm": 0.0630301982164383,
|
| 458 |
+
"learning_rate": 0.0001773471592733964,
|
| 459 |
+
"epoch": 0.6995142262317835,
|
| 460 |
+
"step": 63
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"loss": 0.03109092079102993,
|
| 464 |
+
"grad_norm": 0.1972433626651764,
|
| 465 |
+
"learning_rate": 0.0001766044443118978,
|
| 466 |
+
"epoch": 0.710617626648161,
|
| 467 |
+
"step": 64
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"loss": 0.01442437432706356,
|
| 471 |
+
"grad_norm": 0.2614825665950775,
|
| 472 |
+
"learning_rate": 0.00017585135833488692,
|
| 473 |
+
"epoch": 0.7217210270645386,
|
| 474 |
+
"step": 65
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"loss": 0.02434384636580944,
|
| 478 |
+
"grad_norm": 0.11953062564134598,
|
| 479 |
+
"learning_rate": 0.00017508800329814995,
|
| 480 |
+
"epoch": 0.732824427480916,
|
| 481 |
+
"step": 66
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"loss": 0.003504230175167322,
|
| 485 |
+
"grad_norm": 0.056314583867788315,
|
| 486 |
+
"learning_rate": 0.00017431448254773944,
|
| 487 |
+
"epoch": 0.7439278278972935,
|
| 488 |
+
"step": 67
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"loss": 0.01270079892128706,
|
| 492 |
+
"grad_norm": 0.08628875762224197,
|
| 493 |
+
"learning_rate": 0.0001735309008059829,
|
| 494 |
+
"epoch": 0.7550312283136711,
|
| 495 |
+
"step": 68
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"loss": 0.013141262345016003,
|
| 499 |
+
"grad_norm": 0.09234391897916794,
|
| 500 |
+
"learning_rate": 0.00017273736415730488,
|
| 501 |
+
"epoch": 0.7661346287300486,
|
| 502 |
+
"step": 69
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"loss": 0.021891677752137184,
|
| 506 |
+
"grad_norm": 0.1135958731174469,
|
| 507 |
+
"learning_rate": 0.0001719339800338651,
|
| 508 |
+
"epoch": 0.7772380291464261,
|
| 509 |
+
"step": 70
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"loss": 0.03220357373356819,
|
| 513 |
+
"grad_norm": 0.12099183350801468,
|
| 514 |
+
"learning_rate": 0.00017112085720101373,
|
| 515 |
+
"epoch": 0.7883414295628036,
|
| 516 |
+
"step": 71
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"loss": 0.006556101609021425,
|
| 520 |
+
"grad_norm": 0.1038985624909401,
|
| 521 |
+
"learning_rate": 0.0001702981057425662,
|
| 522 |
+
"epoch": 0.7994448299791811,
|
| 523 |
+
"step": 72
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"loss": 0.0022207903675734997,
|
| 527 |
+
"grad_norm": 0.02890835516154766,
|
| 528 |
+
"learning_rate": 0.00016946583704589973,
|
| 529 |
+
"epoch": 0.8105482303955587,
|
| 530 |
+
"step": 73
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"loss": 0.006729573477059603,
|
| 534 |
+
"grad_norm": 0.07326466590166092,
|
| 535 |
+
"learning_rate": 0.0001686241637868734,
|
| 536 |
+
"epoch": 0.8216516308119362,
|
| 537 |
+
"step": 74
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"loss": 0.009240851737558842,
|
| 541 |
+
"grad_norm": 0.09405697882175446,
|
| 542 |
+
"learning_rate": 0.00016777319991457325,
|
| 543 |
+
"epoch": 0.8327550312283136,
|
| 544 |
+
"step": 75
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"loss": 0.0017095255898311734,
|
| 548 |
+
"grad_norm": 0.018018698319792747,
|
| 549 |
+
"learning_rate": 0.00016691306063588583,
|
| 550 |
+
"epoch": 0.8438584316446912,
|
| 551 |
+
"step": 76
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"loss": 0.003485671477392316,
|
| 555 |
+
"grad_norm": 0.09172848612070084,
|
| 556 |
+
"learning_rate": 0.00016604386239990078,
|
| 557 |
+
"epoch": 0.8549618320610687,
|
| 558 |
+
"step": 77
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"loss": 0.006044326350092888,
|
| 562 |
+
"grad_norm": 0.08144538104534149,
|
| 563 |
+
"learning_rate": 0.00016516572288214552,
|
| 564 |
+
"epoch": 0.8660652324774463,
|
| 565 |
+
"step": 78
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"loss": 0.019570548087358475,
|
| 569 |
+
"grad_norm": 0.12051013112068176,
|
| 570 |
+
"learning_rate": 0.00016427876096865394,
|
| 571 |
+
"epoch": 0.8771686328938237,
|
| 572 |
+
"step": 79
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"loss": 0.029929179698228836,
|
| 576 |
+
"grad_norm": 0.12691369652748108,
|
| 577 |
+
"learning_rate": 0.00016338309673987101,
|
| 578 |
+
"epoch": 0.8882720333102012,
|
| 579 |
+
"step": 80
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"loss": 0.009485170245170593,
|
| 583 |
+
"grad_norm": 0.07689705491065979,
|
| 584 |
+
"learning_rate": 0.000162478851454396,
|
| 585 |
+
"epoch": 0.8993754337265788,
|
| 586 |
+
"step": 81
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"loss": 0.007876642979681492,
|
| 590 |
+
"grad_norm": 0.07503333687782288,
|
| 591 |
+
"learning_rate": 0.0001615661475325658,
|
| 592 |
+
"epoch": 0.9104788341429563,
|
| 593 |
+
"step": 82
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"loss": 0.005961703136563301,
|
| 597 |
+
"grad_norm": 0.05012199282646179,
|
| 598 |
+
"learning_rate": 0.00016064510853988138,
|
| 599 |
+
"epoch": 0.9215822345593337,
|
| 600 |
+
"step": 83
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"loss": 0.020787648856639862,
|
| 604 |
+
"grad_norm": 0.1281776875257492,
|
| 605 |
+
"learning_rate": 0.00015971585917027862,
|
| 606 |
+
"epoch": 0.9326856349757113,
|
| 607 |
+
"step": 84
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"loss": 0.000465249118860811,
|
| 611 |
+
"grad_norm": 0.008060788735747337,
|
| 612 |
+
"learning_rate": 0.00015877852522924732,
|
| 613 |
+
"epoch": 0.9437890353920888,
|
| 614 |
+
"step": 85
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"loss": 0.016680452972650528,
|
| 618 |
+
"grad_norm": 0.08592799305915833,
|
| 619 |
+
"learning_rate": 0.00015783323361679864,
|
| 620 |
+
"epoch": 0.9548924358084664,
|
| 621 |
+
"step": 86
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"loss": 0.009535298682749271,
|
| 625 |
+
"grad_norm": 0.0659300684928894,
|
| 626 |
+
"learning_rate": 0.00015688011231028518,
|
| 627 |
+
"epoch": 0.9659958362248439,
|
| 628 |
+
"step": 87
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"loss": 0.01739584095776081,
|
| 632 |
+
"grad_norm": 0.1285824328660965,
|
| 633 |
+
"learning_rate": 0.0001559192903470747,
|
| 634 |
+
"epoch": 0.9770992366412213,
|
| 635 |
+
"step": 88
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"loss": 0.013622242957353592,
|
| 639 |
+
"grad_norm": 0.1113409548997879,
|
| 640 |
+
"learning_rate": 0.0001549508978070806,
|
| 641 |
+
"epoch": 0.9882026370575989,
|
| 642 |
+
"step": 89
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"loss": 0.012220436707139015,
|
| 646 |
+
"grad_norm": 0.0880318358540535,
