Upload folder using huggingface_hub
Browse files- 133_128_e3_3e-5/adapter_config.json +39 -0
- 133_128_e3_3e-5/adapter_model.safetensors +3 -0
- 133_128_e3_3e-5/added_tokens.json +9 -0
- 133_128_e3_3e-5/all_results.json +9 -0
- 133_128_e3_3e-5/chat_template.jinja +62 -0
- 133_128_e3_3e-5/config.json +32 -0
- 133_128_e3_3e-5/merges.txt +0 -0
- 133_128_e3_3e-5/special_tokens_map.json +33 -0
- 133_128_e3_3e-5/tokenizer.json +0 -0
- 133_128_e3_3e-5/tokenizer_config.json +234 -0
- 133_128_e3_3e-5/train_results.json +9 -0
- 133_128_e3_3e-5/trainer_state.json +2430 -0
- 133_128_e3_3e-5/training_args.bin +3 -0
- 133_128_e3_3e-5/vocab.json +0 -0
133_128_e3_3e-5/adapter_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "ibm-granite/granite-3.3-8b-instruct",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"corda_config": null,
|
| 7 |
+
"eva_config": null,
|
| 8 |
+
"exclude_modules": null,
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 256,
|
| 17 |
+
"lora_bias": false,
|
| 18 |
+
"lora_dropout": 0.05,
|
| 19 |
+
"megatron_config": null,
|
| 20 |
+
"megatron_core": "megatron.core",
|
| 21 |
+
"modules_to_save": null,
|
| 22 |
+
"peft_type": "LORA",
|
| 23 |
+
"r": 128,
|
| 24 |
+
"rank_pattern": {},
|
| 25 |
+
"revision": null,
|
| 26 |
+
"target_modules": [
|
| 27 |
+
"down_proj",
|
| 28 |
+
"q_proj",
|
| 29 |
+
"up_proj",
|
| 30 |
+
"k_proj",
|
| 31 |
+
"v_proj",
|
| 32 |
+
"gate_proj",
|
| 33 |
+
"o_proj"
|
| 34 |
+
],
|
| 35 |
+
"task_type": "CAUSAL_LM",
|
| 36 |
+
"trainable_token_indices": null,
|
| 37 |
+
"use_dora": false,
|
| 38 |
+
"use_rslora": false
|
| 39 |
+
}
|
133_128_e3_3e-5/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6808b380955b965b5aa0964a8773a857aa41039a9c03e3cea7d5370923976836
|
| 3 |
+
size 791751704
|
133_128_e3_3e-5/added_tokens.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|end_of_cite|>": 49156,
|
| 3 |
+
"<|end_of_plugin|>": 49158,
|
| 4 |
+
"<|end_of_role|>": 49153,
|
| 5 |
+
"<|start_of_cite|>": 49155,
|
| 6 |
+
"<|start_of_plugin|>": 49157,
|
| 7 |
+
"<|start_of_role|>": 49152,
|
| 8 |
+
"<|tool_call|>": 49154
|
| 9 |
+
}
|
133_128_e3_3e-5/all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.251473000899543e+18,
|
| 4 |
+
"train_loss": 0.3827222754505551,
|
| 5 |
+
"train_runtime": 1075.2003,
|
| 6 |
+
"train_samples": 18191,
|
| 7 |
+
"train_samples_per_second": 50.756,
|
| 8 |
+
"train_steps_per_second": 1.588
|
| 9 |
+
}
|
133_128_e3_3e-5/chat_template.jinja
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{# Alias tools -> available_tools #}
|
| 2 |
+
{%- if tools and not available_tools -%}
|
| 3 |
+
{%- set available_tools = tools -%}
|
| 4 |
+
{%- endif -%}
|
| 5 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 6 |
+
{%- set system_message = messages[0]['content'] %}
|
| 7 |
+
{%- set loop_messages = messages[1:] %}
|
| 8 |
+
{%- else %}
|
| 9 |
+
{%- set system_message = "Knowledge Cutoff Date: April 2024.
|
| 10 |
+
Today's Date: " + strftime_now('%B %d, %Y') + ".
|
| 11 |
+
You are Granite, developed by IBM." %}
|
| 12 |
+
{%- if available_tools and documents %}
|
| 13 |
+
{%- set system_message = system_message + " You are a helpful assistant with access to the following tools. When a tool is required to answer the user's query, respond only with <|tool_call|> followed by a JSON list of tools used. If a tool does not exist in the provided list of tools, notify the user that you do not have the ability to fulfill the request.
|
| 14 |
+
Write the response to the user's input by strictly aligning with the facts in the provided documents. If the information needed to answer the question is not available in the documents, inform the user that the question cannot be answered based on the available data." %}
|
| 15 |
+
{%- elif available_tools %}
|
| 16 |
+
{%- set system_message = system_message + " You are a helpful assistant with access to the following tools. When a tool is required to answer the user's query, respond only with <|tool_call|> followed by a JSON list of tools used. If a tool does not exist in the provided list of tools, notify the user that you do not have the ability to fulfill the request." %}
|
| 17 |
+
{%- elif documents %}
|
| 18 |
+
{%- set system_message = system_message + " Write the response to the user's input by strictly aligning with the facts in the provided documents. If the information needed to answer the question is not available in the documents, inform the user that the question cannot be answered based on the available data." %}
|
| 19 |
+
{%- elif thinking %}
|
| 20 |
+
{%- set system_message = system_message + " You are a helpful AI assistant.
|
| 21 |
+
Respond to every user query in a comprehensive and detailed way. You can write down your thoughts and reasoning process before responding. In the thought process, engage in a comprehensive cycle of analysis, summarization, exploration, reassessment, reflection, backtracing, and iteration to develop well-considered thinking process. In the response section, based on various attempts, explorations, and reflections from the thoughts section, systematically present the final solution that you deem correct. The response should summarize the thought process. Write your thoughts between <think></think> and write your response between <response></response> for each user query." %}
|
| 22 |
+
{%- else %}
|
| 23 |
+
{%- set system_message = system_message + " You are a helpful AI assistant." %}
|
| 24 |
+
{%- endif %}
|
| 25 |
+
{%- if 'citations' in controls and documents %}
|
| 26 |
+
{%- set system_message = system_message + '
|
| 27 |
+
Use the symbols <|start_of_cite|> and <|end_of_cite|> to indicate when a fact comes from a document in the search result, e.g <|start_of_cite|> {document_id: 1}my fact <|end_of_cite|> for a fact from document 1. Afterwards, list all the citations with their corresponding documents in an ordered list.' %}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{%- if 'hallucinations' in controls and documents %}
|
| 30 |
+
{%- set system_message = system_message + '
|
| 31 |
+
Finally, after the response is written, include a numbered list of sentences from the response with a corresponding risk value that are hallucinated and not based in the documents.' %}
|
| 32 |
+
{%- endif %}
|
| 33 |
+
{%- set loop_messages = messages %}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{{- '<|start_of_role|>system<|end_of_role|>' + system_message + '<|end_of_text|>
|
| 36 |
+
' }}
|
| 37 |
+
{%- if available_tools %}
|
| 38 |
+
{{- '<|start_of_role|>available_tools<|end_of_role|>' }}
|
| 39 |
+
{{- available_tools | tojson(indent=4) }}
|
| 40 |
+
{{- '<|end_of_text|>
|
| 41 |
+
' }}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{%- if documents %}
|
| 44 |
+
{%- for document in documents %}
|
| 45 |
+
{{- '<|start_of_role|>document {"document_id": "' + document['doc_id'] | string + '"}<|end_of_role|>
|
| 46 |
+
' }}
|
| 47 |
+
{{- document['text'] }}
|
| 48 |
+
{{- '<|end_of_text|>
|
| 49 |
+
' }}
|
| 50 |
+
{%- endfor %}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- for message in loop_messages %}
|
| 53 |
+
{{- '<|start_of_role|>' + message['role'] + '<|end_of_role|>' + message['content'] + '<|end_of_text|>
|
| 54 |
+
' }}
|
| 55 |
+
{%- if loop.last and add_generation_prompt %}
|
| 56 |
+
{{- '<|start_of_role|>assistant' }}
|
| 57 |
+
{%- if controls %}
|
| 58 |
+
{{- ' ' + controls | tojson()}}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|end_of_role|>' }}
|
| 61 |
+
{%- endif %}
|
| 62 |
+
{%- endfor %}
|
133_128_e3_3e-5/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"GraniteForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"attention_multiplier": 0.0078125,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"embedding_multiplier": 12.0,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 4096,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 12800,
|
| 15 |
+
"logits_scaling": 16.0,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "granite",
|
| 19 |
+
"num_attention_heads": 32,
|
| 20 |
+
"num_hidden_layers": 40,
|
| 21 |
+
"num_key_value_heads": 8,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"residual_multiplier": 0.22,
|
| 24 |
+
"rms_norm_eps": 1e-05,
|
| 25 |
+
"rope_scaling": null,
|
| 26 |
+
"rope_theta": 10000000.0,
|
| 27 |
+
"tie_word_embeddings": true,
|
| 28 |
+
"torch_dtype": "bfloat16",
|
| 29 |
+
"transformers_version": "4.52.4",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 49159
|
| 32 |
+
}
|
133_128_e3_3e-5/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
133_128_e3_3e-5/special_tokens_map.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|start_of_role|>",
|
| 4 |
+
"<|end_of_role|>",
|
| 5 |
+
"<|tool_call|>",
|
| 6 |
+
"<|start_of_cite|>",
|
| 7 |
+
"<|end_of_cite|>",
|
| 8 |
+
"<|start_of_plugin|>",
|
| 9 |
+
"<|end_of_plugin|>"
|
| 10 |
+
],
|
| 11 |
+
"bos_token": {
|
| 12 |
+
"content": "<|end_of_text|>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false
|
| 17 |
+
},
|
| 18 |
+
"eos_token": {
|
| 19 |
+
"content": "<|end_of_text|>",
|
| 20 |
+
"lstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"rstrip": false,
|
| 23 |
+
"single_word": false
|
| 24 |
+
},
|
| 25 |
+
"pad_token": "<|end_of_plugin|>",
|
| 26 |
+
"unk_token": {
|
| 27 |
+
"content": "<|end_of_text|>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": false,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
}
|
| 33 |
+
}
|
133_128_e3_3e-5/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
133_128_e3_3e-5/tokenizer_config.json
ADDED
|
@@ -0,0 +1,234 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<|end_of_text|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<fim_prefix>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "<fim_middle>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"3": {
|
| 30 |
+
"content": "<fim_suffix>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"4": {
|
| 38 |
+
"content": "<fim_pad>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"5": {
|
| 46 |
+
"content": "<filename>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"6": {
|
| 54 |
+
"content": "<gh_stars>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"7": {
|
| 62 |
+
"content": "<issue_start>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"8": {
|
| 70 |
+
"content": "<issue_comment>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"9": {
|
| 78 |
+
"content": "<issue_closed>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"10": {
|
| 86 |
+
"content": "<jupyter_start>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"11": {
|
| 94 |
+
"content": "<jupyter_text>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"12": {
|
| 102 |
+
"content": "<jupyter_code>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"13": {
|
| 110 |
+
"content": "<jupyter_output>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"14": {
|
| 118 |
+
"content": "<empty_output>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": true
|
| 124 |
+
},
|
| 125 |
+
"15": {
|
| 126 |
+
"content": "<commit_before>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": true
|
| 132 |
+
},
|
| 133 |
+
"16": {
|
| 134 |
+
"content": "<commit_msg>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": true
|
| 140 |
+
},
|
| 141 |
+
"17": {
|
| 142 |
+
