Upload folder using huggingface_hub
Browse files
.gitattributes
CHANGED
|
@@ -37,3 +37,4 @@ D_full_emf1_seed22/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
| 37 |
D_full_emf1_seed23/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
D_full_corrected_seed22/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 39 |
D_full_corrected_seed23/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 37 |
D_full_emf1_seed23/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
D_full_corrected_seed22/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 39 |
D_full_corrected_seed23/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
D_verified_seed22/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
D_verified_seed22/adapter_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": null,
|
| 6 |
+
"base_model_name_or_path": "/home/yuxu/.cache/huggingface/hub/models--Qwen--Qwen2.5-1.5B-Instruct/snapshots/989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
|
| 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 |
+
"layer_replication": null,
|
| 16 |
+
"layers_pattern": null,
|
| 17 |
+
"layers_to_transform": null,
|
| 18 |
+
"loftq_config": {},
|
| 19 |
+
"lora_alpha": 32,
|
| 20 |
+
"lora_bias": false,
|
| 21 |
+
"lora_dropout": 0.05,
|
| 22 |
+
"megatron_config": null,
|
| 23 |
+
"megatron_core": "megatron.core",
|
| 24 |
+
"modules_to_save": null,
|
| 25 |
+
"peft_type": "LORA",
|
| 26 |
+
"peft_version": "0.18.1",
|
| 27 |
+
"qalora_group_size": 16,
|
| 28 |
+
"r": 16,
|
| 29 |
+
"rank_pattern": {},
|
| 30 |
+
"revision": null,
|
| 31 |
+
"target_modules": [
|
| 32 |
+
"k_proj",
|
| 33 |
+
"o_proj",
|
| 34 |
+
"v_proj",
|
| 35 |
+
"q_proj"
|
| 36 |
+
],
|
| 37 |
+
"target_parameters": null,
|
| 38 |
+
"task_type": "CAUSAL_LM",
|
| 39 |
+
"trainable_token_indices": null,
|
| 40 |
+
"use_dora": false,
|
| 41 |
+
"use_qalora": false,
|
| 42 |
+
"use_rslora": false
|
| 43 |
+
}
|
D_verified_seed22/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd47dff0747cfda9274e5ad16738842c11c9e94c7ee75bb78af45424d074a25a
|
| 3 |
+
size 17462432
|
D_verified_seed22/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 %}
|
D_verified_seed22/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b4360dd6a184650ffc48056c2569bc603f896c5adfe94b10f1c79f809638aa5
|
| 3 |
+
size 11422166
|
D_verified_seed22/tokenizer_config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": true,
|
| 24 |
+
"model_max_length": 131072,
|
| 25 |
+
"pad_token": "<|endoftext|>",
|
| 26 |
+
"split_special_tokens": false,
|
| 27 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 28 |
+
"unk_token": null
|
| 29 |
+
}
|
D_verified_seed22/training_summary.json
ADDED
|
@@ -0,0 +1,772 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_name": "Qwen/Qwen2.5-1.5B-Instruct",
|
| 3 |
+
"resolved_model": "/home/yuxu/.cache/huggingface/hub/models--Qwen--Qwen2.5-1.5B-Instruct/snapshots/989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
|
| 4 |
+
"num_train_records": 3185,
|
| 5 |
+
"num_validation_records_available": 0,
|
| 6 |
+
"training_config": {
|
| 7 |
+
"lora_rank": 16,
|
| 8 |
+
"lora_alpha": 32,
|
| 9 |
+
"learning_rate": 1e-05,
|
| 10 |
+
"num_epochs": 2,
|
| 11 |
+
"per_device_batch_size": 1,
|
| 12 |
+
"per_device_eval_batch_size": 1,
|
| 13 |
+
"gradient_accumulation_steps": 16,
|
| 14 |
+
"beta": 0.1,
|
| 15 |
+
"max_length": 512,
|
| 16 |
+
"max_prompt_length": 384,
|
| 17 |
+
"seed": 22,
|
| 18 |
+
"save_steps": 250,
|
| 19 |
+
"save_total_limit": 2,
|
| 20 |
+
"resume_from_checkpoint": null,
|
| 21 |
+
"precision_dtype": "torch.bfloat16",
|
| 22 |
+
"bf16": true,
|
| 23 |
+
"fp16": false,
|
| 24 |
+
"tokenizer_add_bos_token": false
|
| 25 |
+
},
|
| 26 |
+
"global_step": 400,
|
| 27 |
+
"best_metric": null,
|
| 28 |
+
"best_model_checkpoint": null,
|
| 29 |
+
"train_metrics": {
|
| 30 |
+
"train_runtime": 1812.7376,
|
| 31 |
+
"train_samples_per_second": 3.514,
|
| 32 |
+
"train_steps_per_second": 0.221,
|
| 33 |
+
"total_flos": 4.273969110335078e+16,
|
| 34 |
+
"train_loss": 0.3211171677708626
|
| 35 |
+
},
|
| 36 |
+
"log_history": [
|
| 37 |
+
{
|
| 38 |
+
"loss": 0.6896423816680908,
|
| 39 |
+
"grad_norm": 0.13094085454940796,
|
| 40 |
+
"learning_rate": 9.775e-06,
|
| 41 |
+
"entropy": 0.689751333463937,
|
| 42 |
+
"num_tokens": 131970.0,
|
| 43 |
+
"logits/chosen": -1.1523253123048298,
|
| 44 |
+
"logits/rejected": -0.9633123579481915,
|
| 45 |
+