|
| 647 |
+
"learning_rate": 0.0001539750657951513,
|
| 648 |
+
"epoch": 0.9993060374739764,
|
| 649 |
+
"step": 90
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"loss": 0.001605391502380371,
|
| 653 |
+
"grad_norm": 0.08987312018871307,
|
| 654 |
+
"learning_rate": 0.0001529919264233205,
|
| 655 |
+
"epoch": 1.0,
|
| 656 |
+
"step": 91
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"loss": 0.005238894373178482,
|
| 660 |
+
"grad_norm": 0.05442668870091438,
|
| 661 |
+
"learning_rate": 0.00015200161279292155,
|
| 662 |
+
"epoch": 1.0111034004163775,
|
| 663 |
+
"step": 92
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"loss": 0.003774596843868494,
|
| 667 |
+
"grad_norm": 0.051331259310245514,
|
| 668 |
+
"learning_rate": 0.00015100425897656753,
|
| 669 |
+
"epoch": 1.022206800832755,
|
| 670 |
+
"step": 93
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"loss": 0.0033705425448715687,
|
| 674 |
+
"grad_norm": 0.13363640010356903,
|
| 675 |
+
"learning_rate": 0.00015000000000000001,
|
| 676 |
+
"epoch": 1.0333102012491326,
|
| 677 |
+
"step": 94
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"loss": 0.004251929931342602,
|
| 681 |
+
"grad_norm": 0.04953952506184578,
|
| 682 |
+
"learning_rate": 0.0001489889718238087,
|
| 683 |
+
"epoch": 1.0444136016655101,
|
| 684 |
+
"step": 95
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"loss": 0.0062233032658696175,
|
| 688 |
+
"grad_norm": 0.08337115496397018,
|
| 689 |
+
"learning_rate": 0.00014797131132502465,
|
| 690 |
+
"epoch": 1.0555170020818876,
|
| 691 |
+
"step": 96
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"loss": 0.00801208708435297,
|
| 695 |
+
"grad_norm": 0.07984863221645355,
|
| 696 |
+
"learning_rate": 0.00014694715627858908,
|
| 697 |
+
"epoch": 1.066620402498265,
|
| 698 |
+
"step": 97
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"loss": 0.002963041188195348,
|
| 702 |
+
"grad_norm": 0.05126333609223366,
|
| 703 |
+
"learning_rate": 0.00014591664533870118,
|
| 704 |
+
"epoch": 1.0777238029146425,
|
| 705 |
+
"step": 98
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"loss": 0.008451717905700207,
|
| 709 |
+
"grad_norm": 0.0980011373758316,
|
| 710 |
+
"learning_rate": 0.00014487991802004623,
|
| 711 |
+
"epoch": 1.0888272033310202,
|
| 712 |
+
"step": 99
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"loss": 0.0033840062096714973,
|
| 716 |
+
"grad_norm": 0.09044548869132996,
|
| 717 |
+
"learning_rate": 0.00014383711467890774,
|
| 718 |
+
"epoch": 1.0999306037473977,
|
| 719 |
+
"step": 100
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"loss": 0.01799582503736019,
|
| 723 |
+
"grad_norm": 0.15540874004364014,
|
| 724 |
+
"learning_rate": 0.00014278837649416544,
|
| 725 |
+
"epoch": 1.1110340041637752,
|
| 726 |
+
"step": 101
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"loss": 0.005631173960864544,
|
| 730 |
+
"grad_norm": 0.19040288031101227,
|
| 731 |
+
"learning_rate": 0.0001417338454481818,
|
| 732 |
+
"epoch": 1.1221374045801527,
|
| 733 |
+
"step": 102
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"loss": 0.006984387058764696,
|
| 737 |
+
"grad_norm": 0.09872876852750778,
|
| 738 |
+
"learning_rate": 0.00014067366430758004,
|
| 739 |
+
"epoch": 1.1332408049965301,
|
| 740 |
+
"step": 103
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"loss": 0.0009191720164380968,
|
| 744 |
+
"grad_norm": 0.015590984374284744,
|
| 745 |
+
"learning_rate": 0.0001396079766039157,
|
| 746 |
+
"epoch": 1.1443442054129078,
|
| 747 |
+
"step": 104
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"loss": 0.0014551086351275444,
|
| 751 |
+
"grad_norm": 0.05302764102816582,
|
| 752 |
+
"learning_rate": 0.00013853692661424484,
|
| 753 |
+
"epoch": 1.1554476058292853,
|
| 754 |
+
"step": 105
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"loss": 0.0014657770516350865,
|
| 758 |
+
"grad_norm": 0.03392955660820007,
|
| 759 |
+
"learning_rate": 0.00013746065934159123,
|
| 760 |
+
"epoch": 1.1665510062456628,
|
| 761 |
+
"step": 106
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"loss": 0.0008271295810118318,
|
| 765 |
+
"grad_norm": 0.039895109832286835,
|
| 766 |
+
"learning_rate": 0.00013637932049531516,
|
| 767 |
+
"epoch": 1.1776544066620402,
|
| 768 |
+
"step": 107
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"loss": 0.0202441718429327,
|
| 772 |
+
"grad_norm": 0.3854893147945404,
|
| 773 |
+
"learning_rate": 0.00013529305647138687,
|
| 774 |
+
"epoch": 1.1887578070784177,
|
| 775 |
+
"step": 108
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"loss": 0.0037672286853194237,
|
| 779 |
+
"grad_norm": 0.09699325263500214,
|
| 780 |
+
"learning_rate": 0.00013420201433256689,
|
| 781 |
+
"epoch": 1.1998612074947952,
|
| 782 |
+
"step": 109
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"loss": 0.0019683365244418383,
|
| 786 |
+
"grad_norm": 0.038851648569107056,
|
| 787 |
+
"learning_rate": 0.0001331063417884958,
|
| 788 |
+
"epoch": 1.2109646079111729,
|
| 789 |
+
"step": 110
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"loss": 0.003456352511420846,
|
| 793 |
+
"grad_norm": 0.059991780668497086,
|
| 794 |
+
"learning_rate": 0.00013200618717569714,
|
| 795 |
+
"epoch": 1.2220680083275504,
|
| 796 |
+
"step": 111
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"loss": 0.010525468736886978,
|
| 800 |
+
"grad_norm": 0.0657477006316185,
|
| 801 |
+
"learning_rate": 0.00013090169943749476,
|
| 802 |
+
"epoch": 1.2331714087439278,
|
| 803 |
+
"step": 112
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"loss": 0.0004884903901256621,
|
| 807 |
+
"grad_norm": 0.011057283729314804,
|
| 808 |
+
"learning_rate": 0.0001297930281038482,
|
| 809 |
+
"epoch": 1.2442748091603053,
|
| 810 |
+
"step": 113
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"loss": 0.001567146391607821,
|
| 814 |
+
"grad_norm": 0.02416708879172802,
|
| 815 |
+
"learning_rate": 0.00012868032327110904,
|
| 816 |
+
"epoch": 1.2553782095766828,
|
| 817 |
+
"step": 114
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"loss": 0.0027189922984689474,
|
| 821 |
+
"grad_norm": 0.027829235419631004,
|
| 822 |
+
"learning_rate": 0.0001275637355816999,
|
| 823 |
+
"epoch": 1.2664816099930603,
|
| 824 |
+
"step": 115
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"loss": 0.0011546590831130743,
|
| 828 |
+
"grad_norm": 0.020313648506999016,
|
| 829 |
+
"learning_rate": 0.00012644341620372023,
|
| 830 |
+
"epoch": 1.277585010409438,
|
| 831 |
+
"step": 116
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"loss": 0.008269840851426125,
|
| 835 |
+