"content": "<commit_after>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": true
|
| 148 |
+
},
|
| 149 |
+
"18": {
|
| 150 |
+
"content": "<reponame>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": true
|
| 156 |
+
},
|
| 157 |
+
"49152": {
|
| 158 |
+
"content": "<|start_of_role|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": true
|
| 164 |
+
},
|
| 165 |
+
"49153": {
|
| 166 |
+
"content": "<|end_of_role|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": true
|
| 172 |
+
},
|
| 173 |
+
"49154": {
|
| 174 |
+
"content": "<|tool_call|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": true
|
| 180 |
+
},
|
| 181 |
+
"49155": {
|
| 182 |
+
"content": "<|start_of_cite|>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": true
|
| 188 |
+
},
|
| 189 |
+
"49156": {
|
| 190 |
+
"content": "<|end_of_cite|>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": true
|
| 196 |
+
},
|
| 197 |
+
"49157": {
|
| 198 |
+
"content": "<|start_of_plugin|>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": true
|
| 204 |
+
},
|
| 205 |
+
"49158": {
|
| 206 |
+
"content": "<|end_of_plugin|>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": true
|
| 212 |
+
}
|
| 213 |
+
},
|
| 214 |
+
"additional_special_tokens": [
|
| 215 |
+
"<|start_of_role|>",
|
| 216 |
+
"<|end_of_role|>",
|
| 217 |
+
"<|tool_call|>",
|
| 218 |
+
"<|start_of_cite|>",
|
| 219 |
+
"<|end_of_cite|>",
|
| 220 |
+
"<|start_of_plugin|>",
|
| 221 |
+
"<|end_of_plugin|>"
|
| 222 |
+
],
|
| 223 |
+
"bos_token": "<|end_of_text|>",
|
| 224 |
+
"clean_up_tokenization_spaces": true,
|
| 225 |
+
"eos_token": "<|end_of_text|>",
|
| 226 |
+
"errors": "replace",
|
| 227 |
+
"extra_special_tokens": {},
|
| 228 |
+
"model_max_length": 8192,
|
| 229 |
+
"pad_token": "<|end_of_plugin|>",
|
| 230 |
+
"padding_side": "left",
|
| 231 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 232 |
+
"unk_token": "<|end_of_text|>",
|
| 233 |
+
"vocab_size": 49152
|
| 234 |
+
}
|
133_128_e3_3e-5/train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.251473000899543e+18,
|
| 4 |
+
"train_loss": 0.3827222754505551,
|
| 5 |
+
"train_runtime": 1075.2003,
|
| 6 |
+
"train_samples": 18191,
|
| 7 |
+
"train_samples_per_second": 50.756,
|
| 8 |
+
"train_steps_per_second": 1.588
|
| 9 |
+
}
|
133_128_e3_3e-5/trainer_state.json
ADDED
|
@@ -0,0 +1,2430 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1707,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008795074758135445,
|
| 14 |
+
"grad_norm": 2.3172574043273926,
|
| 15 |
+
"learning_rate": 1.3953488372093023e-06,
|
| 16 |
+
"loss": 1.6437,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.01759014951627089,
|
| 21 |
+
"grad_norm": 2.0111255645751953,
|
| 22 |
+
"learning_rate": 3.1395348837209307e-06,
|
| 23 |
+
"loss": 1.5798,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.026385224274406333,
|
| 28 |
+
"grad_norm": 1.5366212129592896,
|
| 29 |
+
"learning_rate": 4.883720930232559e-06,
|
| 30 |
+
"loss": 1.5641,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03518029903254178,
|
| 35 |
+
"grad_norm": 0.689163863658905,
|
| 36 |
+
"learning_rate": 6.627906976744187e-06,
|
| 37 |
+
"loss": 1.4937,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04397537379067722,
|
| 42 |
+
"grad_norm": 0.4792795181274414,
|
| 43 |
+
"learning_rate": 8.372093023255815e-06,
|
| 44 |
+
"loss": 1.4707,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.052770448548812667,
|
| 49 |
+
"grad_norm": 0.4393588602542877,
|
| 50 |
+
"learning_rate": 1.0116279069767442e-05,
|
| 51 |
+
"loss": 1.4469,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06156552330694811,
|
| 56 |
+
"grad_norm": 0.42879006266593933,
|
| 57 |
+
"learning_rate": 1.1860465116279069e-05,
|
| 58 |
+
"loss": 1.4175,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07036059806508356,
|
| 63 |
+
"grad_norm": 0.38449549674987793,
|
| 64 |
+
"learning_rate": 1.3604651162790698e-05,
|
| 65 |
+
"loss": 1.4829,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.079155672823219,
|
| 70 |
+
"grad_norm": 0.42500245571136475,
|
| 71 |
+
"learning_rate": 1.5348837209302328e-05,
|
| 72 |
+
"loss": 1.3551,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08795074758135445,
|
| 77 |
+
"grad_norm": 0.3925383687019348,
|
| 78 |
+
"learning_rate": 1.7093023255813955e-05,
|
| 79 |
+
"loss": 1.3672,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09674582233948989,
|
| 84 |
+
"grad_norm": 0.4350813925266266,
|
| 85 |
+
"learning_rate": 1.8837209302325582e-05,
|
| 86 |
+
"loss": 1.3511,
|
| 87 |
+
"step": 55
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10554089709762533,
|
| 91 |
+
"grad_norm": 0.4195137619972229,
|
| 92 |
+
"learning_rate": 2.058139534883721e-05,
|
| 93 |
+
"loss": 1.3704,
|
| 94 |
+
"step": 60
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.11433597185576078,
|
| 98 |
+
"grad_norm": 0.46276843547821045,
|
| 99 |
+
"learning_rate": 2.2325581395348837e-05,
|
| 100 |
+
"loss": 1.347,
|
| 101 |
+
"step": 65
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.12313104661389622,
|
| 105 |
+
"grad_norm": 0.45429137349128723,
|
| 106 |
+
"learning_rate": 2.4069767441860464e-05,
|
| 107 |
+
"loss": 1.3127,
|
| 108 |
+
"step": 70
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.13192612137203166,
|
| 112 |
+
"grad_norm": 0.4145204722881317,
|
| 113 |
+
"learning_rate": 2.5813953488372094e-05,
|
| 114 |
+
"loss": 1.3861,
|
| 115 |
+
"step": 75
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.14072119613016712,
|
| 119 |
+
"grad_norm": 0.3961552083492279,
|
| 120 |
+
"learning_rate": 2.755813953488372e-05,
|
| 121 |
+
"loss": 1.3232,
|
| 122 |
+
"step": 80
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.14951627088830255,
|
| 126 |
+
"grad_norm": 0.44028440117836,
|
| 127 |
+
"learning_rate": 2.930232558139535e-05,
|
| 128 |
+
"loss": 1.2873,
|
| 129 |
+
"step": 85
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.158311345646438,
|
| 133 |
+
"grad_norm": 0.4683926999568939,
|
| 134 |
+
"learning_rate": 2.9999746465966867e-05,
|
| 135 |
+
"loss": 1.2919,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.16710642040457344,
|
| 140 |
+
"grad_norm": 0.48119649291038513,
|
| 141 |
+
"learning_rate": 2.999819712235739e-05,
|
| 142 |
+
"loss": 1.2423,
|
| 143 |
+
"step": 95
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.1759014951627089,
|
| 147 |
+
"grad_norm": 0.5247952938079834,
|
| 148 |
+
"learning_rate": 2.9995239432687672e-05,
|
| 149 |
+
"loss": 1.3046,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.18469656992084432,
|
| 154 |
+
"grad_norm": 0.6130931377410889,
|
| 155 |
+
"learning_rate": 2.999087367468779e-05,
|
| 156 |
+
"loss": 1.2565,
|
| 157 |
+
"step": 105
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.19349164467897978,
|
| 161 |
+
"grad_norm": 0.5812133550643921,
|
| 162 |
+
"learning_rate": 2.9985100258306897e-05,
|
| 163 |
+
"loss": 1.2144,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.2022867194371152,
|
| 168 |
+
"grad_norm": 0.5057661533355713,
|
| 169 |
+
"learning_rate": 2.9977919725674697e-05,
|
| 170 |
+
"loss": 1.1772,
|
| 171 |
+
"step": 115
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.21108179419525067,
|
| 175 |
+
"grad_norm": 0.5832656025886536,
|
| 176 |
+
"learning_rate": 2.996933275105055e-05,
|
| 177 |
+
"loss": 1.2302,
|
| 178 |
+
"step": 120
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.2198768689533861,
|
| 182 |
+
"grad_norm": 0.5832039713859558,
|
| 183 |
+
"learning_rate": 2.995934014076017e-05,
|
| 184 |
+
"loss": 1.188,
|
| 185 |
+
"step": 125
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.22867194371152155,
|
| 189 |
+
"grad_norm": 0.7051421999931335,
|
| 190 |
+
"learning_rate": 2.9947942833119882e-05,
|
| 191 |
+
"loss": 1.1913,
|
| 192 |
+
"step": 130
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.23746701846965698,
|
| 196 |
+
"grad_norm": 0.5919679403305054,
|
| 197 |
+
"learning_rate": 2.993514189834855e-05,
|
| 198 |
+
"loss": 1.0842,
|
| 199 |
+
"step": 135
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.24626209322779244,
|
| 203 |
+
"grad_norm": 0.6445920467376709,
|
| 204 |
+
"learning_rate": 2.992093853846704e-05,
|
| 205 |
+
"loss": 1.0922,
|
| 206 |
+
"step": 140
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.25505716798592787,
|
| 210 |
+
"grad_norm": 0.6736803650856018,
|
| 211 |
+
"learning_rate": 2.9905334087185382e-05,
|
| 212 |
+
"loss": 1.1188,
|
| 213 |
+
"step": 145
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.2638522427440633,
|
| 217 |
+
"grad_norm": 0.7049084305763245,
|
| 218 |
+
"learning_rate": 2.9888330009777523e-05,
|
| 219 |
+
"loss": 1.08,
|
| 220 |
+
"step": 150
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2726473175021988,
|
| 224 |
+
"grad_norm": 0.7496076226234436,
|
| 225 |
+
"learning_rate": 2.9869927902943743e-05,
|
| 226 |
+
"loss": 1.0627,
|
| 227 |
+
"step": 155
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.28144239226033424,
|
| 231 |
+
"grad_norm": 0.7592334151268005,
|
| 232 |
+
"learning_rate": 2.9850129494660702e-05,
|
| 233 |
+
"loss": 1.0522,
|
| 234 |
+
"step": 160
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.29023746701846964,
|
| 238 |
+
"grad_norm": 0.7949243187904358,
|
| 239 |
+
"learning_rate": 2.9828936644019197e-05,
|
| 240 |
+
"loss": 1.0407,
|
| 241 |
+
"step": 165
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2990325417766051,
|
| 245 |
+
"grad_norm": 0.6946238875389099,
|
| 246 |
+
"learning_rate": 2.9806351341049602e-05,
|
| 247 |
+
"loss": 1.036,
|
| 248 |
+
"step": 170
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.30782761653474056,
|
| 252 |
+
"grad_norm": 0.7097417116165161,
|
| 253 |
+
"learning_rate": 2.9782375706534982e-05,
|
| 254 |
+
"loss": 0.9961,
|
| 255 |
+
"step": 175
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.316622691292876,
|
| 259 |
+
"grad_norm": 0.7381428480148315,
|
| 260 |
+
"learning_rate": 2.9757011991811945e-05,
|
| 261 |
+
"loss": 0.9881,
|
| 262 |
+
"step": 180
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3254177660510114,
|
| 266 |
+
"grad_norm": 0.8711695075035095,
|
| 267 |
+
"learning_rate": 2.9730262578559273e-05,
|
| 268 |
+
"loss": 0.9976,
|
| 269 |
+
"step": 185
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.33421284080914687,
|
| 273 |
+
"grad_norm": 0.816605269908905,
|
| 274 |
+
"learning_rate": 2.9702129978574248e-05,
|
| 275 |
+
"loss": 0.9711,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.34300791556728233,
|
| 280 |
+
"grad_norm": 0.8204250931739807,
|
| 281 |
+
"learning_rate": 2.96726168335368e-05,
|
| 282 |
+
"loss": 0.9536,
|
| 283 |
+
"step": 195
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3518029903254178,
|
| 287 |
+
"grad_norm": 0.7636730670928955,
|
| 288 |
+
"learning_rate": 2.9641725914761447e-05,
|
| 289 |
+
"loss": 0.9297,
|
| 290 |
+
"step": 200
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3605980650835532,
|
| 294 |
+
"grad_norm": 0.9243521094322205,
|
| 295 |
+
"learning_rate": 2.960946012293709e-05,
|
| 296 |
+
"loss": 0.9397,
|
| 297 |
+
"step": 205
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.36939313984168864,
|
| 301 |
+
"grad_norm": 0.9590564370155334,
|
| 302 |
+
"learning_rate": 2.9575822487854602e-05,
|
| 303 |
+
"loss": 0.9083,
|
| 304 |
+
"step": 210