"mean_token_accuracy": 0.7518096381798387,
|
| 46 |
+
"rewards/chosen": 0.002586326666369132,
|
| 47 |
+
"rewards/rejected": -0.004760412030287852,
|
| 48 |
+
"rewards/accuracies": 0.51875,
|
| 49 |
+
"rewards/margins": 0.007346738569322042,
|
| 50 |
+
"logps/chosen": -39.62088840007782,
|
| 51 |
+
"logps/rejected": -47.373421096801756,
|
| 52 |
+
"epoch": 0.05023547880690738,
|
| 53 |
+
"step": 10
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"loss": 0.667300033569336,
|
| 57 |
+
"grad_norm": 0.11734313517808914,
|
| 58 |
+
"learning_rate": 9.525000000000001e-06,
|
| 59 |
+
"entropy": 0.7374365024734288,
|
| 60 |
+
"num_tokens": 264625.0,
|
| 61 |
+
"logits/chosen": -1.1952187331536142,
|
| 62 |
+
"logits/rejected": -0.8921757166187223,
|
| 63 |
+
"mean_token_accuracy": 0.7644516002386809,
|
| 64 |
+
"rewards/chosen": -0.014206019972789363,
|
| 65 |
+
"rewards/rejected": -0.06760225054007378,
|
| 66 |
+
"rewards/accuracies": 0.80625,
|
| 67 |
+
"rewards/margins": 0.053396230508224105,
|
| 68 |
+
"logps/chosen": -41.583077657222745,
|
| 69 |
+
"logps/rejected": -52.24968423843384,
|
| 70 |
+
"epoch": 0.10047095761381476,
|
| 71 |
+
"step": 20
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"loss": 0.6271667003631591,
|
| 75 |
+
"grad_norm": 0.11499293893575668,
|
| 76 |
+
"learning_rate": 9.275e-06,
|
| 77 |
+
"entropy": 0.708477851934731,
|
| 78 |
+
"num_tokens": 397228.0,
|
| 79 |
+
"logits/chosen": -1.150268087316235,
|
| 80 |
+
"logits/rejected": -0.951233671020332,
|
| 81 |
+
"mean_token_accuracy": 0.7672501749359071,
|
| 82 |
+
"rewards/chosen": -0.05163738687588193,
|
| 83 |
+
"rewards/rejected": -0.19341393218783196,
|
| 84 |
+
"rewards/accuracies": 0.90625,
|
| 85 |
+
"rewards/margins": 0.14177654468803666,
|
| 86 |
+
"logps/chosen": -42.39029086828232,
|
| 87 |
+
"logps/rejected": -51.21238135099411,
|
| 88 |
+
"epoch": 0.15070643642072212,
|
| 89 |
+
"step": 30
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"loss": 0.579903507232666,
|
| 93 |
+
"grad_norm": 0.11684582382440567,
|
| 94 |
+
"learning_rate": 9.025e-06,
|
| 95 |
+
"entropy": 0.6974922352470457,
|
| 96 |
+
"num_tokens": 529886.0,
|
| 97 |
+
"logits/chosen": -1.2177221530071947,
|
| 98 |
+
"logits/rejected": -1.0076879473921485,
|
| 99 |
+
"mean_token_accuracy": 0.7614046652801335,
|
| 100 |
+
"rewards/chosen": -0.12291871903926221,
|
| 101 |
+
"rewards/rejected": -0.37786176332156174,
|
| 102 |
+
"rewards/accuracies": 0.90625,
|
| 103 |
+
"rewards/margins": 0.2549430426210165,
|
| 104 |
+
"logps/chosen": -46.055079424381255,
|
| 105 |
+
"logps/rejected": -52.941705322265626,
|
| 106 |
+
"epoch": 0.20094191522762953,
|
| 107 |
+
"step": 40
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"loss": 0.5364655017852783,
|
| 111 |
+
"grad_norm": 0.09873486310243607,
|
| 112 |
+
"learning_rate": 8.775e-06,
|
| 113 |
+
"entropy": 0.731591874291189,
|
| 114 |
+
"num_tokens": 662980.0,
|
| 115 |
+
"logits/chosen": -1.3299434378967778,
|
| 116 |
+
"logits/rejected": -1.076280700265539,
|
| 117 |
+
"mean_token_accuracy": 0.7357267120853066,
|
| 118 |
+
"rewards/chosen": -0.2148243856656336,
|
| 119 |
+
"rewards/rejected": -0.5955711705770227,
|
| 120 |
+
"rewards/accuracies": 0.8875,
|
| 121 |
+
"rewards/margins": 0.38074678448028865,
|
| 122 |
+
"logps/chosen": -51.49660888910294,
|
| 123 |
+
"logps/rejected": -54.77330160140991,
|
| 124 |
+
"epoch": 0.25117739403453687,
|
| 125 |
+
"step": 50
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"loss": 0.5024271488189698,
|
| 129 |
+
"grad_norm": 0.12186943739652634,
|
| 130 |
+
"learning_rate": 8.525e-06,
|
| 131 |
+
"entropy": 0.7015456688357518,
|
| 132 |
+
"num_tokens": 796107.0,
|
| 133 |
+
"logits/chosen": -1.3661355904385775,
|
| 134 |
+
"logits/rejected": -1.1044401187614372,
|
| 135 |
+
"mean_token_accuracy": 0.7381971096619964,
|
| 136 |
+
"rewards/chosen": -0.4195166851852264,
|
| 137 |
+
"rewards/rejected": -0.929675487615168,
|
| 138 |
+
"rewards/accuracies": 0.8875,
|
| 139 |
+
"rewards/margins": 0.510158799123019,
|
| 140 |
+
"logps/chosen": -55.01945576667786,
|
| 141 |
+
"logps/rejected": -59.47034276723862,
|
| 142 |
+
"epoch": 0.30141287284144425,
|
| 143 |
+
"step": 60
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"loss": 0.44471449851989747,
|
| 147 |
+
"grad_norm": 0.12061455100774765,
|
| 148 |
+
"learning_rate": 8.275000000000001e-06,
|
| 149 |
+
"entropy": 0.6852886462933384,
|