"grad_norm": 0.06646428257226944,
|
| 836 |
+
"learning_rate": 0.0001253195168104802,
|
| 837 |
+
"epoch": 1.2886884108258154,
|
| 838 |
+
"step": 117
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"loss": 0.006803714670240879,
|
| 842 |
+
"grad_norm": 0.04964790865778923,
|
| 843 |
+
"learning_rate": 0.00012419218955996676,
|
| 844 |
+
"epoch": 1.299791811242193,
|
| 845 |
+
"step": 118
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"loss": 0.000572250341065228,
|
| 849 |
+
"grad_norm": 0.012129321694374084,
|
| 850 |
+
"learning_rate": 0.00012306158707424403,
|
| 851 |
+
"epoch": 1.3108952116585704,
|
| 852 |
+
"step": 119
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"loss": 0.0030361602548509836,
|
| 856 |
+
"grad_norm": 0.0791798084974289,
|
| 857 |
+
"learning_rate": 0.00012192786241879033,
|
| 858 |
+
"epoch": 1.3219986120749478,
|
| 859 |
+
"step": 120
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"loss": 0.0032888527493923903,
|
| 863 |
+
"grad_norm": 0.04054338485002518,
|
| 864 |
+
"learning_rate": 0.00012079116908177593,
|
| 865 |
+
"epoch": 1.3331020124913255,
|
| 866 |
+
"step": 121
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"loss": 0.004093429539352655,
|
| 870 |
+
"grad_norm": 0.044847872108221054,
|
| 871 |
+
"learning_rate": 0.00011965166095328301,
|
| 872 |
+
"epoch": 1.344205412907703,
|
| 873 |
+
"step": 122
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"loss": 0.0018080237787216902,
|
| 877 |
+
"grad_norm": 0.030936224386096,
|
| 878 |
+
"learning_rate": 0.00011850949230447145,
|
| 879 |
+
"epoch": 1.3553088133240805,
|
| 880 |
+
"step": 123
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"loss": 0.0028136586770415306,
|
| 884 |
+
"grad_norm": 0.036087218672037125,
|
| 885 |
+
"learning_rate": 0.00011736481776669306,
|
| 886 |
+
"epoch": 1.366412213740458,
|
| 887 |
+
"step": 124
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"loss": 0.00630409549921751,
|
| 891 |
+
"grad_norm": 0.09387701749801636,
|
| 892 |
+
"learning_rate": 0.00011621779231055676,
|
| 893 |
+
"epoch": 1.3775156141568354,
|
| 894 |
+
"step": 125
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"loss": 0.0028767576441168785,
|
| 898 |
+
"grad_norm": 0.10042953491210938,
|
| 899 |
+
"learning_rate": 0.00011506857122494831,
|
| 900 |
+
"epoch": 1.3886190145732131,
|
| 901 |
+
"step": 126
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"loss": 0.0023748106323182583,
|
| 905 |
+
"grad_norm": 0.031763847917318344,
|
| 906 |
+
"learning_rate": 0.00011391731009600654,
|
| 907 |
+
"epoch": 1.3997224149895906,
|
| 908 |
+
"step": 127
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"loss": 0.0006115291034802794,
|
| 912 |
+
"grad_norm": 0.013790605589747429,
|
| 913 |
+
"learning_rate": 0.00011276416478605949,
|
| 914 |
+
"epoch": 1.410825815405968,
|
| 915 |
+
"step": 128
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"loss": 0.0006597705651074648,
|
| 919 |
+
"grad_norm": 0.023788312450051308,
|
| 920 |
+
"learning_rate": 0.00011160929141252303,
|
| 921 |
+
"epoch": 1.4219292158223455,
|
| 922 |
+
"step": 129
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"loss": 0.0027555343694984913,
|
| 926 |
+
"grad_norm": 0.049149271100759506,
|
| 927 |
+
"learning_rate": 0.00011045284632676536,
|
| 928 |
+
"epoch": 1.433032616238723,
|
| 929 |
+
"step": 130
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"loss": 0.0011079043615609407,
|
| 933 |
+
"grad_norm": 0.017563462257385254,
|
| 934 |
+
"learning_rate": 0.00010929498609293924,
|
| 935 |
+
"epoch": 1.4441360166551007,
|
| 936 |
+
"step": 131
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"loss": 0.00791978556662798,
|
| 940 |
+
"grad_norm": 0.13089224696159363,
|
| 941 |
+
"learning_rate": 0.00010813586746678583,
|
| 942 |
+
"epoch": 1.4552394170714782,
|
| 943 |
+
"step": 132
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"loss": 0.0012466092593967915,
|
| 947 |
+
"grad_norm": 0.03258095309138298,
|
| 948 |
+
"learning_rate": 0.00010697564737441252,
|
| 949 |
+
"epoch": 1.4663428174878557,
|
| 950 |
+
"step": 133
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"loss": 0.0012967974180355668,
|
| 954 |
+
"grad_norm": 0.03307458758354187,
|
| 955 |
+
"learning_rate": 0.00010581448289104758,
|
| 956 |
+
"epoch": 1.4774462179042331,
|
| 957 |
+
"step": 134
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"loss": 0.007309555076062679,
|
| 961 |
+
"grad_norm": 0.08119885623455048,
|
| 962 |
+
"learning_rate": 0.0001046525312197747,
|
| 963 |
+
"epoch": 1.4885496183206106,
|
| 964 |
+
"step": 135
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"loss": 0.0024388134479522705,
|
| 968 |
+
"grad_norm": 0.0365670882165432,
|
| 969 |
+
"learning_rate": 0.00010348994967025012,
|
| 970 |
+
"epoch": 1.4996530187369883,
|
| 971 |
+
"step": 136
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"loss": 0.009372331202030182,
|
| 975 |
+
"grad_norm": 0.11805802583694458,
|
| 976 |
+
"learning_rate": 0.00010232689563740563,
|
| 977 |
+
"epoch": 1.5107564191533656,
|
| 978 |
+
"step": 137
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"loss": 0.0108938692137599,
|
| 982 |
+
"grad_norm": 0.08489461988210678,
|
| 983 |
+
"learning_rate": 0.00010116352658013973,
|
| 984 |
+
"epoch": 1.5218598195697433,
|
| 985 |
+
"step": 138
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"loss": 0.0026753286365419626,
|
| 989 |
+
"grad_norm": 0.057914867997169495,
|
| 990 |
+
"learning_rate": 0.0001,
|
| 991 |
+
"epoch": 1.5329632199861207,
|
| 992 |
+
"step": 139
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"loss": 0.004938989877700806,
|
| 996 |
+
"grad_norm": 0.0661998838186264,
|
| 997 |
+
"learning_rate": 9.883647341986032e-05,
|
| 998 |
+
"epoch": 1.5440666204024982,
|
| 999 |
+
"step": 140
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"loss": 0.0006187134422361851,
|
| 1003 |
+
"grad_norm": 0.012344218790531158,
|
| 1004 |
+
"learning_rate": 9.767310436259438e-05,
|
| 1005 |
+
"epoch": 1.555170020818876,
|
| 1006 |
+
"step": 141
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"loss": 0.0019832479301840067,
|
| 1010 |
+
"grad_norm": 0.07194311171770096,
|
| 1011 |
+
"learning_rate": 9.651005032974994e-05,
|
| 1012 |
+
"epoch": 1.5662734212352531,
|
| 1013 |
+
"step": 142
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"loss": 0.023575376719236374,
|
| 1017 |
+
"grad_norm": 0.10246355086565018,
|
| 1018 |
+
"learning_rate": 9.534746878022534e-05,
|
| 1019 |
+
"epoch": 1.5773768216516308,
|
| 1020 |
+
"step": 143
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"loss": 0.0016534080496057868,