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3781882145998241,
|
| 308 |
+
"grad_norm": 0.9289636015892029,
|
| 309 |
+
"learning_rate": 2.954081616812234e-05,
|
| 310 |
+
"loss": 0.9491,
|
| 311 |
+
"step": 215
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.38698328935795956,
|
| 315 |
+
"grad_norm": 0.852174699306488,
|
| 316 |
+
"learning_rate": 2.950444445086956e-05,
|
| 317 |
+
"loss": 0.855,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.39577836411609496,
|
| 322 |
+
"grad_norm": 0.945195734500885,
|
| 323 |
+
"learning_rate": 2.9466710751437738e-05,
|
| 324 |
+
"loss": 0.8854,
|
| 325 |
+
"step": 225
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.4045734388742304,
|
| 329 |
+
"grad_norm": 0.9900103211402893,
|
| 330 |
+
"learning_rate": 2.9427618613059867e-05,
|
| 331 |
+
"loss": 0.8586,
|
| 332 |
+
"step": 230
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.4133685136323659,
|
| 336 |
+
"grad_norm": 0.9707903265953064,
|
| 337 |
+
"learning_rate": 2.938717170652775e-05,
|
| 338 |
+
"loss": 0.8898,
|
| 339 |
+
"step": 235
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.42216358839050133,
|
| 343 |
+
"grad_norm": 0.9089756011962891,
|
| 344 |
+
"learning_rate": 2.9345373829847318e-05,
|
| 345 |
+
"loss": 0.8765,
|
| 346 |
+
"step": 240
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.4309586631486368,
|
| 350 |
+
"grad_norm": 1.1131296157836914,
|
| 351 |
+
"learning_rate": 2.9302228907881956e-05,
|
| 352 |
+
"loss": 0.8527,
|
| 353 |
+
"step": 245
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.4397537379067722,
|
| 357 |
+
"grad_norm": 1.08473801612854,
|
| 358 |
+
"learning_rate": 2.9257740991984007e-05,
|
| 359 |
+
"loss": 0.8472,
|
| 360 |
+
"step": 250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.44854881266490765,
|
| 364 |
+
"grad_norm": 1.0361857414245605,
|
| 365 |
+
"learning_rate": 2.92119142596143e-05,
|
| 366 |
+
"loss": 0.8416,
|
| 367 |
+
"step": 255
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.4573438874230431,
|
| 371 |
+
"grad_norm": 0.9746695756912231,
|
| 372 |
+
"learning_rate": 2.9164753013949908e-05,
|
| 373 |
+
"loss": 0.882,
|
| 374 |
+
"step": 260
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.46613896218117856,
|
| 378 |
+
"grad_norm": 1.0064749717712402,
|
| 379 |
+
"learning_rate": 2.911626168348007e-05,
|
| 380 |
+
"loss": 0.8178,
|
| 381 |
+
"step": 265
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.47493403693931396,
|
| 385 |
+
"grad_norm": 0.9450291991233826,
|
| 386 |
+
"learning_rate": 2.9066444821590345e-05,
|
| 387 |
+
"loss": 0.8397,
|
| 388 |
+
"step": 270
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4837291116974494,
|
| 392 |
+
"grad_norm": 1.0043177604675293,
|
| 393 |
+
"learning_rate": 2.9015307106135053e-05,
|
| 394 |
+
"loss": 0.8237,
|
| 395 |
+
"step": 275
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4925241864555849,
|
| 399 |
+
"grad_norm": 1.091946005821228,
|
| 400 |
+
"learning_rate": 2.896285333899802e-05,
|
| 401 |
+
"loss": 0.8131,
|
| 402 |
+
"step": 280
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.5013192612137203,
|
| 406 |
+
"grad_norm": 1.0148231983184814,
|
| 407 |
+
"learning_rate": 2.8909088445641653e-05,
|
| 408 |
+
"loss": 0.8007,
|
| 409 |
+
"step": 285
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5101143359718557,
|
| 413 |
+
"grad_norm": 1.1020045280456543,
|
| 414 |
+
"learning_rate": 2.885401747464447e-05,
|
| 415 |
+
"loss": 0.7936,
|
| 416 |
+
"step": 290
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5189094107299912,
|
| 420 |
+
"grad_norm": 1.1025718450546265,
|
| 421 |
+
"learning_rate": 2.8797645597227017e-05,
|
| 422 |
+
"loss": 0.7182,
|
| 423 |
+
"step": 295
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5277044854881267,
|
| 427 |
+
"grad_norm": 1.1578431129455566,
|
| 428 |
+
"learning_rate": 2.873997810676628e-05,
|
| 429 |
+
"loss": 0.758,
|
| 430 |
+
"step": 300
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5364995602462621,
|
| 434 |
+
"grad_norm": 1.0944195985794067,
|
| 435 |
+
"learning_rate": 2.8681020418298632e-05,
|
| 436 |
+
"loss": 0.7028,
|
| 437 |
+
"step": 305
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5452946350043976,
|
| 441 |
+
"grad_norm": 1.0901892185211182,
|
| 442 |
+
"learning_rate": 2.862077806801136e-05,
|
| 443 |
+
"loss": 0.7702,
|
| 444 |
+
"step": 310
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.554089709762533,
|
| 448 |
+
"grad_norm": 0.9799309372901917,
|
| 449 |
+
"learning_rate": 2.8559256712722818e-05,
|
| 450 |
+
"loss": 0.7511,
|
| 451 |
+
"step": 315
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5628847845206685,
|
| 455 |
+
"grad_norm": 1.1476151943206787,
|
| 456 |
+
"learning_rate": 2.8496462129351225e-05,
|
| 457 |
+
"loss": 0.6869,
|
| 458 |
+
"step": 320
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5716798592788038,
|
| 462 |
+
"grad_norm": 1.0580790042877197,
|
| 463 |
+
"learning_rate": 2.8432400214372244e-05,
|
| 464 |
+
"loss": 0.7467,
|
| 465 |
+
"step": 325
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5804749340369393,
|
| 469 |
+
"grad_norm": 1.0649679899215698,
|
| 470 |
+
"learning_rate": 2.8367076983265247e-05,
|
| 471 |
+
"loss": 0.6765,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5892700087950747,
|
| 476 |
+
"grad_norm": 1.1611366271972656,
|
| 477 |
+
"learning_rate": 2.8300498569948493e-05,
|
| 478 |
+
"loss": 0.6556,
|
| 479 |
+
"step": 335
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5980650835532102,
|
| 483 |
+
"grad_norm": 1.144186019897461,
|
| 484 |
+
"learning_rate": 2.8232671226203143e-05,
|
| 485 |
+
"loss": 0.7133,
|
| 486 |
+
"step": 340
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.6068601583113457,
|
| 490 |
+
"grad_norm": 1.2232329845428467,
|
| 491 |
+
"learning_rate": 2.816360132108618e-05,
|
| 492 |
+
"loss": 0.6986,
|
| 493 |
+
"step": 345
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6156552330694811,
|
| 497 |
+
"grad_norm": 1.1806381940841675,
|
| 498 |
+
"learning_rate": 2.80932953403324e-05,
|
| 499 |
+
"loss": 0.6784,
|
| 500 |
+
"step": 350
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6244503078276166,
|
| 504 |
+
"grad_norm": 1.061988115310669,
|
| 505 |
+
"learning_rate": 2.802175988574535e-05,
|
| 506 |
+
"loss": 0.6563,
|
| 507 |
+
"step": 355
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.633245382585752,
|
| 511 |
+
"grad_norm": 1.1837117671966553,
|
| 512 |
+
"learning_rate": 2.7949001674577424e-05,
|
| 513 |
+
"loss": 0.7434,
|
| 514 |
+
"step": 360
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6420404573438874,
|
| 518 |
+
"grad_norm": 1.1339837312698364,
|
| 519 |
+
"learning_rate": 2.787502753889913e-05,
|
| 520 |
+
"loss": 0.6089,
|
| 521 |
+
"step": 365
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.6508355321020228,
|
| 525 |
+
"grad_norm": 1.0632498264312744,
|
| 526 |
+
"learning_rate": 2.779984442495751e-05,
|
| 527 |
+
"loss": 0.6433,
|
| 528 |
+
"step": 370
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6596306068601583,
|
| 532 |
+
"grad_norm": 1.2514731884002686,
|
| 533 |
+
"learning_rate": 2.7723459392523912e-05,
|
| 534 |
+
"loss": 0.6439,
|
| 535 |
+
"step": 375
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6684256816182937,
|
| 539 |
+
"grad_norm": 1.1749696731567383,
|
| 540 |
+
"learning_rate": 2.764587961423107e-05,
|
| 541 |
+
"loss": 0.668,
|
| 542 |
+
"step": 380
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6772207563764292,
|
| 546 |
+
"grad_norm": 1.2721859216690063,
|
| 547 |
+
"learning_rate": 2.7567112374899563e-05,
|
| 548 |
+
"loss": 0.6273,
|
| 549 |
+
"step": 385
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6860158311345647,
|
| 553 |
+
"grad_norm": 1.1397324800491333,
|
| 554 |
+
"learning_rate": 2.748716507085377e-05,
|
| 555 |
+
"loss": 0.5531,
|
| 556 |
+
"step": 390
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6948109058927001,
|
| 560 |
+
"grad_norm": 1.2371467351913452,
|
| 561 |
+
"learning_rate": 2.740604520922738e-05,
|
| 562 |
+
"loss": 0.6199,
|
| 563 |
+
"step": 395
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7036059806508356,
|
| 567 |
+
"grad_norm": 1.2496663331985474,
|
| 568 |
+
"learning_rate": 2.7323760407258426e-05,
|
| 569 |
+
"loss": 0.6362,
|
| 570 |
+
"step": 400
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.712401055408971,
|
| 574 |
+
"grad_norm": 1.1109861135482788,
|
| 575 |
+
"learning_rate": 2.724031839157402e-05,
|
| 576 |
+
"loss": 0.5899,
|
| 577 |
+
"step": 405
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.7211961301671064,
|
| 581 |
+
"grad_norm": 1.3056610822677612,
|
| 582 |
+
"learning_rate": 2.7155726997464843e-05,
|
| 583 |
+
"loss": 0.5928,
|
| 584 |
+
"step": 410
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7299912049252418,
|
| 588 |
+
"grad_norm": 1.0721070766448975,
|
| 589 |
+
"learning_rate": 2.706999416814938e-05,
|
| 590 |
+
"loss": 0.5649,
|
| 591 |
+
"step": 415
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7387862796833773,
|
| 595 |
+
"grad_norm": 1.2519431114196777,
|
| 596 |
+
"learning_rate": 2.698312795402804e-05,
|
| 597 |
+
"loss": 0.5888,
|
| 598 |
+
"step": 420
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7475813544415127,
|
| 602 |
+
"grad_norm": 1.0946921110153198,
|
| 603 |
+
"learning_rate": 2.6895136511927253e-05,
|
| 604 |
+
"loss": 0.5633,
|
| 605 |
+
"step": 425
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.7563764291996482,
|
| 609 |
+
"grad_norm": 1.158353567123413,
|
| 610 |
+
"learning_rate": 2.680602810433348e-05,
|
| 611 |
+
"loss": 0.607,
|
| 612 |
+
"step": 430
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7651715039577837,
|
| 616 |
+
"grad_norm": 1.1864335536956787,
|
| 617 |
+
"learning_rate": 2.6715811098617392e-05,
|
| 618 |
+
"loss": 0.5726,
|
| 619 |
+
"step": 435
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7739665787159191,
|
| 623 |
+
"grad_norm": 1.268796682357788,
|
| 624 |
+
"learning_rate": 2.6624493966248172e-05,
|
| 625 |
+
"loss": 0.5444,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7827616534740546,
|
| 630 |
+
"grad_norm": 1.2135546207427979,
|
| 631 |
+
"learning_rate": 2.6532085281998002e-05,
|
| 632 |
+
"loss": 0.5603,
|
| 633 |
+
"step": 445
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7915567282321899,
|
| 637 |
+
"grad_norm": 1.2693135738372803,
|
| 638 |
+
"learning_rate": 2.6438593723136922e-05,
|
| 639 |
+
"loss": 0.5302,
|
| 640 |
+
"step": 450
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8003518029903254,
|
| 644 |
+
"grad_norm": 1.2406128644943237,
|
| 645 |
+
"learning_rate": 2.6344028068617984e-05,
|
| 646 |
+
"loss": 0.5396,
|
| 647 |
+
"step": 455
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.8091468777484608,
|
| 651 |
+
"grad_norm": 1.2119282484054565,
|
| 652 |
+
"learning_rate": 2.6248397198252943e-05,
|
| 653 |
+
"loss": 0.5345,
|
| 654 |
+
"step": 460
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.8179419525065963,
|
| 658 |
+
"grad_norm": 1.4239540100097656,
|
| 659 |
+
"learning_rate": 2.6151710091878395e-05,
|
| 660 |
+
"loss": 0.5026,
|
| 661 |
+
"step": 465
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.8267370272647317,
|
| 665 |
+
"grad_norm": 1.332635760307312,
|
| 666 |
+
"learning_rate": 2.60539758285126e-05,
|
| 667 |
