Upload folder using huggingface_hub
Browse files- instruct/13_128_e3_3e-5/adapter_config.json +39 -0
- instruct/13_128_e3_3e-5/adapter_model.safetensors +3 -0
- instruct/13_128_e3_3e-5/added_tokens.json +9 -0
- instruct/13_128_e3_3e-5/all_results.json +9 -0
- instruct/13_128_e3_3e-5/chat_template.jinja +62 -0
- instruct/13_128_e3_3e-5/config.json +32 -0
- instruct/13_128_e3_3e-5/merges.txt +0 -0
- instruct/13_128_e3_3e-5/special_tokens_map.json +33 -0
- instruct/13_128_e3_3e-5/tokenizer.json +0 -0
- instruct/13_128_e3_3e-5/tokenizer_config.json +234 -0
- instruct/13_128_e3_3e-5/train_results.json +9 -0
- instruct/13_128_e3_3e-5/trainer_state.json +2696 -0
- instruct/13_128_e3_3e-5/training_args.bin +3 -0
- instruct/13_128_e3_3e-5/vocab.json +0 -0
instruct/13_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 |
+
"up_proj",
|
| 28 |
+
"k_proj",
|
| 29 |
+
"gate_proj",
|
| 30 |
+
"down_proj",
|
| 31 |
+
"o_proj",
|
| 32 |
+
"q_proj",
|
| 33 |
+
"v_proj"
|
| 34 |
+
],
|
| 35 |
+
"task_type": "CAUSAL_LM",
|
| 36 |
+
"trainable_token_indices": null,
|
| 37 |
+
"use_dora": false,
|
| 38 |
+
"use_rslora": false
|
| 39 |
+
}
|
instruct/13_128_e3_3e-5/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e492cd4161e04bdba078e9e35b28d785472a85db03fa83a455f6bd1535a1592
|
| 3 |
+
size 791751704
|
instruct/13_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 |
+
}
|
instruct/13_128_e3_3e-5/all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.47550395020424e+18,
|
| 4 |
+
"train_loss": 0.3481526232446359,
|
| 5 |
+
"train_runtime": 1204.0905,
|
| 6 |
+
"train_samples": 20214,
|
| 7 |
+
"train_samples_per_second": 50.363,
|
| 8 |
+
"train_steps_per_second": 1.575
|
| 9 |
+
}
|
instruct/13_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 %}
|
instruct/13_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 |
+
}
|
instruct/13_128_e3_3e-5/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
instruct/13_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 |
+
}
|
instruct/13_128_e3_3e-5/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
instruct/13_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 |
+
}
|
instruct/13_128_e3_3e-5/train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.47550395020424e+18,
|
| 4 |
+
"train_loss": 0.3481526232446359,
|
| 5 |
+
"train_runtime": 1204.0905,
|
| 6 |
+
"train_samples": 20214,
|
| 7 |
+
"train_samples_per_second": 50.363,
|
| 8 |
+
"train_steps_per_second": 1.575
|
| 9 |
+
}
|
instruct/13_128_e3_3e-5/trainer_state.json
ADDED
|
@@ -0,0 +1,2696 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1896,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.007911392405063292,
|
| 14 |
+
"grad_norm": 2.403468608856201,
|
| 15 |
+
"learning_rate": 1.2631578947368422e-06,
|
| 16 |
+
"loss": 1.5122,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.015822784810126583,
|
| 21 |
+
"grad_norm": 2.066967487335205,
|
| 22 |
+
"learning_rate": 2.842105263157895e-06,
|
| 23 |
+
"loss": 1.5476,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.023734177215189875,
|
| 28 |
+
"grad_norm": 1.4056240320205688,
|
| 29 |
+
"learning_rate": 4.421052631578947e-06,
|
| 30 |
+
"loss": 1.5042,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03164556962025317,
|
| 35 |
+
"grad_norm": 0.7333563566207886,
|
| 36 |
+
"learning_rate": 6e-06,
|
| 37 |
+
"loss": 1.447,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03955696202531646,
|
| 42 |
+
"grad_norm": 0.48863300681114197,
|
| 43 |
+
"learning_rate": 7.578947368421053e-06,
|
| 44 |
+
"loss": 1.4462,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.04746835443037975,
|
| 49 |
+
"grad_norm": 0.4544259309768677,
|
| 50 |
+
"learning_rate": 9.157894736842105e-06,
|
| 51 |
+
"loss": 1.3882,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.055379746835443035,
|
| 56 |
+
"grad_norm": 0.4853452742099762,
|
| 57 |
+
"learning_rate": 1.0736842105263158e-05,
|
| 58 |
+
"loss": 1.394,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06329113924050633,
|
| 63 |
+
"grad_norm": 0.4000134766101837,
|
| 64 |
+
"learning_rate": 1.231578947368421e-05,
|
| 65 |
+
"loss": 1.3724,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.07120253164556962,
|
| 70 |
+
"grad_norm": 0.43093422055244446,
|
| 71 |
+
"learning_rate": 1.3894736842105263e-05,
|
| 72 |
+
"loss": 1.3343,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.07911392405063292,
|
| 77 |
+
"grad_norm": 0.40417981147766113,
|
| 78 |
+
"learning_rate": 1.547368421052632e-05,
|
| 79 |
+
"loss": 1.3434,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.08702531645569621,
|
| 84 |
+
"grad_norm": 0.4168269634246826,
|
| 85 |
+
"learning_rate": 1.705263157894737e-05,
|
| 86 |
+
"loss": 1.3325,
|
| 87 |
+
"step": 55
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0949367088607595,
|
| 91 |
+
"grad_norm": 0.3852977752685547,
|
| 92 |
+