|
| 1024 |
+
"grad_norm": 0.027751371264457703,
|
| 1025 |
+
"learning_rate": 9.418551710895243e-05,
|
| 1026 |
+
"epoch": 1.5884802220680083,
|
| 1027 |
+
"step": 144
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"loss": 0.010032778605818748,
|
| 1031 |
+
"grad_norm": 0.08701073378324509,
|
| 1032 |
+
"learning_rate": 9.302435262558747e-05,
|
| 1033 |
+
"epoch": 1.5995836224843858,
|
| 1034 |
+
"step": 145
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"loss": 0.004825797863304615,
|
| 1038 |
+
"grad_norm": 0.07686074078083038,
|
| 1039 |
+
"learning_rate": 9.186413253321418e-05,
|
| 1040 |
+
"epoch": 1.6106870229007635,
|
| 1041 |
+
"step": 146
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"loss": 0.007475260179489851,
|
| 1045 |
+
"grad_norm": 0.0755486860871315,
|
| 1046 |
+
"learning_rate": 9.070501390706079e-05,
|
| 1047 |
+
"epoch": 1.6217904233171407,
|
| 1048 |
+
"step": 147
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"loss": 0.0029515489004552364,
|
| 1052 |
+
"grad_norm": 0.04138359799981117,
|
| 1053 |
+
"learning_rate": 8.954715367323468e-05,
|
| 1054 |
+
"epoch": 1.6328938237335184,
|
| 1055 |
+
"step": 148
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"loss": 0.02156146615743637,
|
| 1059 |
+
"grad_norm": 0.14557036757469177,
|
| 1060 |
+
"learning_rate": 8.839070858747697e-05,
|
| 1061 |
+
"epoch": 1.643997224149896,
|
| 1062 |
+
"step": 149
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"loss": 0.0013141734525561333,
|
| 1066 |
+
"grad_norm": 0.033466584980487823,
|
| 1067 |
+
"learning_rate": 8.723583521394054e-05,
|
| 1068 |
+
"epoch": 1.6551006245662734,
|
| 1069 |
+
"step": 150
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"loss": 0.009912682697176933,
|
| 1073 |
+
"grad_norm": 0.10021013021469116,
|
| 1074 |
+
"learning_rate": 8.608268990399349e-05,
|
| 1075 |
+
"epoch": 1.666204024982651,
|
| 1076 |
+
"step": 151
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"loss": 0.002557775005698204,
|
| 1080 |
+
"grad_norm": 0.06423213332891464,
|
| 1081 |
+
"learning_rate": 8.49314287750517e-05,
|
| 1082 |
+
"epoch": 1.6773074253990283,
|
| 1083 |
+
"step": 152
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"loss": 0.007192865014076233,
|
| 1087 |
+
"grad_norm": 0.05610248073935509,
|
| 1088 |
+
"learning_rate": 8.378220768944327e-05,
|
| 1089 |
+
"epoch": 1.688410825815406,
|
| 1090 |
+
"step": 153
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"loss": 0.00357620557770133,
|
| 1094 |
+
"grad_norm": 0.0698416456580162,
|
| 1095 |
+
"learning_rate": 8.263518223330697e-05,
|
| 1096 |
+
"epoch": 1.6995142262317835,
|
| 1097 |
+
"step": 154
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"loss": 0.001066345488652587,
|
| 1101 |
+
"grad_norm": 0.022768152877688408,
|
| 1102 |
+
"learning_rate": 8.149050769552856e-05,
|
| 1103 |
+
"epoch": 1.710617626648161,
|
| 1104 |
+
"step": 155
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"loss": 0.002087992150336504,
|
| 1108 |
+
"grad_norm": 0.05063783749938011,
|
| 1109 |
+
"learning_rate": 8.034833904671698e-05,
|
| 1110 |
+
"epoch": 1.7217210270645387,
|
| 1111 |
+
"step": 156
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"loss": 0.005414406768977642,
|
| 1115 |
+
"grad_norm": 0.07123545557260513,
|
| 1116 |
+
"learning_rate": 7.920883091822408e-05,
|
| 1117 |
+
"epoch": 1.732824427480916,
|
| 1118 |
+
"step": 157
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"loss": 0.0005238516023382545,
|
| 1122 |
+
"grad_norm": 0.013095811009407043,
|
| 1123 |
+
"learning_rate": 7.807213758120966e-05,
|
| 1124 |
+
"epoch": 1.7439278278972936,
|
| 1125 |
+
"step": 158
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"loss": 0.0023119491524994373,
|
| 1129 |
+
"grad_norm": 0.02543606422841549,
|
| 1130 |
+
"learning_rate": 7.693841292575598e-05,
|
| 1131 |
+
"epoch": 1.755031228313671,
|
| 1132 |
+
"step": 159
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"loss": 0.0012464482570067048,
|
| 1136 |
+
"grad_norm": 0.023485898971557617,
|
| 1137 |
+
"learning_rate": 7.580781044003324e-05,
|
| 1138 |
+
"epoch": 1.7661346287300486,
|
| 1139 |
+
"step": 160
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"loss": 0.010121936909854412,
|
| 1143 |
+
"grad_norm": 0.11865287274122238,
|
| 1144 |
+
"learning_rate": 7.468048318951983e-05,
|
| 1145 |
+
"epoch": 1.7772380291464263,
|
| 1146 |
+
"step": 161
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"loss": 0.010622553527355194,
|
| 1150 |
+
"grad_norm": 0.10067789256572723,
|
| 1151 |
+
"learning_rate": 7.35565837962798e-05,
|
| 1152 |
+
"epoch": 1.7883414295628035,
|
| 1153 |
+
"step": 162
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"loss": 0.01093344110995531,
|
| 1157 |
+
"grad_norm": 0.11214911192655563,
|
| 1158 |
+
"learning_rate": 7.243626441830009e-05,
|
| 1159 |
+
"epoch": 1.7994448299791812,
|
| 1160 |
+
"step": 163
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"loss": 0.002597471931949258,
|
| 1164 |
+
"grad_norm": 0.03298800811171532,
|
| 1165 |
+
"learning_rate": 7.131967672889101e-05,
|
| 1166 |
+
"epoch": 1.8105482303955587,
|
| 1167 |
+
"step": 164
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"loss": 0.0004867906682193279,
|
| 1171 |
+
"grad_norm": 0.009381928481161594,
|
| 1172 |
+
"learning_rate": 7.02069718961518e-05,
|
| 1173 |
+
"epoch": 1.8216516308119362,
|
| 1174 |
+
"step": 165
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"loss": 0.0009496217826381326,
|
| 1178 |
+
"grad_norm": 0.03668565675616264,
|
| 1179 |
+
"learning_rate": 6.909830056250527e-05,
|
| 1180 |
+
"epoch": 1.8327550312283136,
|
| 1181 |
+
"step": 166
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"loss": 0.005561809055507183,
|
| 1185 |
+
"grad_norm": 0.06403116881847382,
|
| 1186 |
+
"learning_rate": 6.799381282430284e-05,
|
| 1187 |
+
"epoch": 1.843858431644691,
|
| 1188 |
+
"step": 167
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"loss": 0.00022523854568134993,
|
| 1192 |
+
"grad_norm": 0.004085203632712364,
|
| 1193 |
+
"learning_rate": 6.68936582115042e-05,
|
| 1194 |
+
"epoch": 1.8549618320610688,
|
| 1195 |
+
"step": 168
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"loss": 0.0006040821899659932,
|
| 1199 |
+
"grad_norm": 0.016563741490244865,
|
| 1200 |
+
"learning_rate": 6.579798566743314e-05,
|
| 1201 |
+
"epoch": 1.8660652324774463,
|
| 1202 |
+
"step": 169
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"loss": 0.006220725830644369,
|
| 1206 |
+
"grad_norm": 0.09190915524959564,
|
| 1207 |
+
"learning_rate": 6.470694352861312e-05,
|