+
"loss": 0.5343,
|
| 668 |
+
"step": 470
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.8355321020228672,
|
| 672 |
+
"grad_norm": 1.3118253946304321,
|
| 673 |
+
"learning_rate": 2.5955203585502902e-05,
|
| 674 |
+
"loss": 0.5229,
|
| 675 |
+
"step": 475
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.8443271767810027,
|
| 679 |
+
"grad_norm": 1.2909690141677856,
|
| 680 |
+
"learning_rate": 2.5855402637664026e-05,
|
| 681 |
+
"loss": 0.508,
|
| 682 |
+
"step": 480
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8531222515391381,
|
| 686 |
+
"grad_norm": 1.2916340827941895,
|
| 687 |
+
"learning_rate": 2.575458235640711e-05,
|
| 688 |
+
"loss": 0.4559,
|
| 689 |
+
"step": 485
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8619173262972736,
|
| 693 |
+
"grad_norm": 1.1892061233520508,
|
| 694 |
+
"learning_rate": 2.5652752208859753e-05,
|
| 695 |
+
"loss": 0.5089,
|
| 696 |
+
"step": 490
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.8707124010554089,
|
| 700 |
+
"grad_norm": 1.3271726369857788,
|
| 701 |
+
"learning_rate": 2.5549921756977036e-05,
|
| 702 |
+
"loss": 0.5063,
|
| 703 |
+
"step": 495
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.8795074758135444,
|
| 707 |
+
"grad_norm": 1.41947603225708,
|
| 708 |
+
"learning_rate": 2.5446100656643638e-05,
|
| 709 |
+
"loss": 0.4628,
|
| 710 |
+
"step": 500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8883025505716798,
|
| 714 |
+
"grad_norm": 1.3488316535949707,
|
| 715 |
+
"learning_rate": 2.5341298656767123e-05,
|
| 716 |
+
"loss": 0.4831,
|
| 717 |
+
"step": 505
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8970976253298153,
|
| 721 |
+
"grad_norm": 1.2644160985946655,
|
| 722 |
+
"learning_rate": 2.5235525598362564e-05,
|
| 723 |
+
"loss": 0.4533,
|
| 724 |
+
"step": 510
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.9058927000879508,
|
| 728 |
+
"grad_norm": 1.3000518083572388,
|
| 729 |
+
"learning_rate": 2.5128791413628395e-05,
|
| 730 |
+
"loss": 0.4682,
|
| 731 |
+
"step": 515
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.9146877748460862,
|
| 735 |
+
"grad_norm": 1.3574652671813965,
|
| 736 |
+
"learning_rate": 2.50211061250138e-05,
|
| 737 |
+
"loss": 0.4433,
|
| 738 |
+
"step": 520
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.9234828496042217,
|
| 742 |
+
"grad_norm": 1.3162592649459839,
|
| 743 |
+
"learning_rate": 2.4912479844277595e-05,
|
| 744 |
+
"loss": 0.4675,
|
| 745 |
+
"step": 525
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.9322779243623571,
|
| 749 |
+
"grad_norm": 1.3580745458602905,
|
| 750 |
+
"learning_rate": 2.480292277153872e-05,
|
| 751 |
+
"loss": 0.4444,
|
| 752 |
+
"step": 530
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.9410729991204925,
|
| 756 |
+
"grad_norm": 1.2309595346450806,
|
| 757 |
+
"learning_rate": 2.4692445194318442e-05,
|
| 758 |
+
"loss": 0.4235,
|
| 759 |
+
"step": 535
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.9498680738786279,
|
| 763 |
+
"grad_norm": 1.1799525022506714,
|
| 764 |
+
"learning_rate": 2.4581057486574322e-05,
|
| 765 |
+
"loss": 0.4279,
|
| 766 |
+
"step": 540
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.9586631486367634,
|
| 770 |
+
"grad_norm": 1.3797575235366821,
|
| 771 |
+
"learning_rate": 2.446877010772613e-05,
|
| 772 |
+
"loss": 0.4084,
|
| 773 |
+
"step": 545
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.9674582233948988,
|
| 777 |
+
"grad_norm": 1.371621012687683,
|
| 778 |
+
"learning_rate": 2.435559360167366e-05,
|
| 779 |
+
"loss": 0.4281,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9762532981530343,
|
| 784 |
+
"grad_norm": 1.2333793640136719,
|
| 785 |
+
"learning_rate": 2.424153859580667e-05,
|
| 786 |
+
"loss": 0.4242,
|
| 787 |
+
"step": 555
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.9850483729111698,
|
| 791 |
+
"grad_norm": 1.423797607421875,
|
| 792 |
+
"learning_rate": 2.412661580000694e-05,
|
| 793 |
+
"loss": 0.4284,
|
| 794 |
+
"step": 560
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.9938434476693052,
|
| 798 |
+
"grad_norm": 1.3837473392486572,
|
| 799 |
+
"learning_rate": 2.4010836005642617e-05,
|
| 800 |
+
"loss": 0.4253,
|
| 801 |
+
"step": 565
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.0017590149516271,
|
| 805 |
+
"grad_norm": 1.1022980213165283,
|
| 806 |
+
"learning_rate": 2.3894210084554894e-05,
|
| 807 |
+
"loss": 0.4032,
|
| 808 |
+
"step": 570
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.0105540897097625,
|
| 812 |
+
"grad_norm": 1.429387092590332,
|
| 813 |
+
"learning_rate": 2.3776748988037126e-05,
|
| 814 |
+
"loss": 0.3836,
|
| 815 |
+
"step": 575
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.019349164467898,
|
| 819 |
+
"grad_norm": 1.2948155403137207,
|
| 820 |
+
"learning_rate": 2.3658463745806495e-05,
|
| 821 |
+
"loss": 0.3615,
|
| 822 |
+
"step": 580
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.0281442392260334,
|
| 826 |
+
"grad_norm": 1.4466732740402222,
|
| 827 |
+
"learning_rate": 2.353936546496831e-05,
|
| 828 |
+
"loss": 0.337,
|
| 829 |
+
"step": 585
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.036939313984169,
|
| 833 |
+
"grad_norm": 1.1502208709716797,
|
| 834 |
+
"learning_rate": 2.3419465328973034e-05,
|
| 835 |
+
"loss": 0.3167,
|
| 836 |
+
"step": 590
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.0457343887423043,
|
| 840 |
+
"grad_norm": 1.1986420154571533,
|
| 841 |
+
"learning_rate": 2.329877459656616e-05,
|
| 842 |
+
"loss": 0.3596,
|
| 843 |
+
"step": 595
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.0545294635004399,
|
| 847 |
+
"grad_norm": 1.5961248874664307,
|
| 848 |
+
"learning_rate": 2.317730460073098e-05,
|
| 849 |
+
"loss": 0.3258,
|
| 850 |
+
"step": 600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.0633245382585752,
|
| 854 |
+
"grad_norm": 1.2916852235794067,
|
| 855 |
+
"learning_rate": 2.3055066747624428e-05,
|
| 856 |
+
"loss": 0.2901,
|
| 857 |
+
"step": 605
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.0721196130167105,
|
| 861 |
+
"grad_norm": 1.2745351791381836,
|
| 862 |
+
"learning_rate": 2.293207251550602e-05,
|
| 863 |
+
"loss": 0.3033,
|
| 864 |
+
"step": 610
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.0809146877748461,
|
| 868 |
+
"grad_norm": 1.356088638305664,
|
| 869 |
+
"learning_rate": 2.2808333453660035e-05,
|
| 870 |
+
"loss": 0.3566,
|
| 871 |
+
"step": 615
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.0897097625329815,
|
| 875 |
+
"grad_norm": 1.396205186843872,
|
| 876 |
+
"learning_rate": 2.268386118131103e-05,
|
| 877 |
+
"loss": 0.3051,
|
| 878 |
+
"step": 620
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.098504837291117,
|
| 882 |
+
"grad_norm": 1.4750607013702393,
|
| 883 |
+
"learning_rate": 2.2558667386532767e-05,
|
| 884 |
+
"loss": 0.3167,
|
| 885 |
+
"step": 625
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.1072999120492524,
|
| 889 |
+
"grad_norm": 1.4152711629867554,
|
| 890 |
+
"learning_rate": 2.243276382515071e-05,
|
| 891 |
+
"loss": 0.3317,
|
| 892 |
+
"step": 630
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.116094986807388,
|
| 896 |
+
"grad_norm": 1.2703274488449097,
|
| 897 |
+
"learning_rate": 2.230616231963813e-05,
|
| 898 |
+
"loss": 0.2994,
|
| 899 |
+
"step": 635
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.1248900615655233,
|
| 903 |
+
"grad_norm": 1.347488522529602,
|
| 904 |
+
"learning_rate": 2.2178874758005987e-05,
|
| 905 |
+
"loss": 0.3119,
|
| 906 |
+
"step": 640
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.1336851363236589,
|
| 910 |
+
"grad_norm": 1.2635931968688965,
|
| 911 |
+
"learning_rate": 2.2050913092686585e-05,
|
| 912 |
+
"loss": 0.321,
|
| 913 |
+
"step": 645
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.1424802110817942,
|
| 917 |
+
"grad_norm": 1.222844123840332,
|
| 918 |
+
"learning_rate": 2.1922289339411254e-05,
|
| 919 |
+
"loss": 0.3098,
|
| 920 |
+
"step": 650
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.1512752858399296,
|
| 924 |
+
"grad_norm": 1.4295434951782227,
|
| 925 |
+
"learning_rate": 2.1793015576082087e-05,
|
| 926 |
+
"loss": 0.317,
|
| 927 |
+
"step": 655
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.1600703605980651,
|
| 931 |
+
"grad_norm": 1.3431787490844727,
|
| 932 |
+
"learning_rate": 2.1663103941637763e-05,
|
| 933 |
+
"loss": 0.2871,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.1688654353562005,
|
| 938 |
+
"grad_norm": 1.477893352508545,
|
| 939 |
+
"learning_rate": 2.1532566634913717e-05,
|
| 940 |
+
"loss": 0.3195,
|
| 941 |
+
"step": 665
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.177660510114336,
|
| 945 |
+
"grad_norm": 1.4877034425735474,
|
| 946 |
+
"learning_rate": 2.1401415913496667e-05,
|
| 947 |
+
"loss": 0.3015,
|
| 948 |
+
"step": 670
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.1864555848724714,
|
| 952 |
+
"grad_norm": 1.3660948276519775,
|
| 953 |
+
"learning_rate": 2.1269664092573568e-05,
|
| 954 |
+
"loss": 0.2928,
|
| 955 |
+
"step": 675
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.195250659630607,
|
| 959 |
+
"grad_norm": 1.2822529077529907,
|
| 960 |
+
"learning_rate": 2.113732354377526e-05,
|
| 961 |
+
"loss": 0.2897,
|
| 962 |
+
"step": 680
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.2040457343887423,
|
| 966 |
+
"grad_norm": 1.3543392419815063,
|
| 967 |
+
"learning_rate": 2.1004406694014722e-05,
|
| 968 |
+
"loss": 0.281,
|
| 969 |
+
"step": 685
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.2128408091468779,
|
| 973 |
+
"grad_norm": 1.135543704032898,
|
| 974 |
+
"learning_rate": 2.087092602432018e-05,
|
| 975 |
+
"loss": 0.2819,
|
| 976 |
+
"step": 690
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.2216358839050132,
|
| 980 |
+
"grad_norm": 1.2797603607177734,
|
| 981 |
+
"learning_rate": 2.0736894068663145e-05,
|
| 982 |
+
"loss": 0.3164,
|
| 983 |
+
"step": 695
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.2304309586631486,
|
| 987 |
+
"grad_norm": 1.3295880556106567,
|
| 988 |
+
"learning_rate": 2.060232341278143e-05,
|
| 989 |
+
"loss": 0.2832,
|
| 990 |
+
"step": 700
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.2392260334212841,
|
| 994 |
+
"grad_norm": 1.2211233377456665,
|
| 995 |
+
"learning_rate": 2.0467226692997347e-05,
|
| 996 |
+
"loss": 0.3079,
|
| 997 |
+
"step": 705
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.2480211081794195,
|
| 1001 |
+
"grad_norm": 1.338828206062317,
|
| 1002 |
+
"learning_rate": 2.0331616595031156e-05,
|
| 1003 |
+
"loss": 0.2881,
|
| 1004 |
+
"step": 710
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.256816182937555,
|
| 1008 |
+
"grad_norm": 1.3628044128417969,
|
| 1009 |
+
"learning_rate": 2.0195505852809855e-05,
|
| 1010 |
+
"loss": 0.2857,
|
| 1011 |
+
"step": 715
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.2656112576956904,
|
| 1015 |
+
"grad_norm": 1.2952430248260498,
|
| 1016 |
+
"learning_rate": 2.0058907247271446e-05,
|
| 1017 |
+
"loss": 0.2628,
|
| 1018 |
+
"step": 720
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.2744063324538257,
|
| 1022 |
+
"grad_norm": 1.4404287338256836,
|
| 1023 |
+
"learning_rate": 1.9921833605164786e-05,
|
| 1024 |
+
"loss": 0.2784,
|
| 1025 |
+
"step": 725
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.2832014072119613,