| 150 |
+
"num_tokens": 929478.0,
|
| 151 |
+
"logits/chosen": -1.4021386362206496,
|
| 152 |
+
"logits/rejected": -1.1261810231833123,
|
| 153 |
+
"mean_token_accuracy": 0.7667396960780024,
|
| 154 |
+
"rewards/chosen": -0.5427155995392241,
|
| 155 |
+
"rewards/rejected": -1.2492919825599529,
|
| 156 |
+
"rewards/accuracies": 0.85625,
|
| 157 |
+
"rewards/margins": 0.7065763819031418,
|
| 158 |
+
"logps/chosen": -53.6978485584259,
|
| 159 |
+
"logps/rejected": -64.24746949672699,
|
| 160 |
+
"epoch": 0.3516483516483517,
|
| 161 |
+
"step": 70
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"loss": 0.42475309371948244,
|
| 165 |
+
"grad_norm": 0.11903995275497437,
|
| 166 |
+
"learning_rate": 8.025e-06,
|
| 167 |
+
"entropy": 0.6787039614515379,
|
| 168 |
+
"num_tokens": 1062022.0,
|
| 169 |
+
"logits/chosen": -1.5759432148317676,
|
| 170 |
+
"logits/rejected": -1.2306955382113542,
|
| 171 |
+
"mean_token_accuracy": 0.7302945245988667,
|
| 172 |
+
"rewards/chosen": -0.6781344773538877,
|
| 173 |
+
"rewards/rejected": -1.488679854432121,
|
| 174 |
+
"rewards/accuracies": 0.8625,
|
| 175 |
+
"rewards/margins": 0.8105453730095178,
|
| 176 |
+
"logps/chosen": -54.423488879203795,
|
| 177 |
+
"logps/rejected": -61.99909480810165,
|
| 178 |
+
"epoch": 0.40188383045525905,
|
| 179 |
+
"step": 80
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"loss": 0.40935330390930175,
|
| 183 |
+
"grad_norm": 0.1066775843501091,
|
| 184 |
+
"learning_rate": 7.775000000000001e-06,
|
| 185 |
+
"entropy": 0.6443115042406135,
|
| 186 |
+
"num_tokens": 1195866.0,
|
| 187 |
+
"logits/chosen": -1.6532846289662317,
|
| 188 |
+
"logits/rejected": -1.2993566104277297,
|
| 189 |
+
"mean_token_accuracy": 0.7505350362509489,
|
| 190 |
+
"rewards/chosen": -0.9569830165567736,
|
| 191 |
+
"rewards/rejected": -1.9420622304081916,
|
| 192 |
+
"rewards/accuracies": 0.84375,
|
| 193 |
+
"rewards/margins": 0.9850792161189019,
|
| 194 |
+
"logps/chosen": -60.308094668388364,
|
| 195 |
+
"logps/rejected": -69.52688453197479,
|
| 196 |
+
"epoch": 0.4521193092621664,
|
| 197 |
+
"step": 90
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"loss": 0.3887610673904419,
|
| 201 |
+
"grad_norm": 0.1309821456670761,
|
| 202 |
+
"learning_rate": 7.525e-06,
|
| 203 |
+
"entropy": 0.6520523697749013,
|
| 204 |
+
"num_tokens": 1328318.0,
|
| 205 |
+
"logits/chosen": -1.7746430284347894,
|
| 206 |
+
"logits/rejected": -1.4033554953496512,
|
| 207 |
+
"mean_token_accuracy": 0.7270496355835349,
|
| 208 |
+
"rewards/chosen": -0.9801554235164076,
|
| 209 |
+
"rewards/rejected": -2.104814527183771,
|
| 210 |
+
"rewards/accuracies": 0.8625,
|
| 211 |
+
"rewards/margins": 1.1246590989641845,
|
| 212 |
+
"logps/chosen": -56.47254922389984,
|
| 213 |
+
"logps/rejected": -67.29142245054246,
|
| 214 |
+
"epoch": 0.5023547880690737,
|
| 215 |
+
"step": 100
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"loss": 0.3660599708557129,
|
| 219 |
+
"grad_norm": 0.10758155584335327,
|
| 220 |
+
"learning_rate": 7.275000000000001e-06,
|
| 221 |
+
"entropy": 0.6838750483802869,
|
| 222 |
+
"num_tokens": 1461973.0,
|
| 223 |
+
"logits/chosen": -1.6473026648423121,
|
| 224 |
+
"logits/rejected": -1.3700423881868464,
|
| 225 |
+
"mean_token_accuracy": 0.7227723866701126,
|
| 226 |
+
"rewards/chosen": -1.3291343829288962,
|
| 227 |
+
"rewards/rejected": -2.621742352005094,
|
| 228 |
+
"rewards/accuracies": 0.88125,
|
| 229 |
+
"rewards/margins": 1.292607967182994,
|
| 230 |
+
"logps/chosen": -65.51949642896652,
|
| 231 |
+
"logps/rejected": -77.87361862659455,
|
| 232 |
+
"epoch": 0.5525902668759811,
|
| 233 |
+
"step": 110
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"loss": 0.30113241672515867,
|
| 237 |
+
"grad_norm": 0.14314669370651245,
|
| 238 |
+
"learning_rate": 7.0250000000000005e-06,
|
| 239 |
+
"entropy": 0.6215112688369118,
|
| 240 |
+
"num_tokens": 1595182.0,
|
| 241 |
+
"logits/chosen": -1.934697893487676,
|
| 242 |
+
"logits/rejected": -1.5154955753295023,
|
| 243 |
+
"mean_token_accuracy": 0.7339814778417348,
|
| 244 |
+
"rewards/chosen": -1.2177859152667225,
|
| 245 |
+
"rewards/rejected": -2.8568852407857777,
|
| 246 |
+
"rewards/accuracies": 0.925,
|
| 247 |
+
"rewards/margins": 1.6390993235632778,
|
| 248 |
+
"logps/chosen": -60.70231634378433,
|
| 249 |
+
"logps/rejected": -79.2034349322319,
|
| 250 |
+
"epoch": 0.6028257456828885,
|
| 251 |
+
"step": 120