| 1208 |
+
"epoch": 1.8771686328938237,
|
| 1209 |
+
"step": 170
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"loss": 0.0005163773312233388,
|
| 1213 |
+
"grad_norm": 0.012748060747981071,
|
| 1214 |
+
"learning_rate": 6.362067950468489e-05,
|
| 1215 |
+
"epoch": 1.8882720333102012,
|
| 1216 |
+
"step": 171
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"loss": 0.009463150054216385,
|
| 1220 |
+
"grad_norm": 0.08283476531505585,
|
| 1221 |
+
"learning_rate": 6.25393406584088e-05,
|
| 1222 |
+
"epoch": 1.8993754337265787,
|
| 1223 |
+
"step": 172
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"loss": 0.005616231355816126,
|
| 1227 |
+
"grad_norm": 0.10440317541360855,
|
| 1228 |
+
"learning_rate": 6.146307338575519e-05,
|
| 1229 |
+
"epoch": 1.9104788341429564,
|
| 1230 |
+
"step": 173
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"loss": 0.0025631391908973455,
|
| 1234 |
+
"grad_norm": 0.03647289052605629,
|
| 1235 |
+
"learning_rate": 6.039202339608432e-05,
|
| 1236 |
+
"epoch": 1.9215822345593336,
|
| 1237 |
+
"step": 174
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"loss": 0.0007256513345055282,
|
| 1241 |
+
"grad_norm": 0.018309885635972023,
|
| 1242 |
+
"learning_rate": 5.9326335692419995e-05,
|
| 1243 |
+
"epoch": 1.9326856349757113,
|
| 1244 |
+
"step": 175
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"loss": 0.006048131268471479,
|
| 1248 |
+
"grad_norm": 0.08872969448566437,
|
| 1249 |
+
"learning_rate": 5.8266154551818216e-05,
|
| 1250 |
+
"epoch": 1.9437890353920888,
|
| 1251 |
+
"step": 176
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"loss": 0.0048399316146969795,
|
| 1255 |
+
"grad_norm": 0.0693540945649147,
|
| 1256 |
+
"learning_rate": 5.72116235058346e-05,
|
| 1257 |
+
"epoch": 1.9548924358084663,
|
| 1258 |
+
"step": 177
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"loss": 0.00079789396841079,
|
| 1262 |
+
"grad_norm": 0.022740449756383896,
|
| 1263 |
+
"learning_rate": 5.616288532109225e-05,
|
| 1264 |
+
"epoch": 1.965995836224844,
|
| 1265 |
+
"step": 178
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"loss": 0.0002554625098127872,
|
| 1269 |
+
"grad_norm": 0.004780248738825321,
|
| 1270 |
+
"learning_rate": 5.5120081979953785e-05,
|
| 1271 |
+
"epoch": 1.9770992366412212,
|
| 1272 |
+
"step": 179
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"loss": 0.004320482723414898,
|
| 1276 |
+
"grad_norm": 0.051174018532037735,
|
| 1277 |
+
"learning_rate": 5.4083354661298814e-05,
|
| 1278 |
+
"epoch": 1.988202637057599,
|
| 1279 |
+
"step": 180
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"loss": 0.00403692526742816,
|
| 1283 |
+
"grad_norm": 0.04460834339261055,
|
| 1284 |
+
"learning_rate": 5.305284372141095e-05,
|
| 1285 |
+
"epoch": 1.9993060374739764,
|
| 1286 |
+
"step": 181
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"loss": 0.000192473380593583,
|
| 1290 |
+
"grad_norm": 0.02723965048789978,
|
| 1291 |
+
"learning_rate": 5.2028688674975415e-05,
|
| 1292 |
+
"epoch": 2.0,
|
| 1293 |
+
"step": 182
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"loss": 0.0031282880809158087,
|
| 1297 |
+
"grad_norm": 0.07167559117078781,
|
| 1298 |
+
"learning_rate": 5.101102817619131e-05,
|
| 1299 |
+
"epoch": 2.0111034004163777,
|
| 1300 |
+
"step": 183
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"loss": 0.001440710504539311,
|
| 1304 |
+
"grad_norm": 0.034675925970077515,
|
| 1305 |
+
"learning_rate": 5.000000000000002e-05,
|
| 1306 |
+
"epoch": 2.022206800832755,
|
| 1307 |
+
"step": 184
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"loss": 0.002487765159457922,
|
| 1311 |
+
"grad_norm": 0.04422704502940178,
|
| 1312 |
+
"learning_rate": 4.899574102343247e-05,
|
| 1313 |
+
"epoch": 2.0333102012491326,
|
| 1314 |
+
"step": 185
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"loss": 0.00018012276268564165,
|
| 1318 |
+
"grad_norm": 0.003625614335760474,
|
| 1319 |
+
"learning_rate": 4.799838720707846e-05,
|
| 1320 |
+
"epoch": 2.04441360166551,
|
| 1321 |
+
"step": 186
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"loss": 0.0008509792387485504,
|
| 1325 |
+
"grad_norm": 0.0641360729932785,
|
| 1326 |
+
"learning_rate": 4.700807357667952e-05,
|
| 1327 |
+
"epoch": 2.0555170020818876,
|
| 1328 |
+
"step": 187
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"loss": 0.0012335458304733038,
|
| 1332 |
+
"grad_norm": 0.03804023936390877,
|
| 1333 |
+
"learning_rate": 4.6024934204848745e-05,
|
| 1334 |
+
"epoch": 2.0666204024982653,
|
| 1335 |
+
"step": 188
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"loss": 0.0007197751547209918,
|
| 1339 |
+
"grad_norm": 0.01487179659307003,
|
| 1340 |
+
"learning_rate": 4.50491021929194e-05,
|
| 1341 |
+
"epoch": 2.0777238029146425,
|
| 1342 |
+
"step": 189
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"loss": 0.000570658827200532,
|
| 1346 |
+
"grad_norm": 0.012829587794840336,
|
| 1347 |
+
"learning_rate": 4.4080709652925336e-05,
|
| 1348 |
+
"epoch": 2.0888272033310202,
|
| 1349 |
+
"step": 190
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"loss": 0.0003685185220092535,
|
| 1353 |
+
"grad_norm": 0.00644276337698102,
|
| 1354 |
+
"learning_rate": 4.3119887689714844e-05,
|
| 1355 |
+
"epoch": 2.0999306037473975,
|
| 1356 |
+
"step": 191
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"loss": 0.00217076507396996,
|
| 1360 |
+
"grad_norm": 0.03593655303120613,
|
| 1361 |
+
"learning_rate": 4.216676638320135e-05,
|
| 1362 |
+
"epoch": 2.111034004163775,
|
| 1363 |
+
"step": 192
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"loss": 0.0021631934214383364,
|
| 1367 |
+
"grad_norm": 0.038568392395973206,
|
| 1368 |
+
"learning_rate": 4.12214747707527e-05,
|
| 1369 |
+
"epoch": 2.122137404580153,
|
| 1370 |
+
"step": 193
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"loss": 0.00018508080393075943,
|
| 1374 |
+
"grad_norm": 0.004468716215342283,
|
| 1375 |
+
"learning_rate": 4.028414082972141e-05,
|
| 1376 |
+
"epoch": 2.13324080499653,
|
| 1377 |
+
"step": 194
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"loss": 0.0009128937381319702,
|
| 1381 |
+
"grad_norm": 0.027085695415735245,
|
| 1382 |
+
"learning_rate": 3.935489146011869e-05,
|
| 1383 |
+
"epoch": 2.144344205412908,
|
| 1384 |
+
"step": 195
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"loss": 0.0002966481843031943,
|
| 1388 |
+
"grad_norm": 0.006275610998272896,
|
| 1389 |
+
"learning_rate": 3.843385246743417e-05,
|
| 1390 |
+
"epoch": 2.155447605829285,
|
| 1391 |
+
"step": 196
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"loss": 0.0011582208098843694,