|
| 1029 |
+
"grad_norm": 1.408872365951538,
|
| 1030 |
+
"learning_rate": 1.9784297797845176e-05,
|
| 1031 |
+
"loss": 0.2665,
|
| 1032 |
+
"step": 730
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.2919964819700969,
|
| 1036 |
+
"grad_norm": 1.2583457231521606,
|
| 1037 |
+
"learning_rate": 1.9646312740065684e-05,
|
| 1038 |
+
"loss": 0.2683,
|
| 1039 |
+
"step": 735
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.3007915567282322,
|
| 1043 |
+
"grad_norm": 1.33659029006958,
|
| 1044 |
+
"learning_rate": 1.9507891388764452e-05,
|
| 1045 |
+
"loss": 0.2638,
|
| 1046 |
+
"step": 740
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.3095866314863676,
|
| 1050 |
+
"grad_norm": 1.3514808416366577,
|
| 1051 |
+
"learning_rate": 1.9369046741848053e-05,
|
| 1052 |
+
"loss": 0.2404,
|
| 1053 |
+
"step": 745
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.3183817062445031,
|
| 1057 |
+
"grad_norm": 1.512386679649353,
|
| 1058 |
+
"learning_rate": 1.9229791836970937e-05,
|
| 1059 |
+
"loss": 0.2477,
|
| 1060 |
+
"step": 750
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.3271767810026385,
|
| 1064 |
+
"grad_norm": 1.15213942527771,
|
| 1065 |
+
"learning_rate": 1.9090139750311185e-05,
|
| 1066 |
+
"loss": 0.2301,
|
| 1067 |
+
"step": 755
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.335971855760774,
|
| 1071 |
+
"grad_norm": 1.336578130722046,
|
| 1072 |
+
"learning_rate": 1.8950103595342685e-05,
|
| 1073 |
+
"loss": 0.2658,
|
| 1074 |
+
"step": 760
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.3447669305189094,
|
| 1078 |
+
"grad_norm": 1.3252836465835571,
|
| 1079 |
+
"learning_rate": 1.8809696521603698e-05,
|
| 1080 |
+
"loss": 0.2523,
|
| 1081 |
+
"step": 765
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.3535620052770447,
|
| 1085 |
+
"grad_norm": 1.1670089960098267,
|
| 1086 |
+
"learning_rate": 1.866893171346216e-05,
|
| 1087 |
+
"loss": 0.2305,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.3623570800351803,
|
| 1092 |
+
"grad_norm": 1.299018383026123,
|
| 1093 |
+
"learning_rate": 1.8527822388877623e-05,
|
| 1094 |
+
"loss": 0.2599,
|
| 1095 |
+
"step": 775
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.3711521547933159,
|
| 1099 |
+
"grad_norm": 1.4729745388031006,
|
| 1100 |
+
"learning_rate": 1.838638179816009e-05,
|
| 1101 |
+
"loss": 0.228,
|
| 1102 |
+
"step": 780
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.3799472295514512,
|
| 1106 |
+
"grad_norm": 1.3241815567016602,
|
| 1107 |
+
"learning_rate": 1.8244623222725797e-05,
|
| 1108 |
+
"loss": 0.2231,
|
| 1109 |
+
"step": 785
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.3887423043095866,
|
| 1113 |
+
"grad_norm": 1.3604726791381836,
|
| 1114 |
+
"learning_rate": 1.810255997385006e-05,
|
| 1115 |
+
"loss": 0.234,
|
| 1116 |
+
"step": 790
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.3975373790677221,
|
| 1120 |
+
"grad_norm": 1.3357797861099243,
|
| 1121 |
+
"learning_rate": 1.7960205391417345e-05,
|
| 1122 |
+
"loss": 0.2288,
|
| 1123 |
+
"step": 795
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.4063324538258575,
|
| 1127 |
+
"grad_norm": 1.3327202796936035,
|
| 1128 |
+
"learning_rate": 1.7817572842668648e-05,
|
| 1129 |
+
"loss": 0.2214,
|
| 1130 |
+
"step": 800
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.415127528583993,
|
| 1134 |
+
"grad_norm": 1.280179500579834,
|
| 1135 |
+
"learning_rate": 1.7674675720946276e-05,
|
| 1136 |
+
"loss": 0.2395,
|
| 1137 |
+
"step": 805
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.4239226033421284,
|
| 1141 |
+
"grad_norm": 1.3980705738067627,
|
| 1142 |
+
"learning_rate": 1.7531527444436216e-05,
|
| 1143 |
+
"loss": 0.2327,
|
| 1144 |
+
"step": 810
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.4327176781002637,
|
| 1148 |
+
"grad_norm": 1.2538130283355713,
|
| 1149 |
+
"learning_rate": 1.738814145490813e-05,
|
| 1150 |
+
"loss": 0.2139,
|
| 1151 |
+
"step": 815
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.4415127528583993,
|
| 1155 |
+
"grad_norm": 1.2887780666351318,
|
| 1156 |
+
"learning_rate": 1.7244531216453204e-05,
|
| 1157 |
+
"loss": 0.2048,
|
| 1158 |
+
"step": 820
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.4503078276165349,
|
| 1162 |
+
"grad_norm": 1.3535557985305786,
|
| 1163 |
+
"learning_rate": 1.7100710214219804e-05,
|
| 1164 |
+
"loss": 0.2451,
|
| 1165 |
+
"step": 825
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.4591029023746702,
|
| 1169 |
+
"grad_norm": 1.202744960784912,
|
| 1170 |
+
"learning_rate": 1.695669195314723e-05,
|
| 1171 |
+
"loss": 0.2076,
|
| 1172 |
+
"step": 830
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.4678979771328056,
|
| 1176 |
+
"grad_norm": 1.6135321855545044,
|
| 1177 |
+
"learning_rate": 1.681248995669762e-05,
|
| 1178 |
+
"loss": 0.2243,
|
| 1179 |
+
"step": 835
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.4766930518909411,
|
| 1183 |
+
"grad_norm": 1.158492088317871,
|
| 1184 |
+
"learning_rate": 1.6668117765586013e-05,
|
| 1185 |
+
"loss": 0.2025,
|
| 1186 |
+
"step": 840
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.4854881266490765,
|
| 1190 |
+
"grad_norm": 1.2682121992111206,
|
| 1191 |
+
"learning_rate": 1.6523588936508916e-05,
|
| 1192 |
+
"loss": 0.2075,
|
| 1193 |
+
"step": 845
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.494283201407212,
|
| 1197 |
+
"grad_norm": 1.224328637123108,
|
| 1198 |
+
"learning_rate": 1.6378917040871315e-05,
|
| 1199 |
+
"loss": 0.2329,
|
| 1200 |
+
"step": 850
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.5030782761653474,
|
| 1204 |
+
"grad_norm": 1.221476674079895,
|
| 1205 |
+
"learning_rate": 1.623411566351227e-05,
|
| 1206 |
+
"loss": 0.2073,
|
| 1207 |
+
"step": 855
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.5118733509234827,
|
| 1211 |
+
"grad_norm": 1.2146347761154175,
|
| 1212 |
+
"learning_rate": 1.608919840142932e-05,
|
| 1213 |
+
"loss": 0.2142,
|
| 1214 |
+
"step": 860
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.5206684256816183,
|
| 1218 |
+
"grad_norm": 1.1274282932281494,
|
| 1219 |
+
"learning_rate": 1.5944178862501692e-05,
|
| 1220 |
+
"loss": 0.1875,
|
| 1221 |
+
"step": 865
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.5294635004397539,
|
| 1225 |
+
"grad_norm": 1.1922705173492432,
|
| 1226 |
+
"learning_rate": 1.5799070664212528e-05,
|
| 1227 |
+
"loss": 0.2007,
|
| 1228 |
+
"step": 870
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.5382585751978892,
|
| 1232 |
+
"grad_norm": 1.0983917713165283,
|
| 1233 |
+
"learning_rate": 1.565388743237015e-05,
|
| 1234 |
+
"loss": 0.1897,
|
| 1235 |
+
"step": 875
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.5470536499560246,
|
| 1239 |
+
"grad_norm": 1.0871819257736206,
|
| 1240 |
+
"learning_rate": 1.550864279982863e-05,
|
| 1241 |
+
"loss": 0.181,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.55584872471416,
|
| 1246 |
+
"grad_norm": 1.2087064981460571,
|
| 1247 |
+
"learning_rate": 1.5363350405207615e-05,
|
| 1248 |
+
"loss": 0.199,
|
| 1249 |
+
"step": 885
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.5646437994722955,
|
| 1253 |
+
"grad_norm": 1.1849874258041382,
|
| 1254 |
+
"learning_rate": 1.5218023891611668e-05,
|
| 1255 |
+
"loss": 0.1789,
|
| 1256 |
+
"step": 890
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.573438874230431,
|
| 1260 |
+
"grad_norm": 1.1807067394256592,
|
| 1261 |
+
"learning_rate": 1.5072676905349155e-05,
|
| 1262 |
+
"loss": 0.1833,
|
| 1263 |
+
"step": 895
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.5822339489885664,
|
| 1267 |
+
"grad_norm": 2.5398900508880615,
|
| 1268 |
+
"learning_rate": 1.492732309465085e-05,
|
| 1269 |
+
"loss": 0.1929,
|
| 1270 |
+
"step": 900
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.5910290237467017,
|
| 1274 |
+
"grad_norm": 1.0452604293823242,
|
| 1275 |
+
"learning_rate": 1.478197610838833e-05,
|
| 1276 |
+
"loss": 0.1864,
|
| 1277 |
+
"step": 905
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.5998240985048373,
|
| 1281 |
+
"grad_norm": 1.291452407836914,
|
| 1282 |
+
"learning_rate": 1.4636649594792386e-05,
|
| 1283 |
+
"loss": 0.1756,
|
| 1284 |
+
"step": 910
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.6086191732629729,
|
| 1288 |
+
"grad_norm": 1.181359887123108,
|
| 1289 |
+
"learning_rate": 1.4491357200171374e-05,
|
| 1290 |
+
"loss": 0.1661,
|
| 1291 |
+
"step": 915
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.6174142480211082,
|
| 1295 |
+
"grad_norm": 0.9952807426452637,
|
| 1296 |
+
"learning_rate": 1.4346112567629849e-05,
|
| 1297 |
+
"loss": 0.182,
|
| 1298 |
+
"step": 920
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.6262093227792436,
|
| 1302 |
+
"grad_norm": 1.1589791774749756,
|
| 1303 |
+
"learning_rate": 1.4200929335787475e-05,
|
| 1304 |
+
"loss": 0.1641,
|
| 1305 |
+
"step": 925
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.635004397537379,
|
| 1309 |
+
"grad_norm": 1.1980968713760376,
|
| 1310 |
+
"learning_rate": 1.4055821137498309e-05,
|
| 1311 |
+
"loss": 0.1799,
|
| 1312 |
+
"step": 930
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.6437994722955145,
|
| 1316 |
+
"grad_norm": 1.3150545358657837,
|
| 1317 |
+
"learning_rate": 1.391080159857068e-05,
|
| 1318 |
+
"loss": 0.1796,
|
| 1319 |
+
"step": 935
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.65259454705365,
|
| 1323 |
+
"grad_norm": 1.2347595691680908,
|
| 1324 |
+
"learning_rate": 1.3765884336487732e-05,
|
| 1325 |
+
"loss": 0.1622,
|
| 1326 |
+
"step": 940
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.6613896218117854,
|
| 1330 |
+
"grad_norm": 1.2973097562789917,
|
| 1331 |
+
"learning_rate": 1.362108295912869e-05,
|
| 1332 |
+
"loss": 0.1673,
|
| 1333 |
+
"step": 945
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.6701846965699207,
|
| 1337 |
+
"grad_norm": 1.1694238185882568,
|
| 1338 |
+
"learning_rate": 1.3476411063491081e-05,
|
| 1339 |
+
"loss": 0.153,
|
| 1340 |
+
"step": 950
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.6789797713280563,
|
| 1344 |
+
"grad_norm": 1.1278259754180908,
|
| 1345 |
+
"learning_rate": 1.3331882234413989e-05,
|
| 1346 |
+
"loss": 0.1596,
|
| 1347 |
+
"step": 955
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.6877748460861919,
|
| 1351 |
+
"grad_norm": 1.130813479423523,
|
| 1352 |
+
"learning_rate": 1.3187510043302384e-05,
|
| 1353 |
+
"loss": 0.1621,
|
| 1354 |
+
"step": 960
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.6965699208443272,
|
| 1358 |
+
"grad_norm": 1.1461126804351807,
|
| 1359 |
+
"learning_rate": 1.3043308046852765e-05,
|
| 1360 |
+
"loss": 0.1733,
|
| 1361 |
+
"step": 965
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.7053649956024626,
|
| 1365 |
+
"grad_norm": 1.139128565788269,
|
| 1366 |
+
"learning_rate": 1.28992897857802e-05,
|
| 1367 |
+
"loss": 0.149,
|
| 1368 |
+
"step": 970
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.714160070360598,
|
| 1372 |
+
"grad_norm": 1.0681076049804688,
|
| 1373 |
+
"learning_rate": 1.27554687835468e-05,
|
| 1374 |
+
"loss": 0.1672,
|
| 1375 |
+
"step": 975
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.7229551451187335,
|
| 1379 |
+
"grad_norm": 1.060549259185791,
|
| 1380 |
+
"learning_rate": 1.261185854509187e-05,
|
| 1381 |
+
"loss": 0.1525,