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"loss": 0.2874736309051514,
|
| 255 |
+
"grad_norm": 0.17616006731987,
|
| 256 |
+
"learning_rate": 6.775e-06,
|
| 257 |
+
"entropy": 0.6661754509317689,
|
| 258 |
+
"num_tokens": 1728787.0,
|
| 259 |
+
"logits/chosen": -1.9861773831873346,
|
| 260 |
+
"logits/rejected": -1.5237576000681574,
|
| 261 |
+
"mean_token_accuracy": 0.7176908973604441,
|
| 262 |
+
"rewards/chosen": -1.315507940785028,
|
| 263 |
+
"rewards/rejected": -3.1794131041038782,
|
| 264 |
+
"rewards/accuracies": 0.8875,
|
| 265 |
+
"rewards/margins": 1.8639051569625735,
|
| 266 |
+
"logps/chosen": -62.42429987192154,
|
| 267 |
+
"logps/rejected": -84.25045642852783,
|
| 268 |
+
"epoch": 0.6530612244897959,
|
| 269 |
+
"step": 130
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"loss": 0.3126381874084473,
|
| 273 |
+
"grad_norm": 0.14193673431873322,
|
| 274 |
+
"learning_rate": 6.525e-06,
|
| 275 |
+
"entropy": 0.6002552698686486,
|
| 276 |
+
"num_tokens": 1861599.0,
|
| 277 |
+
"logits/chosen": -2.122078231381872,
|
| 278 |
+
"logits/rejected": -1.6287145550832007,
|
| 279 |
+
"mean_token_accuracy": 0.7182940527796745,
|
| 280 |
+
"rewards/chosen": -1.3322255573002622,
|
| 281 |
+
"rewards/rejected": -3.025114858860616,
|
| 282 |
+
"rewards/accuracies": 0.8875,
|
| 283 |
+
"rewards/margins": 1.6928893025964498,
|
| 284 |
+
"logps/chosen": -57.53818924427033,
|
| 285 |
+
"logps/rejected": -80.19278188943863,
|
| 286 |
+
"epoch": 0.7032967032967034,
|
| 287 |
+
"step": 140
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"loss": 0.2915832757949829,
|
| 291 |
+
"grad_norm": 0.1305493861436844,
|
| 292 |
+
"learning_rate": 6.275e-06,
|
| 293 |
+
"entropy": 0.6727007152163423,
|
| 294 |
+
"num_tokens": 1995013.0,
|
| 295 |
+
"logits/chosen": -2.0417178543886463,
|
| 296 |
+
"logits/rejected": -1.5001922485515107,
|
| 297 |
+
"mean_token_accuracy": 0.7270697236992418,
|
| 298 |
+
"rewards/chosen": -1.4838186738521473,
|
| 299 |
+
"rewards/rejected": -3.6069449370726945,
|
| 300 |
+
"rewards/accuracies": 0.88125,
|
| 301 |
+
"rewards/margins": 2.1231262689456343,
|
| 302 |
+
"logps/chosen": -64.43461210727692,
|
| 303 |
+
"logps/rejected": -91.32557408809662,
|
| 304 |
+
"epoch": 0.7535321821036107,
|
| 305 |
+
"step": 150
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"loss": 0.2940825462341309,
|
| 309 |
+
"grad_norm": 0.08979333937168121,
|
| 310 |
+
"learning_rate": 6.025000000000001e-06,
|
| 311 |
+
"entropy": 0.5827442438137951,
|
| 312 |
+
"num_tokens": 2127474.0,
|
| 313 |
+
"logits/chosen": -2.099819178823348,
|
| 314 |
+
"logits/rejected": -1.6885847668202878,
|
| 315 |
+
"mean_token_accuracy": 0.7267681088298559,
|
| 316 |
+
"rewards/chosen": -1.4805560540087754,
|
| 317 |
+
"rewards/rejected": -3.3969569090753793,
|
| 318 |
+
"rewards/accuracies": 0.875,
|
| 319 |
+
"rewards/margins": 1.9164008550345897,
|
| 320 |
+
"logps/chosen": -57.75792602300644,
|
| 321 |
+
"logps/rejected": -81.37804629802704,
|
| 322 |
+
"epoch": 0.8037676609105181,
|
| 323 |
+
"step": 160
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"loss": 0.2389240264892578,
|
| 327 |
+
"grad_norm": 0.139341801404953,
|
| 328 |
+
"learning_rate": 5.775000000000001e-06,
|
| 329 |
+
"entropy": 0.5839000599458813,
|
| 330 |
+
"num_tokens": 2260033.0,
|
| 331 |
+
"logits/chosen": -2.2174489045489976,
|
| 332 |
+
"logits/rejected": -1.6226538364130199,
|
| 333 |
+
"mean_token_accuracy": 0.7345416588708759,
|
| 334 |
+
"rewards/chosen": -1.493336088315118,
|
| 335 |
+
"rewards/rejected": -3.709612054377794,
|
| 336 |
+
"rewards/accuracies": 0.9125,
|
| 337 |
+
"rewards/margins": 2.2162759650498627,
|
| 338 |
+
"logps/chosen": -56.21773297786713,
|
| 339 |
+
"logps/rejected": -86.65012849569321,
|
| 340 |
+
"epoch": 0.8540031397174255,
|
| 341 |
+
"step": 170
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"loss": 0.28322443962097166,
|
| 345 |
+
"grad_norm": 0.23025472462177277,
|
| 346 |
+
"learning_rate": 5.5250000000000005e-06,
|
| 347 |
+
"entropy": 0.6710750604979694,
|
| 348 |
+
"num_tokens": 2393967.0,
|
| 349 |
+
"logits/chosen": -2.0555160703694106,
|
| 350 |
+
"logits/rejected": -1.5437636215952086,
|
| 351 |
+
"mean_token_accuracy": 0.7202916046604514,
|
| 352 |
+
"rewards/chosen": -1.7690606065036263,
|
| 353 |
+
"rewards/rejected": -3.9886667074635627,
|
| 354 |
+
"rewards/accuracies": 0.89375,
|
| 355 |
+
"rewards/margins": 2.219606099836528,
|