|
| 1395 |
+
"grad_norm": 0.02687346190214157,
|
| 1396 |
+
"learning_rate": 3.7521148545604e-05,
|
| 1397 |
+
"epoch": 2.1665510062456628,
|
| 1398 |
+
"step": 197
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"loss": 0.0008262763731181622,
|
| 1402 |
+
"grad_norm": 0.014363622292876244,
|
| 1403 |
+
"learning_rate": 3.661690326012897e-05,
|
| 1404 |
+
"epoch": 2.1776544066620405,
|
| 1405 |
+
"step": 198
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"loss": 0.0003042164898943156,
|
| 1409 |
+
"grad_norm": 0.01361521054059267,
|
| 1410 |
+
"learning_rate": 3.5721239031346066e-05,
|
| 1411 |
+
"epoch": 2.1887578070784177,
|
| 1412 |
+
"step": 199
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"loss": 0.00021733059838879853,
|
| 1416 |
+
"grad_norm": 0.00866472814232111,
|
| 1417 |
+
"learning_rate": 3.483427711785449e-05,
|
| 1418 |
+
"epoch": 2.1998612074947954,
|
| 1419 |
+
"step": 200
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"loss": 0.00023889444128144532,
|
| 1423 |
+
"grad_norm": 0.00470183277502656,
|
| 1424 |
+
"learning_rate": 3.395613760009925e-05,
|
| 1425 |
+
"epoch": 2.2109646079111727,
|
| 1426 |
+
"step": 201
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"loss": 0.0022749893832951784,
|
| 1430 |
+
"grad_norm": 0.03821490705013275,
|
| 1431 |
+
"learning_rate": 3.308693936411421e-05,
|
| 1432 |
+
"epoch": 2.2220680083275504,
|
| 1433 |
+
"step": 202
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"loss": 0.00029496397473849356,
|
| 1437 |
+
"grad_norm": 0.007405790500342846,
|
| 1438 |
+
"learning_rate": 3.222680008542678e-05,
|
| 1439 |
+
"epoch": 2.233171408743928,
|
| 1440 |
+
"step": 203
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"loss": 0.00033837920636869967,
|
| 1444 |
+
"grad_norm": 0.008731233887374401,
|
| 1445 |
+
"learning_rate": 3.137583621312665e-05,
|
| 1446 |
+
"epoch": 2.2442748091603053,
|
| 1447 |
+
"step": 204
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"loss": 0.0007719028508290648,
|
| 1451 |
+
"grad_norm": 0.014901618473231792,
|
| 1452 |
+
"learning_rate": 3.053416295410026e-05,
|
| 1453 |
+
"epoch": 2.255378209576683,
|
| 1454 |
+
"step": 205
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"loss": 0.0014813048765063286,
|
| 1458 |
+
"grad_norm": 0.04358435794711113,
|
| 1459 |
+
"learning_rate": 2.9701894257433826e-05,
|
| 1460 |
+
"epoch": 2.2664816099930603,
|
| 1461 |
+
"step": 206
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"loss": 0.0005430534947663546,
|
| 1465 |
+
"grad_norm": 0.011528179049491882,
|
| 1466 |
+
"learning_rate": 2.8879142798986292e-05,
|
| 1467 |
+
"epoch": 2.277585010409438,
|
| 1468 |
+
"step": 207
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"loss": 0.00034370424691587687,
|
| 1472 |
+
"grad_norm": 0.004893212113529444,
|
| 1473 |
+
"learning_rate": 2.8066019966134904e-05,
|
| 1474 |
+
"epoch": 2.2886884108258156,
|
| 1475 |
+
"step": 208
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"loss": 0.0010527092963457108,
|
| 1479 |
+
"grad_norm": 0.0205328818410635,
|
| 1480 |
+
"learning_rate": 2.7262635842695127e-05,
|
| 1481 |
+
"epoch": 2.299791811242193,
|
| 1482 |
+
"step": 209
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"loss": 0.0008116448880173266,
|
| 1486 |
+
"grad_norm": 0.021442309021949768,
|
| 1487 |
+
"learning_rate": 2.6469099194017143e-05,
|
| 1488 |
+
"epoch": 2.3108952116585706,
|
| 1489 |
+
"step": 210
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"loss": 0.0037238351069390774,
|
| 1493 |
+
"grad_norm": 0.04724681004881859,
|
| 1494 |
+
"learning_rate": 2.5685517452260567e-05,
|
| 1495 |
+
"epoch": 2.321998612074948,
|
| 1496 |
+
"step": 211
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"loss": 0.00570558849722147,
|
| 1500 |
+
"grad_norm": 0.06557858735322952,
|
| 1501 |
+
"learning_rate": 2.491199670185008e-05,
|
| 1502 |
+
"epoch": 2.3331020124913255,
|
| 1503 |
+
"step": 212
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"loss": 0.0011785834794864058,
|
| 1507 |
+
"grad_norm": 0.02593190036714077,
|
| 1508 |
+
"learning_rate": 2.4148641665113113e-05,
|
| 1509 |
+
"epoch": 2.344205412907703,
|
| 1510 |
+
"step": 213
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"loss": 0.0018704122630879283,
|
| 1514 |
+
"grad_norm": 0.05805144086480141,
|
| 1515 |
+
"learning_rate": 2.339555568810221e-05,
|
| 1516 |
+
"epoch": 2.3553088133240805,
|
| 1517 |
+
"step": 214
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"loss": 0.0009465268230997026,
|
| 1521 |
+
"grad_norm": 0.029634978622198105,
|
| 1522 |
+
"learning_rate": 2.265284072660362e-05,
|
| 1523 |
+
"epoch": 2.366412213740458,
|
| 1524 |
+
"step": 215
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"loss": 0.0006902794120833278,
|
| 1528 |
+
"grad_norm": 0.017646944150328636,
|
| 1529 |
+
"learning_rate": 2.192059733233408e-05,
|
| 1530 |
+
"epoch": 2.3775156141568354,
|
| 1531 |
+
"step": 216
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"loss": 0.0008653043187223375,
|
| 1535 |
+
"grad_norm": 0.02128649689257145,
|
| 1536 |
+
"learning_rate": 2.119892463932781e-05,
|
| 1537 |
+
"epoch": 2.388619014573213,
|
| 1538 |
+
"step": 217
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"loss": 0.0009620467899367213,
|
| 1542 |
+
"grad_norm": 0.01664072461426258,
|
| 1543 |
+
"learning_rate": 2.0487920350515212e-05,
|
| 1544 |
+
"epoch": 2.3997224149895904,
|
| 1545 |
+
"step": 218
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"loss": 0.00021245694370009005,
|
| 1549 |
+
"grad_norm": 0.004011219833046198,
|
| 1550 |
+
"learning_rate": 1.9787680724495617e-05,
|
| 1551 |
+
"epoch": 2.410825815405968,
|
| 1552 |
+
"step": 219
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"loss": 0.000662539794575423,
|
| 1556 |
+
"grad_norm": 0.013697715476155281,
|
| 1557 |
+
"learning_rate": 1.9098300562505266e-05,
|
| 1558 |
+
"epoch": 2.4219292158223458,
|
| 1559 |
+
"step": 220
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"loss": 0.0013410050887614489,
|
| 1563 |
+
"grad_norm": 0.03492460772395134,
|
| 1564 |
+
"learning_rate": 1.8419873195582814e-05,
|
| 1565 |
+
"epoch": 2.433032616238723,
|
| 1566 |
+
"step": 221
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"loss": 0.0006607146933674812,
|
| 1570 |
+
"grad_norm": 0.016314877197146416,
|
| 1571 |
+
"learning_rate": 1.775249047193377e-05,
|
| 1572 |
+
"epoch": 2.4441360166551007,
|
| 1573 |
+
"step": 222
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"loss": 0.00267712096683681,
|
| 1577 |
+
"grad_norm": 0.03245789557695389,
|
| 1578 |
+