|
| 1382 |
+
"step": 980
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.731750219876869,
|
| 1386 |
+
"grad_norm": 1.050777554512024,
|
| 1387 |
+
"learning_rate": 1.2468472555563788e-05,
|
| 1388 |
+
"loss": 0.1597,
|
| 1389 |
+
"step": 985
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.7405452946350044,
|
| 1393 |
+
"grad_norm": 1.131949543952942,
|
| 1394 |
+
"learning_rate": 1.232532427905373e-05,
|
| 1395 |
+
"loss": 0.1524,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.7493403693931397,
|
| 1400 |
+
"grad_norm": 1.4923933744430542,
|
| 1401 |
+
"learning_rate": 1.2182427157331351e-05,
|
| 1402 |
+
"loss": 0.1786,
|
| 1403 |
+
"step": 995
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.7581354441512753,
|
| 1407 |
+
"grad_norm": 1.153559684753418,
|
| 1408 |
+
"learning_rate": 1.2039794608582659e-05,
|
| 1409 |
+
"loss": 0.1534,
|
| 1410 |
+
"step": 1000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.7669305189094109,
|
| 1414 |
+
"grad_norm": 1.1780742406845093,
|
| 1415 |
+
"learning_rate": 1.1897440026149948e-05,
|
| 1416 |
+
"loss": 0.1621,
|
| 1417 |
+
"step": 1005
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.7757255936675462,
|
| 1421 |
+
"grad_norm": 1.2406305074691772,
|
| 1422 |
+
"learning_rate": 1.1755376777274202e-05,
|
| 1423 |
+
"loss": 0.15,
|
| 1424 |
+
"step": 1010
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.7845206684256816,
|
| 1428 |
+
"grad_norm": 1.1729429960250854,
|
| 1429 |
+
"learning_rate": 1.1613618201839912e-05,
|
| 1430 |
+
"loss": 0.1493,
|
| 1431 |
+
"step": 1015
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.793315743183817,
|
| 1435 |
+
"grad_norm": 1.1011075973510742,
|
| 1436 |
+
"learning_rate": 1.1472177611122381e-05,
|
| 1437 |
+
"loss": 0.1359,
|
| 1438 |
+
"step": 1020
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.8021108179419525,
|
| 1442 |
+
"grad_norm": 0.9851521253585815,
|
| 1443 |
+
"learning_rate": 1.1331068286537844e-05,
|
| 1444 |
+
"loss": 0.1548,
|
| 1445 |
+
"step": 1025
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.810905892700088,
|
| 1449 |
+
"grad_norm": 1.070907473564148,
|
| 1450 |
+
"learning_rate": 1.1190303478396304e-05,
|
| 1451 |
+
"loss": 0.1501,
|
| 1452 |
+
"step": 1030
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.8197009674582234,
|
| 1456 |
+
"grad_norm": 1.2286392450332642,
|
| 1457 |
+
"learning_rate": 1.104989640465732e-05,
|
| 1458 |
+
"loss": 0.1398,
|
| 1459 |
+
"step": 1035
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.8284960422163588,
|
| 1463 |
+
"grad_norm": 1.018330693244934,
|
| 1464 |
+
"learning_rate": 1.0909860249688812e-05,
|
| 1465 |
+
"loss": 0.1268,
|
| 1466 |
+
"step": 1040
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.8372911169744943,
|
| 1470 |
+
"grad_norm": 1.2737165689468384,
|
| 1471 |
+
"learning_rate": 1.0770208163029066e-05,
|
| 1472 |
+
"loss": 0.1552,
|
| 1473 |
+
"step": 1045
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.8460861917326299,
|
| 1477 |
+
"grad_norm": 1.213727355003357,
|
| 1478 |
+
"learning_rate": 1.0630953258151946e-05,
|
| 1479 |
+
"loss": 0.1372,
|
| 1480 |
+
"step": 1050
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.8548812664907652,
|
| 1484 |
+
"grad_norm": 1.4174960851669312,
|
| 1485 |
+
"learning_rate": 1.0492108611235545e-05,
|
| 1486 |
+
"loss": 0.1621,
|
| 1487 |
+
"step": 1055
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.8636763412489006,
|
| 1491 |
+
"grad_norm": 1.1794441938400269,
|
| 1492 |
+
"learning_rate": 1.0353687259934322e-05,
|
| 1493 |
+
"loss": 0.1387,
|
| 1494 |
+
"step": 1060
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.872471416007036,
|
| 1498 |
+
"grad_norm": 1.0497779846191406,
|
| 1499 |
+
"learning_rate": 1.0215702202154828e-05,
|
| 1500 |
+
"loss": 0.1503,
|
| 1501 |
+
"step": 1065
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.8812664907651715,
|
| 1505 |
+
"grad_norm": 1.0930366516113281,
|
| 1506 |
+
"learning_rate": 1.0078166394835213e-05,
|
| 1507 |
+
"loss": 0.1346,
|
| 1508 |
+
"step": 1070
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.890061565523307,
|
| 1512 |
+
"grad_norm": 1.2730180025100708,
|
| 1513 |
+
"learning_rate": 9.94109275272856e-06,
|
| 1514 |
+
"loss": 0.1346,
|
| 1515 |
+
"step": 1075
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.8988566402814424,
|
| 1519 |
+
"grad_norm": 0.9998127818107605,
|
| 1520 |
+
"learning_rate": 9.80449414719015e-06,
|
| 1521 |
+
"loss": 0.1392,
|
| 1522 |
+
"step": 1080
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.9076517150395778,
|
| 1526 |
+
"grad_norm": 1.2404751777648926,
|
| 1527 |
+
"learning_rate": 9.668383404968845e-06,
|
| 1528 |
+
"loss": 0.1234,
|
| 1529 |
+
"step": 1085
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.9164467897977133,
|
| 1533 |
+
"grad_norm": 1.0237677097320557,
|
| 1534 |
+
"learning_rate": 9.532773307002659e-06,
|
| 1535 |
+
"loss": 0.1303,
|
| 1536 |
+
"step": 1090
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.9252418645558487,
|
| 1540 |
+
"grad_norm": 1.0831749439239502,
|
| 1541 |
+
"learning_rate": 9.397676587218577e-06,
|
| 1542 |
+
"loss": 0.1231,
|
| 1543 |
+
"step": 1095
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.9340369393139842,
|
| 1547 |
+
"grad_norm": 1.0309388637542725,
|
| 1548 |
+
"learning_rate": 9.263105931336854e-06,
|
| 1549 |
+
"loss": 0.1309,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.9428320140721196,
|
| 1554 |
+
"grad_norm": 1.0883592367172241,
|
| 1555 |
+
"learning_rate": 9.129073975679822e-06,
|
| 1556 |
+
"loss": 0.1226,
|
| 1557 |
+
"step": 1105
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.951627088830255,
|
| 1561 |
+
"grad_norm": 0.9501267671585083,
|
| 1562 |
+
"learning_rate": 8.995593305985284e-06,
|
| 1563 |
+
"loss": 0.1333,
|
| 1564 |
+
"step": 1110
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.9604221635883905,
|
| 1568 |
+
"grad_norm": 0.9364697337150574,
|
| 1569 |
+
"learning_rate": 8.862676456224744e-06,
|
| 1570 |
+
"loss": 0.1207,
|
| 1571 |
+
"step": 1115
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.969217238346526,
|
| 1575 |
+
"grad_norm": 1.0854167938232422,
|
| 1576 |
+
"learning_rate": 8.730335907426436e-06,
|
| 1577 |
+
"loss": 0.1155,
|
| 1578 |
+
"step": 1120
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.9780123131046614,
|
| 1582 |
+
"grad_norm": 0.8636917471885681,
|
| 1583 |
+
"learning_rate": 8.598584086503342e-06,
|
| 1584 |
+
"loss": 0.1153,
|
| 1585 |
+
"step": 1125
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.9868073878627968,
|
| 1589 |
+
"grad_norm": 1.1015220880508423,
|
| 1590 |
+
"learning_rate": 8.467433365086277e-06,
|
| 1591 |
+
"loss": 0.13,
|
| 1592 |
+
"step": 1130
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.9956024626209323,
|
| 1596 |
+
"grad_norm": 0.9550289511680603,
|
| 1597 |
+
"learning_rate": 8.336896058362238e-06,
|
| 1598 |
+
"loss": 0.1284,
|
| 1599 |
+
"step": 1135
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 2.0035180299032542,
|
| 1603 |
+
"grad_norm": 0.7205578684806824,
|
| 1604 |
+
"learning_rate": 8.206984423917919e-06,
|
| 1605 |
+
"loss": 0.1041,
|
| 1606 |
+
"step": 1140
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 2.0123131046613896,
|
| 1610 |
+
"grad_norm": 0.8555493354797363,
|
| 1611 |
+
"learning_rate": 8.077710660588749e-06,
|
| 1612 |
+
"loss": 0.0941,
|
| 1613 |
+
"step": 1145
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 2.021108179419525,
|
| 1617 |
+
"grad_norm": 1.0212748050689697,
|
| 1618 |
+
"learning_rate": 7.949086907313419e-06,
|
| 1619 |
+
"loss": 0.1005,
|
| 1620 |
+
"step": 1150
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 2.0299032541776607,
|
| 1624 |
+
"grad_norm": 1.0084145069122314,
|
| 1625 |
+
"learning_rate": 7.821125241994017e-06,
|
| 1626 |
+
"loss": 0.0993,
|
| 1627 |
+
"step": 1155
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 2.038698328935796,
|
| 1631 |
+
"grad_norm": 0.9347840547561646,
|
| 1632 |
+
"learning_rate": 7.693837680361868e-06,
|
| 1633 |
+
"loss": 0.0949,
|
| 1634 |
+
"step": 1160
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.0474934036939314,
|
| 1638 |
+
"grad_norm": 0.9160139560699463,
|
| 1639 |
+
"learning_rate": 7.567236174849293e-06,
|
| 1640 |
+
"loss": 0.1001,
|
| 1641 |
+
"step": 1165
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 2.0562884784520667,
|
| 1645 |
+
"grad_norm": 1.0276960134506226,
|
| 1646 |
+
"learning_rate": 7.441332613467242e-06,
|
| 1647 |
+
"loss": 0.091,
|
| 1648 |
+
"step": 1170
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 2.065083553210202,
|
| 1652 |
+
"grad_norm": 1.044572353363037,
|
| 1653 |
+
"learning_rate": 7.316138818688976e-06,
|
| 1654 |
+
"loss": 0.0941,
|
| 1655 |
+
"step": 1175
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 2.073878627968338,
|
| 1659 |
+
"grad_norm": 0.9516727924346924,
|
| 1660 |
+
"learning_rate": 7.191666546339965e-06,
|
| 1661 |
+
"loss": 0.0945,
|
| 1662 |
+
"step": 1180
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 2.0826737027264732,
|
| 1666 |
+
"grad_norm": 0.8449445962905884,
|
| 1667 |
+
"learning_rate": 7.067927484493984e-06,
|
| 1668 |
+
"loss": 0.0934,
|
| 1669 |
+
"step": 1185
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 2.0914687774846086,
|
| 1673 |
+
"grad_norm": 0.8541709780693054,
|
| 1674 |
+
"learning_rate": 6.944933252375575e-06,
|
| 1675 |
+
"loss": 0.1004,
|
| 1676 |
+
"step": 1190
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.100263852242744,
|
| 1680 |
+
"grad_norm": 1.1123250722885132,
|
| 1681 |
+
"learning_rate": 6.822695399269022e-06,
|
| 1682 |
+
"loss": 0.0934,
|
| 1683 |
+
"step": 1195
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.1090589270008797,
|
| 1687 |
+
"grad_norm": 1.0344734191894531,
|
| 1688 |
+
"learning_rate": 6.701225403433846e-06,
|
| 1689 |
+
"loss": 0.0992,
|
| 1690 |
+
"step": 1200
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 2.117854001759015,
|
| 1694 |
+
"grad_norm": 0.7952005863189697,
|
| 1695 |
+
"learning_rate": 6.5805346710269695e-06,
|
| 1696 |
+
"loss": 0.0903,
|
| 1697 |
+
"step": 1205
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.1266490765171504,
|
| 1701 |
+
"grad_norm": 0.8689743876457214,
|
| 1702 |
+
"learning_rate": 6.4606345350316935e-06,
|
| 1703 |
+
"loss": 0.0868,
|
| 1704 |
+
"step": 1210
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.1354441512752858,
|
| 1708 |
+
"grad_norm": 0.9427767395973206,
|
| 1709 |
+
"learning_rate": 6.34153625419351e-06,
|
| 1710 |
+
"loss": 0.0982,
|
| 1711 |
+
"step": 1215
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.144239226033421,
|
| 1715 |
+
"grad_norm": 1.0584067106246948,
|
| 1716 |
+
"learning_rate": 6.223251011962876e-06,
|
| 1717 |
+
"loss": 0.094,
|
| 1718 |
+
"step": 1220
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.153034300791557,
|
| 1722 |
+
"grad_norm": 0.9294475317001343,
|
| 1723 |
+
"learning_rate": 6.105789915445112e-06,
|
| 1724 |
+
"loss": 0.092,
|
| 1725 |
+
"step": 1225
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.1618293755496922,
|
| 1729 |
+
"grad_norm": 0.779670774936676,
|
| 1730 |
+
"learning_rate": 5.989163994357386e-06,
|
| 1731 |
+
"loss": 0.0941,
|
| 1732 |
+
"step": 1230
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.1706244503078276,