| 356 |
+
"logps/chosen": -67.48856694698334,
|
| 357 |
+
"logps/rejected": -92.84689679145814,
|
| 358 |
+
"epoch": 0.9042386185243328,
|
| 359 |
+
"step": 180
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"loss": 0.29533464908599855,
|
| 363 |
+
"grad_norm": 0.12374638020992279,
|
| 364 |
+
"learning_rate": 5.275e-06,
|
| 365 |
+
"entropy": 0.6247204889848945,
|
| 366 |
+
"num_tokens": 2526184.0,
|
| 367 |
+
"logits/chosen": -2.051882173583905,
|
| 368 |
+
"logits/rejected": -1.6121547510662348,
|
| 369 |
+
"mean_token_accuracy": 0.7194008273072541,
|
| 370 |
+
"rewards/chosen": -1.7575724313326646,
|
| 371 |
+
"rewards/rejected": -3.731215941905975,
|
| 372 |
+
"rewards/accuracies": 0.89375,
|
| 373 |
+
"rewards/margins": 1.9736435044556857,
|
| 374 |
+
"logps/chosen": -68.31192799806595,
|
| 375 |
+
"logps/rejected": -83.8547656774521,
|
| 376 |
+
"epoch": 0.9544740973312402,
|
| 377 |
+
"step": 190
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"loss": 0.21644911766052247,
|
| 381 |
+
"grad_norm": 0.011066500097513199,
|
| 382 |
+
"learning_rate": 5.025e-06,
|
| 383 |
+
"entropy": 0.6499112024196777,
|
| 384 |
+
"num_tokens": 2646778.0,
|
| 385 |
+
"logits/chosen": -2.0789769635962476,
|
| 386 |
+
"logits/rejected": -1.6486789713430938,
|
| 387 |
+
"mean_token_accuracy": 0.7179098229983757,
|
| 388 |
+
"rewards/chosen": -1.8985417433821692,
|
| 389 |
+
"rewards/rejected": -4.1633654734184,
|
| 390 |
+
"rewards/accuracies": 0.896551724137931,
|
| 391 |
+
"rewards/margins": 2.2648237388709496,
|
| 392 |
+
"logps/chosen": -66.81492134620403,
|
| 393 |
+
"logps/rejected": -90.29491719213026,
|
| 394 |
+
"epoch": 1.0,
|
| 395 |
+
"step": 200
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"loss": 0.2303311347961426,
|
| 399 |
+
"grad_norm": 0.16805951297283173,
|
| 400 |
+
"learning_rate": 4.775e-06,
|
| 401 |
+
"entropy": 0.6100569258545875,
|
| 402 |
+
"num_tokens": 2779951.0,
|
| 403 |
+
"logits/chosen": -2.18281432171439,
|
| 404 |
+
"logits/rejected": -1.6656628587918842,
|
| 405 |
+
"mean_token_accuracy": 0.7220925435423851,
|
| 406 |
+
"rewards/chosen": -1.9006830145488494,
|
| 407 |
+
"rewards/rejected": -4.296074335649609,
|
| 408 |
+
"rewards/accuracies": 0.9,
|
| 409 |
+
"rewards/margins": 2.3953913148492574,
|
| 410 |
+
"logps/chosen": -65.32953391075134,
|
| 411 |
+
"logps/rejected": -93.77217375040054,
|
| 412 |
+
"epoch": 1.0502354788069075,
|
| 413 |
+
"step": 210
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"loss": 0.25216624736785886,
|
| 417 |
+
"grad_norm": 0.12303531914949417,
|
| 418 |
+
"learning_rate": 4.525000000000001e-06,
|
| 419 |
+
"entropy": 0.568905234970589,
|
| 420 |
+
"num_tokens": 2912772.0,
|
| 421 |
+
"logits/chosen": -2.3079234464466842,
|
| 422 |
+
"logits/rejected": -1.7598141767542064,
|
| 423 |
+
"mean_token_accuracy": 0.7157205916941166,
|
| 424 |
+
"rewards/chosen": -2.1148597109655385,
|
| 425 |
+
"rewards/rejected": -4.674917114898562,
|
| 426 |
+
"rewards/accuracies": 0.8875,
|
| 427 |
+
"rewards/margins": 2.5600573919713496,
|
| 428 |
+
"logps/chosen": -64.76336991786957,
|
| 429 |
+
"logps/rejected": -95.60576952695847,
|
| 430 |
+
"epoch": 1.1004709576138147,
|
| 431 |
+
"step": 220
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"loss": 0.24181344509124755,
|
| 435 |
+
"grad_norm": 0.17596560716629028,
|
| 436 |
+
"learning_rate": 4.2750000000000006e-06,
|
| 437 |
+
"entropy": 0.6560148949036375,
|
| 438 |
+
"num_tokens": 3045642.0,
|
| 439 |
+
"logits/chosen": -2.195677149413134,
|
| 440 |
+
"logits/rejected": -1.6967719311669462,
|
| 441 |
+
"mean_token_accuracy": 0.7058368335943669,
|
| 442 |
+
"rewards/chosen": -2.2872255038586444,
|
| 443 |
+
"rewards/rejected": -4.651171739399433,
|
| 444 |
+
"rewards/accuracies": 0.91875,
|
| 445 |
+
"rewards/margins": 2.363946241140366,
|
| 446 |
+
"logps/chosen": -71.82424494028092,
|
| 447 |
+
"logps/rejected": -95.5311926484108,
|
| 448 |
+
"epoch": 1.1507064364207222,
|
| 449 |
+
"step": 230
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"loss": 0.2337651014328003,
|
| 453 |
+
"grad_norm": 0.12319736927747726,
|
| 454 |
+
"learning_rate": 4.0250000000000004e-06,
|
| 455 |
+
"entropy": 0.5653066983068129,
|
| 456 |
+
"num_tokens": 3177845.0,
|
| 457 |
+
"logits/chosen": -2.365491519276238,
|
| 458 |
+
"logits/rejected": -1.8260216611432554,
|
| 459 |
+
"mean_token_accuracy": 0.7125749061815441,
|
| 460 |
+