"learning_rate": 1.7096242744495837e-05,
|
| 1579 |
+
"epoch": 2.455239417071478,
|
| 1580 |
+
"step": 223
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"loss": 0.0009540443425066769,
|
| 1584 |
+
"grad_norm": 0.01580522023141384,
|
| 1585 |
+
"learning_rate": 1.6451218858706374e-05,
|
| 1586 |
+
"epoch": 2.4663428174878557,
|
| 1587 |
+
"step": 224
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"loss": 0.0002631157694850117,
|
| 1591 |
+
"grad_norm": 0.006987827364355326,
|
| 1592 |
+
"learning_rate": 1.5817506140474247e-05,
|
| 1593 |
+
"epoch": 2.4774462179042334,
|
| 1594 |
+
"step": 225
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"loss": 0.0005279219476506114,
|
| 1598 |
+
"grad_norm": 0.011882264167070389,
|
| 1599 |
+
"learning_rate": 1.5195190384357404e-05,
|
| 1600 |
+
"epoch": 2.4885496183206106,
|
| 1601 |
+
"step": 226
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"loss": 0.0005659055314026773,
|
| 1605 |
+
"grad_norm": 0.01122165285050869,
|
| 1606 |
+
"learning_rate": 1.458435584194745e-05,
|
| 1607 |
+
"epoch": 2.4996530187369883,
|
| 1608 |
+
"step": 227
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"loss": 0.0006534359999932349,
|
| 1612 |
+
"grad_norm": 0.015417223796248436,
|
| 1613 |
+
"learning_rate": 1.3985085210463477e-05,
|
| 1614 |
+
"epoch": 2.5107564191533656,
|
| 1615 |
+
"step": 228
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"loss": 0.000537577667273581,
|
| 1619 |
+
"grad_norm": 0.014984376728534698,
|
| 1620 |
+
"learning_rate": 1.339745962155613e-05,
|
| 1621 |
+
"epoch": 2.5218598195697433,
|
| 1622 |
+
"step": 229
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"loss": 0.0014547649770975113,
|
| 1626 |
+
"grad_norm": 0.04610738530755043,
|
| 1627 |
+
"learning_rate": 1.2821558630323772e-05,
|
| 1628 |
+
"epoch": 2.5329632199861205,
|
| 1629 |
+
"step": 230
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"loss": 0.0030281252693384886,
|
| 1633 |
+
"grad_norm": 0.06278136372566223,
|
| 1634 |
+
"learning_rate": 1.2257460204541794e-05,
|
| 1635 |
+
"epoch": 2.544066620402498,
|
| 1636 |
+
"step": 231
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"loss": 4.744660691358149e-05,
|
| 1640 |
+
"grad_norm": 0.0010382796172052622,
|
| 1641 |
+
"learning_rate": 1.1705240714107302e-05,
|
| 1642 |
+
"epoch": 2.555170020818876,
|
| 1643 |
+
"step": 232
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"loss": 0.0011553519871085882,
|
| 1647 |
+
"grad_norm": 0.03249562904238701,
|
| 1648 |
+
"learning_rate": 1.116497492069961e-05,
|
| 1649 |
+
"epoch": 2.566273421235253,
|
| 1650 |
+
"step": 233
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"loss": 0.0005710968980565667,
|
| 1654 |
+
"grad_norm": 0.040762804448604584,
|
| 1655 |
+
"learning_rate": 1.0636735967658784e-05,
|
| 1656 |
+
"epoch": 2.577376821651631,
|
| 1657 |
+
"step": 234
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"loss": 0.0016874882858246565,
|
| 1661 |
+
"grad_norm": 0.03226013481616974,
|
| 1662 |
+
"learning_rate": 1.0120595370083318e-05,
|
| 1663 |
+
"epoch": 2.588480222068008,
|
| 1664 |
+
"step": 235
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"loss": 0.0004933075397275388,
|
| 1668 |
+
"grad_norm": 0.008815680630505085,
|
| 1669 |
+
"learning_rate": 9.616623005147951e-06,
|
| 1670 |
+
"epoch": 2.599583622484386,
|
| 1671 |
+
"step": 236
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"loss": 0.0004388394590932876,
|
| 1675 |
+
"grad_norm": 0.008908948861062527,
|
| 1676 |
+
"learning_rate": 9.124887102643575e-06,
|
| 1677 |
+
"epoch": 2.6106870229007635,
|
| 1678 |
+
"step": 237
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"loss": 0.00035047775600105524,
|
| 1682 |
+
"grad_norm": 0.008026096038520336,
|
| 1683 |
+
"learning_rate": 8.645454235739903e-06,
|
| 1684 |
+
"epoch": 2.6217904233171407,
|
| 1685 |
+
"step": 238
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"loss": 0.004668767563998699,
|
| 1689 |
+
"grad_norm": 0.09909072518348694,
|
| 1690 |
+
"learning_rate": 8.178389311972612e-06,
|
| 1691 |
+
"epoch": 2.6328938237335184,
|
| 1692 |
+
"step": 239
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"loss": 0.0018479757709428668,
|
| 1696 |
+
"grad_norm": 0.03939471393823624,
|
| 1697 |
+
"learning_rate": 7.72375556445577e-06,
|
| 1698 |
+
"epoch": 2.6439972241498957,
|
| 1699 |
+
"step": 240
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"loss": 0.001170073519460857,
|
| 1703 |
+
"grad_norm": 0.07461482286453247,
|
| 1704 |
+
"learning_rate": 7.281614543321269e-06,
|
| 1705 |
+
"epoch": 2.6551006245662734,
|
| 1706 |
+
"step": 241
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"loss": 0.001040186150930822,
|
| 1710 |
+
"grad_norm": 0.018851246684789658,
|
| 1711 |
+
"learning_rate": 6.852026107385756e-06,
|
| 1712 |
+
"epoch": 2.666204024982651,
|
| 1713 |
+
"step": 242
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"loss": 0.0003154140431433916,
|
| 1717 |
+
"grad_norm": 0.011888038367033005,
|
| 1718 |
+
"learning_rate": 6.435048416046863e-06,
|
| 1719 |
+
"epoch": 2.6773074253990283,
|
| 1720 |
+
"step": 243
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"loss": 0.002783727366477251,
|
| 1724 |
+
"grad_norm": 0.06753791868686676,
|
| 1725 |
+
"learning_rate": 6.030737921409169e-06,
|
| 1726 |
+
"epoch": 2.688410825815406,
|
| 1727 |
+
"step": 244
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"loss": 8.915372018236667e-05,
|
| 1731 |
+
"grad_norm": 0.0017917046789079905,
|
| 1732 |
+
"learning_rate": 5.639149360641649e-06,
|
| 1733 |
+
"epoch": 2.6995142262317833,
|
| 1734 |
+
"step": 245
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"loss": 0.001064301235601306,
|
| 1738 |
+
"grad_norm": 0.025877000764012337,
|
| 1739 |
+
"learning_rate": 5.26033574856708e-06,
|
| 1740 |
+
"epoch": 2.710617626648161,
|
| 1741 |
+
"step": 246
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"loss": 0.00019584805704653263,
|
| 1745 |
+
"grad_norm": 0.013628424145281315,
|
| 1746 |
+
"learning_rate": 4.8943483704846475e-06,
|
| 1747 |
+
"epoch": 2.7217210270645387,
|
| 1748 |
+
"step": 247
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"loss": 0.0002667722583282739,
|
| 1752 |
+
"grad_norm": 0.009531455114483833,
|
| 1753 |
+
"learning_rate": 4.541236775226809e-06,
|
| 1754 |
+
"epoch": 2.732824427480916,
|
| 1755 |
+
"step": 248
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"loss": 0.0014836416812613606,
|
| 1759 |
+
"grad_norm": 0.03923977538943291,
|
| 1760 |
+
"learning_rate": 4.20104876845111e-06,
|
| 1761 |
+
"epoch": 2.7439278278972936,
|
| 1762 |
+
"step": 249