|
| 1736 |
+
"grad_norm": 0.872122585773468,
|
| 1737 |
+
"learning_rate": 5.873384199993063e-06,
|
| 1738 |
+
"loss": 0.0946,
|
| 1739 |
+
"step": 1235
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.179419525065963,
|
| 1743 |
+
"grad_norm": 0.8662841320037842,
|
| 1744 |
+
"learning_rate": 5.758461404193335e-06,
|
| 1745 |
+
"loss": 0.0887,
|
| 1746 |
+
"step": 1240
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 2.1882145998240983,
|
| 1750 |
+
"grad_norm": 0.6998201012611389,
|
| 1751 |
+
"learning_rate": 5.644406398326341e-06,
|
| 1752 |
+
"loss": 0.0861,
|
| 1753 |
+
"step": 1245
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 2.197009674582234,
|
| 1757 |
+
"grad_norm": 0.8021641969680786,
|
| 1758 |
+
"learning_rate": 5.531229892273871e-06,
|
| 1759 |
+
"loss": 0.0894,
|
| 1760 |
+
"step": 1250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.2058047493403694,
|
| 1764 |
+
"grad_norm": 0.7781965732574463,
|
| 1765 |
+
"learning_rate": 5.418942513425682e-06,
|
| 1766 |
+
"loss": 0.0795,
|
| 1767 |
+
"step": 1255
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 2.2145998240985048,
|
| 1771 |
+
"grad_norm": 0.9334102869033813,
|
| 1772 |
+
"learning_rate": 5.307554805681562e-06,
|
| 1773 |
+
"loss": 0.091,
|
| 1774 |
+
"step": 1260
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.22339489885664,
|
| 1778 |
+
"grad_norm": 0.8019357323646545,
|
| 1779 |
+
"learning_rate": 5.197077228461279e-06,
|
| 1780 |
+
"loss": 0.0832,
|
| 1781 |
+
"step": 1265
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.232189973614776,
|
| 1785 |
+
"grad_norm": 0.8955482840538025,
|
| 1786 |
+
"learning_rate": 5.0875201557224085e-06,
|
| 1787 |
+
"loss": 0.0819,
|
| 1788 |
+
"step": 1270
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.2409850483729112,
|
| 1792 |
+
"grad_norm": 0.6904844045639038,
|
| 1793 |
+
"learning_rate": 4.9788938749862025e-06,
|
| 1794 |
+
"loss": 0.0884,
|
| 1795 |
+
"step": 1275
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.2497801231310466,
|
| 1799 |
+
"grad_norm": 0.8131275177001953,
|
| 1800 |
+
"learning_rate": 4.8712085863716065e-06,
|
| 1801 |
+
"loss": 0.0862,
|
| 1802 |
+
"step": 1280
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.258575197889182,
|
| 1806 |
+
"grad_norm": 0.9088744521141052,
|
| 1807 |
+
"learning_rate": 4.7644744016374385e-06,
|
| 1808 |
+
"loss": 0.0888,
|
| 1809 |
+
"step": 1285
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.2673702726473177,
|
| 1813 |
+
"grad_norm": 0.722126305103302,
|
| 1814 |
+
"learning_rate": 4.658701343232876e-06,
|
| 1815 |
+
"loss": 0.0983,
|
| 1816 |
+
"step": 1290
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.276165347405453,
|
| 1820 |
+
"grad_norm": 0.8710489273071289,
|
| 1821 |
+
"learning_rate": 4.553899343356367e-06,
|
| 1822 |
+
"loss": 0.0915,
|
| 1823 |
+
"step": 1295
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.2849604221635884,
|
| 1827 |
+
"grad_norm": 0.7461559176445007,
|
| 1828 |
+
"learning_rate": 4.450078243022967e-06,
|
| 1829 |
+
"loss": 0.0933,
|
| 1830 |
+
"step": 1300
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.2937554969217238,
|
| 1834 |
+
"grad_norm": 0.8261001706123352,
|
| 1835 |
+
"learning_rate": 4.347247791140247e-06,
|
| 1836 |
+
"loss": 0.0891,
|
| 1837 |
+
"step": 1305
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.302550571679859,
|
| 1841 |
+
"grad_norm": 0.8420723080635071,
|
| 1842 |
+
"learning_rate": 4.24541764359289e-06,
|
| 1843 |
+
"loss": 0.0898,
|
| 1844 |
+
"step": 1310
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.311345646437995,
|
| 1848 |
+
"grad_norm": 0.9102108478546143,
|
| 1849 |
+
"learning_rate": 4.144597362335977e-06,
|
| 1850 |
+
"loss": 0.0915,
|
| 1851 |
+
"step": 1315
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.3201407211961302,
|
| 1855 |
+
"grad_norm": 0.8050822615623474,
|
| 1856 |
+
"learning_rate": 4.044796414497097e-06,
|
| 1857 |
+
"loss": 0.0832,
|
| 1858 |
+
"step": 1320
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.3289357959542656,
|
| 1862 |
+
"grad_norm": 0.7277134656906128,
|
| 1863 |
+
"learning_rate": 3.946024171487407e-06,
|
| 1864 |
+
"loss": 0.0844,
|
| 1865 |
+
"step": 1325
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.337730870712401,
|
| 1869 |
+
"grad_norm": 0.6631373167037964,
|
| 1870 |
+
"learning_rate": 3.848289908121605e-06,
|
| 1871 |
+
"loss": 0.079,
|
| 1872 |
+
"step": 1330
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.3465259454705363,
|
| 1876 |
+
"grad_norm": 0.6942434310913086,
|
| 1877 |
+
"learning_rate": 3.75160280174706e-06,
|
| 1878 |
+
"loss": 0.0682,
|
| 1879 |
+
"step": 1335
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.355321020228672,
|
| 1883 |
+
"grad_norm": 0.8732395768165588,
|
| 1884 |
+
"learning_rate": 3.6559719313820205e-06,
|
| 1885 |
+
"loss": 0.0853,
|
| 1886 |
+
"step": 1340
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.3641160949868074,
|
| 1890 |
+
"grad_norm": 0.7472450733184814,
|
| 1891 |
+
"learning_rate": 3.5614062768630827e-06,
|
| 1892 |
+
"loss": 0.0764,
|
| 1893 |
+
"step": 1345
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.3729111697449428,
|
| 1897 |
+
"grad_norm": 0.7423739433288574,
|
| 1898 |
+
"learning_rate": 3.467914718001996e-06,
|
| 1899 |
+
"loss": 0.0842,
|
| 1900 |
+
"step": 1350
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.381706244503078,
|
| 1904 |
+
"grad_norm": 0.6666902899742126,
|
| 1905 |
+
"learning_rate": 3.3755060337518314e-06,
|
| 1906 |
+
"loss": 0.0827,
|
| 1907 |
+
"step": 1355
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.390501319261214,
|
| 1911 |
+
"grad_norm": 0.8937864899635315,
|
| 1912 |
+
"learning_rate": 3.2841889013826074e-06,
|
| 1913 |
+
"loss": 0.0837,
|
| 1914 |
+
"step": 1360
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.3992963940193492,
|
| 1918 |
+
"grad_norm": 0.8662824034690857,
|
| 1919 |
+
"learning_rate": 3.193971895666521e-06,
|
| 1920 |
+
"loss": 0.0787,
|
| 1921 |
+
"step": 1365
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.4080914687774846,
|
| 1925 |
+
"grad_norm": 0.6090977191925049,
|
| 1926 |
+
"learning_rate": 3.104863488072749e-06,
|
| 1927 |
+
"loss": 0.0835,
|
| 1928 |
+
"step": 1370
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.41688654353562,
|
| 1932 |
+
"grad_norm": 0.593767523765564,
|
| 1933 |
+
"learning_rate": 3.016872045971958e-06,
|
| 1934 |
+
"loss": 0.0822,
|
| 1935 |
+
"step": 1375
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.4256816182937557,
|
| 1939 |
+
"grad_norm": 0.7369496822357178,
|
| 1940 |
+
"learning_rate": 2.930005831850623e-06,
|
| 1941 |
+
"loss": 0.0771,
|
| 1942 |
+
"step": 1380
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.434476693051891,
|
| 1946 |
+
"grad_norm": 0.8004661202430725,
|
| 1947 |
+
"learning_rate": 2.8442730025351598e-06,
|
| 1948 |
+
"loss": 0.0752,
|
| 1949 |
+
"step": 1385
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.4432717678100264,
|
| 1953 |
+
"grad_norm": 0.7470291256904602,
|
| 1954 |
+
"learning_rate": 2.7596816084259803e-06,
|
| 1955 |
+
"loss": 0.0731,
|
| 1956 |
+
"step": 1390
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.4520668425681618,
|
| 1960 |
+
"grad_norm": 0.7611671090126038,
|
| 1961 |
+
"learning_rate": 2.676239592741575e-06,
|
| 1962 |
+
"loss": 0.082,
|
| 1963 |
+
"step": 1395
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.460861917326297,
|
| 1967 |
+
"grad_norm": 0.7018731832504272,
|
| 1968 |
+
"learning_rate": 2.593954790772619e-06,
|
| 1969 |
+
"loss": 0.0864,
|
| 1970 |
+
"step": 1400
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.469656992084433,
|
| 1974 |
+
"grad_norm": 0.6412310600280762,
|
| 1975 |
+
"learning_rate": 2.512834929146232e-06,
|
| 1976 |
+
"loss": 0.0771,
|
| 1977 |
+
"step": 1405
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.4784520668425682,
|
| 1981 |
+
"grad_norm": 0.6994911432266235,
|
| 1982 |
+
"learning_rate": 2.4328876251004394e-06,
|
| 1983 |
+
"loss": 0.0683,
|
| 1984 |
+
"step": 1410
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.4872471416007036,
|
| 1988 |
+
"grad_norm": 0.7697396874427795,
|
| 1989 |
+
"learning_rate": 2.3541203857689308e-06,
|
| 1990 |
+
"loss": 0.084,
|
| 1991 |
+
"step": 1415
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.496042216358839,
|
| 1995 |
+
"grad_norm": 0.6429616808891296,
|
| 1996 |
+
"learning_rate": 2.276540607476089e-06,
|
| 1997 |
+
"loss": 0.067,
|
| 1998 |
+
"step": 1420
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.5048372911169743,
|
| 2002 |
+
"grad_norm": 0.7573561668395996,
|
| 2003 |
+
"learning_rate": 2.2001555750424897e-06,
|
| 2004 |
+
"loss": 0.0783,
|
| 2005 |
+
"step": 1425
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.51363236587511,
|
| 2009 |
+
"grad_norm": 0.575046956539154,
|
| 2010 |
+
"learning_rate": 2.1249724611008714e-06,
|
| 2011 |
+
"loss": 0.0753,
|
| 2012 |
+
"step": 1430
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.5224274406332454,
|
| 2016 |
+
"grad_norm": 0.7653530836105347,
|
| 2017 |
+
"learning_rate": 2.0509983254225772e-06,
|
| 2018 |
+
"loss": 0.0816,
|
| 2019 |
+
"step": 1435
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.5312225153913808,
|
| 2023 |
+
"grad_norm": 0.6458086967468262,
|
| 2024 |
+
"learning_rate": 1.978240114254653e-06,
|
| 2025 |
+
"loss": 0.0734,
|
| 2026 |
+
"step": 1440
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.540017590149516,
|
| 2030 |
+
"grad_norm": 0.573203980922699,
|
| 2031 |
+
"learning_rate": 1.9067046596676013e-06,
|
| 2032 |
+
"loss": 0.0729,
|
| 2033 |
+
"step": 1445
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.5488126649076515,
|
| 2037 |
+
"grad_norm": 0.7243578433990479,
|
| 2038 |
+
"learning_rate": 1.8363986789138215e-06,
|
| 2039 |
+
"loss": 0.0673,
|
| 2040 |
+
"step": 1450
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.5576077396657872,
|
| 2044 |
+
"grad_norm": 0.5532898902893066,
|
| 2045 |
+
"learning_rate": 1.7673287737968624e-06,
|
| 2046 |
+
"loss": 0.0719,
|
| 2047 |
+
"step": 1455
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.5664028144239226,
|
| 2051 |
+
"grad_norm": 0.6563058495521545,
|
| 2052 |
+
"learning_rate": 1.6995014300515067e-06,
|
| 2053 |
+
"loss": 0.0757,
|
| 2054 |
+
"step": 1460
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.575197889182058,
|
| 2058 |
+
"grad_norm": 0.7555197477340698,
|
| 2059 |
+
"learning_rate": 1.6329230167347559e-06,
|
| 2060 |
+
"loss": 0.0856,
|
| 2061 |
+
"step": 1465
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.5839929639401937,
|
| 2065 |
+
"grad_norm": 0.8473305702209473,
|
| 2066 |
+
"learning_rate": 1.567599785627758e-06,
|
| 2067 |
+
"loss": 0.0824,
|
| 2068 |
+
"step": 1470
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.592788038698329,
|
| 2072 |
+
"grad_norm": 0.817717969417572,
|
| 2073 |
+
"learning_rate": 1.5035378706487712e-06,
|
| 2074 |
+
"loss": 0.0869,
|
| 2075 |
+
"step": 1475
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.6015831134564644,
|
| 2079 |
+
"grad_norm": 0.6018490791320801,
|
| 2080 |
+
"learning_rate": 1.4407432872771852e-06,
|
| 2081 |
+
"loss": 0.077,
|
| 2082 |
+
"step": 1480
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.6103781882145998,
|
| 2086 |
+
"grad_norm": 0.7161872386932373,