"rewards/chosen": -1.9638693497210624,
|
| 461 |
+
"rewards/rejected": -4.608601247519255,
|
| 462 |
+
"rewards/accuracies": 0.925,
|
| 463 |
+
"rewards/margins": 2.644731895625591,
|
| 464 |
+
"logps/chosen": -61.488043415546414,
|
| 465 |
+
"logps/rejected": -94.4761459350586,
|
| 466 |
+
"epoch": 1.2009419152276295,
|
| 467 |
+
"step": 240
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"loss": 0.22149863243103027,
|
| 471 |
+
"grad_norm": 0.15987302362918854,
|
| 472 |
+
"learning_rate": 3.7750000000000003e-06,
|
| 473 |
+
"entropy": 0.6203280808171258,
|
| 474 |
+
"num_tokens": 3310262.0,
|
| 475 |
+
"logits/chosen": -2.275773576342021,
|
| 476 |
+
"logits/rejected": -1.804882297520518,
|
| 477 |
+
"mean_token_accuracy": 0.7059518062509597,
|
| 478 |
+
"rewards/chosen": -2.0083323230064707,
|
| 479 |
+
"rewards/rejected": -4.590591602027416,
|
| 480 |
+
"rewards/accuracies": 0.93125,
|
| 481 |
+
"rewards/margins": 2.5822592776268722,
|
| 482 |
+
"logps/chosen": -69.69174835681915,
|
| 483 |
+
"logps/rejected": -93.78124842643737,
|
| 484 |
+
"epoch": 1.2511773940345368,
|
| 485 |
+
"step": 250
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"loss": 0.2983433246612549,
|
| 489 |
+
"grad_norm": 0.30712154507637024,
|
| 490 |
+
"learning_rate": 3.525e-06,
|
| 491 |
+
"entropy": 0.6800256669637748,
|
| 492 |
+
"num_tokens": 3444453.0,
|
| 493 |
+
"logits/chosen": -2.1269210300066197,
|
| 494 |
+
"logits/rejected": -1.644955600095911,
|
| 495 |
+
"mean_token_accuracy": 0.7150781577453017,
|
| 496 |
+
"rewards/chosen": -2.3532459220557937,
|
| 497 |
+
"rewards/rejected": -5.054403532575816,
|
| 498 |
+
"rewards/accuracies": 0.86875,
|
| 499 |
+
"rewards/margins": 2.7011575967073442,
|
| 500 |
+
"logps/chosen": -76.04198013544082,
|
| 501 |
+
"logps/rejected": -107.49599938392639,
|
| 502 |
+
"epoch": 1.3014128728414442,
|
| 503 |
+
"step": 260
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"loss": 0.24631202220916748,
|
| 507 |
+
"grad_norm": 0.09332051128149033,
|
| 508 |
+
"learning_rate": 3.2750000000000004e-06,
|
| 509 |
+
"entropy": 0.5894242083537392,
|
| 510 |
+
"num_tokens": 3577246.0,
|
| 511 |
+
"logits/chosen": -2.3688034717069137,
|
| 512 |
+
"logits/rejected": -1.7370439701261446,
|
| 513 |
+
"mean_token_accuracy": 0.7299393502995372,
|
| 514 |
+
"rewards/chosen": -1.845114409609232,
|
| 515 |
+
"rewards/rejected": -4.53761124163866,
|
| 516 |
+
"rewards/accuracies": 0.9125,
|
| 517 |
+
"rewards/margins": 2.6924968369305136,
|
| 518 |
+
"logps/chosen": -63.63579627275467,
|
| 519 |
+
"logps/rejected": -94.37624535560607,
|
| 520 |
+
"epoch": 1.3516483516483517,
|
| 521 |
+
"step": 270
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"loss": 0.23863513469696046,
|
| 525 |
+
"grad_norm": 0.10272336006164551,
|
| 526 |
+
"learning_rate": 3.0250000000000003e-06,
|
| 527 |
+
"entropy": 0.5985479053982999,
|
| 528 |
+
"num_tokens": 3709965.0,
|
| 529 |
+
"logits/chosen": -2.1876354155981135,
|
| 530 |
+
"logits/rejected": -1.6744618003037917,
|
| 531 |
+
"mean_token_accuracy": 0.7436435533687472,
|
| 532 |
+
"rewards/chosen": -2.027228345174808,
|
| 533 |
+
"rewards/rejected": -4.674324031174183,
|
| 534 |
+
"rewards/accuracies": 0.93125,
|
| 535 |
+
"rewards/margins": 2.6470956701785324,
|
| 536 |
+
"logps/chosen": -65.39346297979355,
|
| 537 |
+
"logps/rejected": -96.76267764568328,
|
| 538 |
+
"epoch": 1.401883830455259,
|
| 539 |
+
"step": 280
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"loss": 0.2838656187057495,
|
| 543 |
+
"grad_norm": 0.11879245936870575,
|
| 544 |
+
"learning_rate": 2.7750000000000005e-06,
|
| 545 |
+
"entropy": 0.6604614646392293,
|
| 546 |
+
"num_tokens": 3843149.0,
|
| 547 |
+
"logits/chosen": -2.038344956790353,
|
| 548 |
+
"logits/rejected": -1.5576915145458856,
|
| 549 |
+
"mean_token_accuracy": 0.700537676177919,
|
| 550 |
+
"rewards/chosen": -1.9881086508707084,
|
| 551 |
+
"rewards/rejected": -4.222263621725142,
|
| 552 |
+
"rewards/accuracies": 0.86875,
|
| 553 |
+
"rewards/margins": 2.2341549627482893,
|
| 554 |
+
"logps/chosen": -68.3739852309227,
|
| 555 |
+
"logps/rejected": -91.22785549163818,
|
| 556 |
+
"epoch": 1.4521193092621665,
|
| 557 |
+
"step": 290
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"loss": 0.22575137615203858,
|
| 561 |
+
"grad_norm": 0.1327482908964157,
|
| 562 |
+
"learning_rate": 2.5250000000000004e-06,
|
| 563 |
+
"entropy": 0.5960212936508469,
|