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"loss": 0.0010388277005404234,
|
| 1766 |
+
"grad_norm": 0.029680265113711357,
|
| 1767 |
+
"learning_rate": 3.873830406168111e-06,
|
| 1768 |
+
"epoch": 2.755031228313671,
|
| 1769 |
+
"step": 250
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"loss": 0.000178680318640545,
|
| 1773 |
+
"grad_norm": 0.00518755754455924,
|
| 1774 |
+
"learning_rate": 3.5596259885061102e-06,
|
| 1775 |
+
"epoch": 2.7661346287300486,
|
| 1776 |
+
"step": 251
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"loss": 0.0009329643216915429,
|
| 1780 |
+
"grad_norm": 0.024954449385404587,
|
| 1781 |
+
"learning_rate": 3.2584780537136207e-06,
|
| 1782 |
+
"epoch": 2.7772380291464263,
|
| 1783 |
+
"step": 252
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"loss": 0.0006331746117211878,
|
| 1787 |
+
"grad_norm": 0.012082848697900772,
|
| 1788 |
+
"learning_rate": 2.970427372400353e-06,
|
| 1789 |
+
"epoch": 2.7883414295628035,
|
| 1790 |
+
"step": 253
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"loss": 0.00015647141844965518,
|
| 1794 |
+
"grad_norm": 0.004532885737717152,
|
| 1795 |
+
"learning_rate": 2.6955129420176196e-06,
|
| 1796 |
+
"epoch": 2.799444829979181,
|
| 1797 |
+
"step": 254
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"loss": 0.0001516131014795974,
|
| 1801 |
+
"grad_norm": 0.0039943912997841835,
|
| 1802 |
+
"learning_rate": 2.433771981578581e-06,
|
| 1803 |
+
"epoch": 2.8105482303955585,
|
| 1804 |
+
"step": 255
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"loss": 0.0005077925743535161,
|
| 1808 |
+
"grad_norm": 0.014085155911743641,
|
| 1809 |
+
"learning_rate": 2.1852399266194314e-06,
|
| 1810 |
+
"epoch": 2.821651630811936,
|
| 1811 |
+
"step": 256
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"loss": 0.0010776021517813206,
|
| 1815 |
+
"grad_norm": 0.03057239204645157,
|
| 1816 |
+
"learning_rate": 1.9499504244020693e-06,
|
| 1817 |
+
"epoch": 2.832755031228314,
|
| 1818 |
+
"step": 257
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"loss": 0.001233020331710577,
|
| 1822 |
+
"grad_norm": 0.0293893925845623,
|
| 1823 |
+
"learning_rate": 1.7279353293586765e-06,
|
| 1824 |
+
"epoch": 2.843858431644691,
|
| 1825 |
+
"step": 258
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"loss": 0.00045728698023594916,
|
| 1829 |
+
"grad_norm": 0.012387805618345737,
|
| 1830 |
+
"learning_rate": 1.5192246987791981e-06,
|
| 1831 |
+
"epoch": 2.854961832061069,
|
| 1832 |
+
"step": 259
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"loss": 0.006937120109796524,
|
| 1836 |
+
"grad_norm": 0.051831986755132675,
|
| 1837 |
+
"learning_rate": 1.323846788742078e-06,
|
| 1838 |
+
"epoch": 2.866065232477446,
|
| 1839 |
+
"step": 260
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"loss": 0.0007000218611210585,
|
| 1843 |
+
"grad_norm": 0.02848062850534916,
|
| 1844 |
+
"learning_rate": 1.14182805028884e-06,
|
| 1845 |
+
"epoch": 2.8771686328938237,
|
| 1846 |
+
"step": 261
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"loss": 0.003899730509147048,
|
| 1850 |
+
"grad_norm": 0.06617487221956253,
|
| 1851 |
+
"learning_rate": 9.731931258429638e-07,
|
| 1852 |
+
"epoch": 2.8882720333102014,
|
| 1853 |
+
"step": 262
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"loss": 0.002479749033227563,
|
| 1857 |
+
"grad_norm": 0.042154788970947266,
|
| 1858 |
+
"learning_rate": 8.17964845873831e-07,
|
| 1859 |
+
"epoch": 2.8993754337265787,
|
| 1860 |
+
"step": 263
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"loss": 0.0009774853242561221,
|
| 1864 |
+
"grad_norm": 0.029534008353948593,
|
| 1865 |
+
"learning_rate": 6.761642258056978e-07,
|
| 1866 |
+
"epoch": 2.9104788341429564,
|
| 1867 |
+
"step": 264
|
| 1868 |
+
},
|
| 1869 |
+
{
|
| 1870 |
+
"loss": 0.0017800165805965662,
|
| 1871 |
+
"grad_norm": 0.04713388904929161,
|
| 1872 |
+
"learning_rate": 5.478104631726711e-07,
|
| 1873 |
+
"epoch": 2.9215822345593336,
|
| 1874 |
+
"step": 265
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"loss": 0.00025485060177743435,
|
| 1878 |
+
"grad_norm": 0.0057479869574308395,
|
| 1879 |
+
"learning_rate": 4.329209350195651e-07,
|
| 1880 |
+
"epoch": 2.9326856349757113,
|
| 1881 |
+
"step": 266
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"loss": 0.003669393016025424,
|
| 1885 |
+
"grad_norm": 0.07603989541530609,
|
| 1886 |
+
"learning_rate": 3.315111955493944e-07,
|
| 1887 |
+
"epoch": 2.943789035392089,
|
| 1888 |
+
"step": 267
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"loss": 0.0010033966973423958,
|
| 1892 |
+
"grad_norm": 0.026915691792964935,
|
| 1893 |
+
"learning_rate": 2.4359497401758024e-07,
|
| 1894 |
+
"epoch": 2.9548924358084663,
|
| 1895 |
+
"step": 268
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"loss": 0.0012706794077530503,
|
| 1899 |
+
"grad_norm": 0.023083670064806938,
|
| 1900 |
+
"learning_rate": 1.6918417287318245e-07,
|
| 1901 |
+
"epoch": 2.965995836224844,
|
| 1902 |
+
"step": 269
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"loss": 0.0004130644374527037,
|
| 1906 |
+
"grad_norm": 0.00994869228452444,
|
| 1907 |
+
"learning_rate": 1.0828886614754341e-07,
|
| 1908 |
+
"epoch": 2.9770992366412212,
|
| 1909 |
+
"step": 270
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"loss": 0.0012230968568474054,
|
| 1913 |
+
"grad_norm": 0.026720473542809486,
|
| 1914 |
+
"learning_rate": 6.09172980904238e-08,
|
| 1915 |
+
"epoch": 2.988202637057599,
|
| 1916 |
+
"step": 271
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"loss": 0.000913719180971384,
|
| 1920 |
+
"grad_norm": 0.01431743148714304,
|
| 1921 |
+
"learning_rate": 2.7075882053828605e-08,
|
| 1922 |
+
"epoch": 2.9993060374739766,
|
| 1923 |
+
"step": 272
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"loss": 0.01702476479113102,
|
| 1927 |
+
"grad_norm": 0.5101261734962463,
|
| 1928 |
+
"learning_rate": 6.769199623779532e-09,
|
| 1929 |
+
"epoch": 3.0,
|
| 1930 |
+
"step": 273
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"train_runtime": 9101.4738,
|
| 1934 |
+
"train_samples_per_second": 0.475,
|
| 1935 |
+
"train_steps_per_second": 0.03,
|
| 1936 |
+
"total_flos": 5.568150387086623e+17,
|
| 1937 |
+
"train_loss": 0.01574120909053851,
|
| 1938 |
+
"epoch": 3.0,
|
| 1939 |
+
"step": 273
|
| 1940 |
+
}
|
| 1941 |
+
]
|
| 1942 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8
|
| 3 |
+
size 11421892
|
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": 32768,
|
| 26 |
+
"pad_token": "<|endoftext|>",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null
|
| 30 |
+
}
|