|
| 2087 |
+
"learning_rate": 1.3792219319886413e-06,
|
| 2088 |
+
"loss": 0.0724,
|
| 2089 |
+
"step": 1485
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.619173262972735,
|
| 2093 |
+
"grad_norm": 0.7263765335083008,
|
| 2094 |
+
"learning_rate": 1.3189795817013695e-06,
|
| 2095 |
+
"loss": 0.0818,
|
| 2096 |
+
"step": 1490
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.627968337730871,
|
| 2100 |
+
"grad_norm": 0.6535840034484863,
|
| 2101 |
+
"learning_rate": 1.2600218932337204e-06,
|
| 2102 |
+
"loss": 0.0693,
|
| 2103 |
+
"step": 1495
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.6367634124890063,
|
| 2107 |
+
"grad_norm": 0.7487488985061646,
|
| 2108 |
+
"learning_rate": 1.202354402772981e-06,
|
| 2109 |
+
"loss": 0.0694,
|
| 2110 |
+
"step": 1500
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.6455584872471416,
|
| 2114 |
+
"grad_norm": 0.5553443431854248,
|
| 2115 |
+
"learning_rate": 1.1459825253555306e-06,
|
| 2116 |
+
"loss": 0.063,
|
| 2117 |
+
"step": 1505
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.654353562005277,
|
| 2121 |
+
"grad_norm": 0.6657333374023438,
|
| 2122 |
+
"learning_rate": 1.0909115543583508e-06,
|
| 2123 |
+
"loss": 0.0793,
|
| 2124 |
+
"step": 1510
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.6631486367634123,
|
| 2128 |
+
"grad_norm": 0.6260197758674622,
|
| 2129 |
+
"learning_rate": 1.0371466610019837e-06,
|
| 2130 |
+
"loss": 0.0648,
|
| 2131 |
+
"step": 1515
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.671943711521548,
|
| 2135 |
+
"grad_norm": 0.6683763265609741,
|
| 2136 |
+
"learning_rate": 9.846928938649475e-07,
|
| 2137 |
+
"loss": 0.0786,
|
| 2138 |
+
"step": 1520
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.6807387862796834,
|
| 2142 |
+
"grad_norm": 0.6568044424057007,
|
| 2143 |
+
"learning_rate": 9.335551784096552e-07,
|
| 2144 |
+
"loss": 0.0637,
|
| 2145 |
+
"step": 1525
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.6895338610378188,
|
| 2149 |
+
"grad_norm": 0.5744944214820862,
|
| 2150 |
+
"learning_rate": 8.837383165199308e-07,
|
| 2151 |
+
"loss": 0.0751,
|
| 2152 |
+
"step": 1530
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.698328935795954,
|
| 2156 |
+
"grad_norm": 0.5950682759284973,
|
| 2157 |
+
"learning_rate": 8.35246986050095e-07,
|
| 2158 |
+
"loss": 0.0683,
|
| 2159 |
+
"step": 1535
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.7071240105540895,
|
| 2163 |
+
"grad_norm": 0.5924869775772095,
|
| 2164 |
+
"learning_rate": 7.880857403857028e-07,
|
| 2165 |
+
"loss": 0.0715,
|
| 2166 |
+
"step": 1540
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.7159190853122253,
|
| 2170 |
+
"grad_norm": 0.48297256231307983,
|
| 2171 |
+
"learning_rate": 7.422590080159947e-07,
|
| 2172 |
+
"loss": 0.0707,
|
| 2173 |
+
"step": 1545
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.7247141600703606,
|
| 2177 |
+
"grad_norm": 0.6890292167663574,
|
| 2178 |
+
"learning_rate": 6.977710921180452e-07,
|
| 2179 |
+
"loss": 0.0744,
|
| 2180 |
+
"step": 1550
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.733509234828496,
|
| 2184 |
+
"grad_norm": 0.6717385649681091,
|
| 2185 |
+
"learning_rate": 6.546261701526851e-07,
|
| 2186 |
+
"loss": 0.0679,
|
| 2187 |
+
"step": 1555
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.7423043095866317,
|
| 2191 |
+
"grad_norm": 0.7569487690925598,
|
| 2192 |
+
"learning_rate": 6.128282934722479e-07,
|
| 2193 |
+
"loss": 0.0873,
|
| 2194 |
+
"step": 1560
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.751099384344767,
|
| 2198 |
+
"grad_norm": 0.6629652976989746,
|
| 2199 |
+
"learning_rate": 5.723813869401362e-07,
|
| 2200 |
+
"loss": 0.0693,
|
| 2201 |
+
"step": 1565
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.7598944591029024,
|
| 2205 |
+
"grad_norm": 0.6372948288917542,
|
| 2206 |
+
"learning_rate": 5.332892485622648e-07,
|
| 2207 |
+
"loss": 0.0636,
|
| 2208 |
+
"step": 1570
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.7686895338610378,
|
| 2212 |
+
"grad_norm": 0.594325065612793,
|
| 2213 |
+
"learning_rate": 4.955555491304409e-07,
|
| 2214 |
+
"loss": 0.0702,
|
| 2215 |
+
"step": 1575
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.777484608619173,
|
| 2219 |
+
"grad_norm": 0.6359429359436035,
|
| 2220 |
+
"learning_rate": 4.5918383187766143e-07,
|
| 2221 |
+
"loss": 0.0834,
|
| 2222 |
+
"step": 1580
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.786279683377309,
|
| 2226 |
+
"grad_norm": 0.746578574180603,
|
| 2227 |
+
"learning_rate": 4.2417751214540115e-07,
|
| 2228 |
+
"loss": 0.0766,
|
| 2229 |
+
"step": 1585
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.7950747581354443,
|
| 2233 |
+
"grad_norm": 0.8064423203468323,
|
| 2234 |
+
"learning_rate": 3.905398770629104e-07,
|
| 2235 |
+
"loss": 0.0723,
|
| 2236 |
+
"step": 1590
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.8038698328935796,
|
| 2240 |
+
"grad_norm": 0.7364261150360107,
|
| 2241 |
+
"learning_rate": 3.582740852385541e-07,
|
| 2242 |
+
"loss": 0.0758,
|
| 2243 |
+
"step": 1595
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.812664907651715,
|
| 2247 |
+
"grad_norm": 0.6186119318008423,
|
| 2248 |
+
"learning_rate": 3.2738316646320543e-07,
|
| 2249 |
+
"loss": 0.0658,
|
| 2250 |
+
"step": 1600
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.8214599824098503,
|
| 2254 |
+
"grad_norm": 0.6899746060371399,
|
| 2255 |
+
"learning_rate": 2.9787002142575516e-07,
|
| 2256 |
+
"loss": 0.0663,
|
| 2257 |
+
"step": 1605
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.830255057167986,
|
| 2261 |
+
"grad_norm": 0.6665034890174866,
|
| 2262 |
+
"learning_rate": 2.6973742144072775e-07,
|
| 2263 |
+
"loss": 0.0696,
|
| 2264 |
+
"step": 1610
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.8390501319261214,
|
| 2268 |
+
"grad_norm": 0.5943213701248169,
|
| 2269 |
+
"learning_rate": 2.4298800818805477e-07,
|
| 2270 |
+
"loss": 0.0625,
|
| 2271 |
+
"step": 1615
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.847845206684257,
|
| 2275 |
+
"grad_norm": 0.5624052882194519,
|
| 2276 |
+
"learning_rate": 2.1762429346502132e-07,
|
| 2277 |
+
"loss": 0.0685,
|
| 2278 |
+
"step": 1620
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 2.856640281442392,
|
| 2282 |
+
"grad_norm": 0.6240286231040955,
|
| 2283 |
+
"learning_rate": 1.9364865895039664e-07,
|
| 2284 |
+
"loss": 0.0751,
|
| 2285 |
+
"step": 1625
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.8654353562005275,
|
| 2289 |
+
"grad_norm": 0.6763429045677185,
|
| 2290 |
+
"learning_rate": 1.7106335598080115e-07,
|
| 2291 |
+
"loss": 0.0721,
|
| 2292 |
+
"step": 1630
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.8742304309586633,
|
| 2296 |
+
"grad_norm": 0.6858879923820496,
|
| 2297 |
+
"learning_rate": 1.4987050533929902e-07,
|
| 2298 |
+
"loss": 0.0727,
|
| 2299 |
+
"step": 1635
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.8830255057167986,
|
| 2303 |
+
"grad_norm": 0.8514876961708069,
|
| 2304 |
+
"learning_rate": 1.3007209705625745e-07,
|
| 2305 |
+
"loss": 0.076,
|
| 2306 |
+
"step": 1640
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 2.891820580474934,
|
| 2310 |
+
"grad_norm": 0.7763850092887878,
|
| 2311 |
+
"learning_rate": 1.1166999022247438e-07,
|
| 2312 |
+
"loss": 0.076,
|
| 2313 |
+
"step": 1645
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.9006156552330697,
|
| 2317 |
+
"grad_norm": 0.6178231239318848,
|
| 2318 |
+
"learning_rate": 9.466591281461989e-08,
|
| 2319 |
+
"loss": 0.063,
|
| 2320 |
+
"step": 1650
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.909410729991205,
|
| 2324 |
+
"grad_norm": 0.8232988119125366,
|
| 2325 |
+
"learning_rate": 7.906146153296267e-08,
|
| 2326 |
+
"loss": 0.0731,
|
| 2327 |
+
"step": 1655
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.9182058047493404,
|
| 2331 |
+
"grad_norm": 0.8093054890632629,
|
| 2332 |
+
"learning_rate": 6.485810165145156e-08,
|
| 2333 |
+
"loss": 0.08,
|
| 2334 |
+
"step": 1660
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.927000879507476,
|
| 2338 |
+
"grad_norm": 0.7137056589126587,
|
| 2339 |
+
"learning_rate": 5.205716688011564e-08,
|
| 2340 |
+
"loss": 0.0655,
|
| 2341 |
+
"step": 1665
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.935795954265611,
|
| 2345 |
+
"grad_norm": 0.7028983235359192,
|
| 2346 |
+
"learning_rate": 4.0659859239831065e-08,
|
| 2347 |
+
"loss": 0.0738,
|
| 2348 |
+
"step": 1670
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.944591029023747,
|
| 2352 |
+
"grad_norm": 0.7251346111297607,
|
| 2353 |
+
"learning_rate": 3.0667248949449725e-08,
|
| 2354 |
+
"loss": 0.0672,
|
| 2355 |
+
"step": 1675
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.9533861037818823,
|
| 2359 |
+
"grad_norm": 0.6649497151374817,
|
| 2360 |
+
"learning_rate": 2.2080274325306237e-08,
|
| 2361 |
+
"loss": 0.0715,
|
| 2362 |
+
"step": 1680
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.9621811785400176,
|
| 2366 |
+
"grad_norm": 0.6889492273330688,
|
| 2367 |
+
"learning_rate": 1.4899741693105152e-08,
|
| 2368 |
+
"loss": 0.0753,
|
| 2369 |
+
"step": 1685
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.970976253298153,
|
| 2373 |
+
"grad_norm": 0.592121958732605,
|
| 2374 |
+
"learning_rate": 9.12632531221147e-09,
|
| 2375 |
+
"loss": 0.0663,
|
| 2376 |
+
"step": 1690
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.9797713280562883,
|
| 2380 |
+
"grad_norm": 0.7029461860656738,
|
| 2381 |
+
"learning_rate": 4.760567312331321e-09,
|
| 2382 |
+
"loss": 0.0692,
|
| 2383 |
+
"step": 1695
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.988566402814424,
|
| 2387 |
+
"grad_norm": 0.5134033560752869,
|
| 2388 |
+
"learning_rate": 1.8028776426110005e-09,
|
| 2389 |
+
"loss": 0.0726,
|
| 2390 |
+
"step": 1700
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.9973614775725594,
|
| 2394 |
+
"grad_norm": 0.6766219735145569,
|
| 2395 |
+
"learning_rate": 2.5353403313443934e-10,
|
| 2396 |
+
"loss": 0.0671,
|
| 2397 |
+
"step": 1705
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 3.0,
|
| 2401 |
+
"step": 1707,
|
| 2402 |
+
"total_flos": 2.251473000899543e+18,
|
| 2403 |
+
"train_loss": 0.3827222754505551,
|
| 2404 |
+
"train_runtime": 1075.2003,
|
| 2405 |
+
"train_samples_per_second": 50.756,
|
| 2406 |
+
"train_steps_per_second": 1.588
|
| 2407 |
+
}
|
| 2408 |
+
],
|
| 2409 |
+
"logging_steps": 5,
|
| 2410 |
+
"max_steps": 1707,
|
| 2411 |
+
"num_input_tokens_seen": 0,
|
| 2412 |
+
"num_train_epochs": 3,
|
| 2413 |
+
"save_steps": 20000,
|
| 2414 |
+
"stateful_callbacks": {
|
| 2415 |
+
"TrainerControl": {
|
| 2416 |
+
"args": {
|
| 2417 |
+
"should_epoch_stop": false,
|
| 2418 |
+
"should_evaluate": false,
|
| 2419 |
+
"should_log": false,
|
| 2420 |
+
"should_save": false,
|
| 2421 |
+
"should_training_stop": false
|
| 2422 |
+
},
|
| 2423 |
+
"attributes": {}
|
| 2424 |
+
}
|
| 2425 |
+
},
|
| 2426 |
+
"total_flos": 2.251473000899543e+18,
|
| 2427 |
+
"train_batch_size": 2,
|
| 2428 |
+
"trial_name": null,
|
| 2429 |
+
"trial_params": null
|
| 2430 |
+
}
|
133_128_e3_3e-5/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c85cb79b9d185abc8d55849d3679a41bf99a7c5a8afd7ae1155d3c09a8a9789f
|
| 3 |
+
size 8273
|
133_128_e3_3e-5/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|