| 564 |
+
"num_tokens": 3976367.0,
|
| 565 |
+
"logits/chosen": -2.2630690155633295,
|
| 566 |
+
"logits/rejected": -1.6568873585799537,
|
| 567 |
+
"mean_token_accuracy": 0.7286339377984404,
|
| 568 |
+
"rewards/chosen": -1.9000442525139078,
|
| 569 |
+
"rewards/rejected": -4.470313359051943,
|
| 570 |
+
"rewards/accuracies": 0.93125,
|
| 571 |
+
"rewards/margins": 2.570269102603197,
|
| 572 |
+
"logps/chosen": -63.67682906389236,
|
| 573 |
+
"logps/rejected": -96.12785203456879,
|
| 574 |
+
"epoch": 1.5023547880690737,
|
| 575 |
+
"step": 300
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"loss": 0.22741954326629638,
|
| 579 |
+
"grad_norm": 0.08919738233089447,
|
| 580 |
+
"learning_rate": 2.2750000000000002e-06,
|
| 581 |
+
"entropy": 0.6995795883238316,
|
| 582 |
+
"num_tokens": 4110311.0,
|
| 583 |
+
"logits/chosen": -1.9476948164859302,
|
| 584 |
+
"logits/rejected": -1.6104303648197447,
|
| 585 |
+
"mean_token_accuracy": 0.6787193703465164,
|
| 586 |
+
"rewards/chosen": -1.9925882068826468,
|
| 587 |
+
"rewards/rejected": -4.589907981455326,
|
| 588 |
+
"rewards/accuracies": 0.89375,
|
| 589 |
+
"rewards/margins": 2.597319772094488,
|
| 590 |
+
"logps/chosen": -77.3451806306839,
|
| 591 |
+
"logps/rejected": -95.16323103904725,
|
| 592 |
+
"epoch": 1.5525902668759812,
|
| 593 |
+
"step": 310
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"loss": 0.19917845726013184,
|
| 597 |
+
"grad_norm": 0.1700974851846695,
|
| 598 |
+
"learning_rate": 2.025e-06,
|
| 599 |
+
"entropy": 0.6563272744067945,
|
| 600 |
+
"num_tokens": 4243362.0,
|
| 601 |
+
"logits/chosen": -2.196371844299108,
|
| 602 |
+
"logits/rejected": -1.6771500761138032,
|
| 603 |
+
"mean_token_accuracy": 0.7145074622705578,
|
| 604 |
+
"rewards/chosen": -1.8976672918070108,
|
| 605 |
+
"rewards/rejected": -4.560050521045923,
|
| 606 |
+
"rewards/accuracies": 0.91875,
|
| 607 |
+
"rewards/margins": 2.6623832277953623,
|
| 608 |
+
"logps/chosen": -69.23130304813385,
|
| 609 |
+
"logps/rejected": -95.48282803297043,
|
| 610 |
+
"epoch": 1.6028257456828885,
|
| 611 |
+
"step": 320
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"loss": 0.22364423274993897,
|
| 615 |
+
"grad_norm": 0.20068404078483582,
|
| 616 |
+
"learning_rate": 1.7750000000000002e-06,
|
| 617 |
+
"entropy": 0.5802150918229018,
|
| 618 |
+
"num_tokens": 4375925.0,
|
| 619 |
+
"logits/chosen": -2.3776315103846715,
|
| 620 |
+
"logits/rejected": -1.7971645487014372,
|
| 621 |
+
"mean_token_accuracy": 0.7232824593782425,
|
| 622 |
+
"rewards/chosen": -1.9919385116634658,
|
| 623 |
+
"rewards/rejected": -4.705703271180392,
|
| 624 |
+
"rewards/accuracies": 0.91875,
|
| 625 |
+
"rewards/margins": 2.713764737918973,
|
| 626 |
+
"logps/chosen": -61.739083194732665,
|
| 627 |
+
"logps/rejected": -96.16682741641998,
|
| 628 |
+
"epoch": 1.6530612244897958,
|
| 629 |
+
"step": 330
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"loss": 0.16870708465576173,
|
| 633 |
+
"grad_norm": 0.18509013950824738,
|
| 634 |
+
"learning_rate": 1.525e-06,
|
| 635 |
+
"entropy": 0.559020329132909,
|
| 636 |
+
"num_tokens": 4508168.0,
|
| 637 |
+
"logits/chosen": -2.515471136572378,
|
| 638 |
+
"logits/rejected": -1.8837550746889142,
|
| 639 |
+
"mean_token_accuracy": 0.7255307767540217,
|
| 640 |
+
"rewards/chosen": -1.8211148633650738,
|
| 641 |
+
"rewards/rejected": -4.7719439506530765,
|
| 642 |
+
"rewards/accuracies": 0.94375,
|
| 643 |
+
"rewards/margins": 2.9508290767669676,
|
| 644 |
+
"logps/chosen": -61.37455749511719,
|
| 645 |
+
"logps/rejected": -96.88579313755035,
|
| 646 |
+
"epoch": 1.7032967032967035,
|
| 647 |
+
"step": 340
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"loss": 0.27746694087982177,
|
| 651 |
+
"grad_norm": 0.18750888109207153,
|
| 652 |
+
"learning_rate": 1.275e-06,
|
| 653 |
+
"entropy": 0.6220451932924334,
|
| 654 |
+
"num_tokens": 4641537.0,
|
| 655 |
+
"logits/chosen": -2.28468026263477,
|
| 656 |
+
"logits/rejected": -1.757515696813504,
|
| 657 |
+
"mean_token_accuracy": 0.7095036102458835,
|
| 658 |
+
"rewards/chosen": -2.0999434778117574,
|
| 659 |
+
"rewards/rejected": -4.736582234501839,
|
| 660 |
+
"rewards/accuracies": 0.8875,
|
| 661 |
+
"rewards/margins": 2.636638741567731,
|
| 662 |
+
"logps/chosen": -69.99055384397506,
|
| 663 |
+
"logps/rejected": -97.3722316145897,
|
| 664 |
+
"epoch": 1.7535321821036107,
|
| 665 |
+
"step": 350
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"loss": 0.25263917446136475,
|
| 669 |
+
"grad_norm": 0.11750508099794388,
|
| 670 |
+
"learning_rate": 1.025e-06,
|
| 671 |
+
"entropy": 0.5796636143466458,
|
| 672 |
+
"num_tokens": 4774347.0,
|
| 673 |
+
"logits/chosen": -2.3597667018883772,
|
| 674 |
+
"logits/rejected": -1.8791617116259407,
|
| 675 |
+
"mean_token_accuracy": 0.7264608392491937,
|
| 676 |
+
"rewards/chosen": -1.9807785405311733,
|
| 677 |
+
"rewards/rejected": -4.672142916731536,
|
| 678 |
+
"rewards/accuracies": 0.90625,
|
| 679 |
+
"rewards/margins": 2.691364372894168,
|
| 680 |
+
"logps/chosen": -61.77672256231308,
|
| 681 |
+
"logps/rejected": -95.34002628326417,
|
| 682 |
+
"epoch": 1.803767660910518,
|
| 683 |
+
"step": 360
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"loss": 0.2696484804153442,
|
| 687 |
+
"grad_norm": 0.2888956665992737,
|
| 688 |
+
"learning_rate": 7.750000000000001e-07,
|
| 689 |
+
"entropy": 0.5511329970850056,
|
| 690 |
+
"num_tokens": 4906316.0,
|
| 691 |
+
"logits/chosen": -2.3675168545459426,
|
| 692 |
+
"logits/rejected": -1.815755491003597,
|
| 693 |
+
"mean_token_accuracy": 0.7197678625583649,
|
| 694 |
+
"rewards/chosen": -2.0591997408890164,
|
| 695 |
+
"rewards/rejected": -4.720358964102343,
|
| 696 |
+
"rewards/accuracies": 0.89375,
|
| 697 |
+
"rewards/margins": 2.6611592227593066,
|
| 698 |
+
"logps/chosen": -62.35932513475418,
|
| 699 |
+
"logps/rejected": -94.21393445730209,
|
| 700 |
+
"epoch": 1.8540031397174255,
|
| 701 |
+
"step": 370
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"loss": 0.19859030246734619,
|
| 705 |
+
"grad_norm": 0.2091812938451767,
|
| 706 |
+
"learning_rate": 5.250000000000001e-07,
|
| 707 |
+
"entropy": 0.5907206964795477,
|
| 708 |
+
"num_tokens": 5039566.0,
|
| 709 |
+
"logits/chosen": -2.3285244386117294,
|
| 710 |
+
"logits/rejected": -1.7605950618332606,
|
| 711 |
+
"mean_token_accuracy": 0.734518223721534,
|
| 712 |
+
"rewards/chosen": -1.9436712248352706,
|
| 713 |
+
"rewards/rejected": -4.841772738844156,
|
| 714 |
+
"rewards/accuracies": 0.91875,
|
| 715 |
+
"rewards/margins": 2.898101507127285,
|
| 716 |
+
"logps/chosen": -68.03990856409072,
|
| 717 |
+
"logps/rejected": -97.62951198816299,
|
| 718 |
+
"epoch": 1.904238618524333,
|
| 719 |
+
"step": 380
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"loss": 0.2376657247543335,
|
| 723 |
+
"grad_norm": 0.22157330811023712,
|
| 724 |
+
"learning_rate": 2.75e-07,
|
| 725 |
+
"entropy": 0.6367816166239209,
|
| 726 |
+
"num_tokens": 5172584.0,
|
| 727 |
+
"logits/chosen": -2.2245480681783,
|
| 728 |
+
"logits/rejected": -1.7170647767244063,
|
| 729 |
+
"mean_token_accuracy": 0.7177026910707355,
|
| 730 |
+
"rewards/chosen": -2.2113265432557228,
|
| 731 |
+
"rewards/rejected": -5.099810367077589,
|
| 732 |
+
"rewards/accuracies": 0.9125,
|
| 733 |
+
"rewards/margins": 2.8884838223457336,
|
| 734 |
+
"logps/chosen": -67.23016082048416,
|
| 735 |
+
"logps/rejected": -102.15162888765335,
|
| 736 |
+
"epoch": 1.9544740973312402,
|
| 737 |
+
"step": 390
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"loss": 0.1598552346229553,
|
| 741 |
+
"grad_norm": 0.019359465688467026,
|
| 742 |
+
"learning_rate": 2.5000000000000002e-08,
|
| 743 |
+
"entropy": 0.6207083983764309,
|
| 744 |
+
"num_tokens": 5293556.0,
|
| 745 |
+
"logits/chosen": -2.3401597344127856,
|
| 746 |
+
"logits/rejected": -1.8184490606216015,
|
| 747 |
+
"mean_token_accuracy": 0.7158373423691453,
|
| 748 |
+
"rewards/chosen": -2.0682149272387975,
|
| 749 |
+
"rewards/rejected": -4.944644331932068,
|
| 750 |
+
"rewards/accuracies": 0.9517241379310345,
|
| 751 |
+
"rewards/margins": 2.8764294094052807,
|
| 752 |
+
"logps/chosen": -73.4231296276224,
|
| 753 |
+
"logps/rejected": -101.61757607295596,
|
| 754 |
+
"epoch": 2.0,
|
| 755 |
+
"step": 400
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"train_runtime": 1812.7376,
|
| 759 |
+
"train_samples_per_second": 3.514,
|
| 760 |
+
"train_steps_per_second": 0.221,
|
| 761 |
+
"total_flos": 4.273969110335078e+16,
|
| 762 |
+
"train_loss": 0.3211171677708626,
|
| 763 |
+
"epoch": 2.0,
|
| 764 |
+
"step": 400
|
| 765 |
+
}
|
| 766 |
+
],
|
| 767 |
+
"validation_monitor_size": 0,
|
| 768 |
+
"validation_monitor_selection": "fixed_seed_random_without_replacement",
|
| 769 |
+
"validation_monitor_seed": 22,
|
| 770 |
+
"validation_monitor_indices": [],
|
| 771 |
+
"early_stopping_patience": null
|
| 772 |
+
}
|