"learning_rate": 1.8631578947368424e-05,
|
| 93 |
+
"loss": 1.2944,
|
| 94 |
+
"step": 60
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.10284810126582279,
|
| 98 |
+
"grad_norm": 0.4567843973636627,
|
| 99 |
+
"learning_rate": 2.0210526315789475e-05,
|
| 100 |
+
"loss": 1.3179,
|
| 101 |
+
"step": 65
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.11075949367088607,
|
| 105 |
+
"grad_norm": 0.4385586977005005,
|
| 106 |
+
"learning_rate": 2.178947368421053e-05,
|
| 107 |
+
"loss": 1.292,
|
| 108 |
+
"step": 70
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.11867088607594936,
|
| 112 |
+
"grad_norm": 0.43308112025260925,
|
| 113 |
+
"learning_rate": 2.336842105263158e-05,
|
| 114 |
+
"loss": 1.2736,
|
| 115 |
+
"step": 75
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.12658227848101267,
|
| 119 |
+
"grad_norm": 0.43022000789642334,
|
| 120 |
+
"learning_rate": 2.4947368421052635e-05,
|
| 121 |
+
"loss": 1.3027,
|
| 122 |
+
"step": 80
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.13449367088607594,
|
| 126 |
+
"grad_norm": 0.49931344389915466,
|
| 127 |
+
"learning_rate": 2.6526315789473685e-05,
|
| 128 |
+
"loss": 1.2567,
|
| 129 |
+
"step": 85
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.14240506329113925,
|
| 133 |
+
"grad_norm": 0.5501281023025513,
|
| 134 |
+
"learning_rate": 2.810526315789474e-05,
|
| 135 |
+
"loss": 1.2604,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.15031645569620253,
|
| 140 |
+
"grad_norm": 0.5215017795562744,
|
| 141 |
+
"learning_rate": 2.968421052631579e-05,
|
| 142 |
+
"loss": 1.2336,
|
| 143 |
+
"step": 95
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.15822784810126583,
|
| 147 |
+
"grad_norm": 0.5216394066810608,
|
| 148 |
+
"learning_rate": 2.9999634866395974e-05,
|
| 149 |
+
"loss": 1.2325,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.1661392405063291,
|
| 154 |
+
"grad_norm": 0.48694348335266113,
|
| 155 |
+
"learning_rate": 2.9998151541595854e-05,
|
| 156 |
+
"loss": 1.1987,
|
| 157 |
+
"step": 105
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.17405063291139242,
|
| 161 |
+
"grad_norm": 0.614189624786377,
|
| 162 |
+
"learning_rate": 2.9995527317498098e-05,
|
| 163 |
+
"loss": 1.1697,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.1819620253164557,
|
| 168 |
+
"grad_norm": 0.5633684992790222,
|
| 169 |
+
"learning_rate": 2.999176239372568e-05,
|
| 170 |
+
"loss": 1.2195,
|
| 171 |
+
"step": 115
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.189873417721519,
|
| 175 |
+
"grad_norm": 0.5750209093093872,
|
| 176 |
+
"learning_rate": 2.998685705667381e-05,
|
| 177 |
+
"loss": 1.1112,
|
| 178 |
+
"step": 120
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.19778481012658228,
|
| 182 |
+
"grad_norm": 0.5884466767311096,
|
| 183 |
+
"learning_rate": 2.9980811679488186e-05,
|
| 184 |
+
"loss": 1.1197,
|
| 185 |
+
"step": 125
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.20569620253164558,
|
| 189 |
+
"grad_norm": 0.5795807242393494,
|
| 190 |
+
"learning_rate": 2.997362672203657e-05,
|
| 191 |
+
"loss": 1.0747,
|
| 192 |
+
"step": 130
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.21360759493670886,
|
| 196 |
+
"grad_norm": 0.5831396579742432,
|
| 197 |
+
"learning_rate": 2.9965302730873856e-05,
|
| 198 |
+
"loss": 1.1612,
|
| 199 |
+
"step": 135
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.22151898734177214,
|
| 203 |
+
"grad_norm": 0.6353974342346191,
|
| 204 |
+
"learning_rate": 2.9955840339200444e-05,
|
| 205 |
+
"loss": 1.1168,
|
| 206 |
+
"step": 140
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.22943037974683544,
|
| 210 |
+
"grad_norm": 0.7513188719749451,
|
| 211 |
+
"learning_rate": 2.9945240266814103e-05,
|
| 212 |
+
"loss": 1.1157,
|
| 213 |
+
"step": 145
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.23734177215189872,
|
| 217 |
+
"grad_norm": 0.660714328289032,
|
| 218 |
+
"learning_rate": 2.9933503320055216e-05,
|
| 219 |
+
"loss": 1.0538,
|
| 220 |
+
"step": 150
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.24525316455696203,
|
| 224 |
+
"grad_norm": 0.6214624047279358,
|
| 225 |
+
"learning_rate": 2.9920630391745415e-05,
|
| 226 |
+
"loss": 1.0613,
|
| 227 |
+
"step": 155
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.25316455696202533,
|
| 231 |
+
"grad_norm": 0.6323642730712891,
|
| 232 |
+
"learning_rate": 2.9906622461119698e-05,
|
| 233 |
+
"loss": 1.0448,
|
| 234 |
+
"step": 160
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2610759493670886,
|
| 238 |
+
"grad_norm": 0.7148507833480835,
|
| 239 |
+
"learning_rate": 2.9891480593751934e-05,
|
| 240 |
+
"loss": 1.0055,
|
| 241 |
+
"step": 165
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2689873417721519,
|
| 245 |
+
"grad_norm": 0.8210672736167908,
|
| 246 |
+
"learning_rate": 2.9875205941473775e-05,
|
| 247 |
+
"loss": 1.0646,
|
| 248 |
+
"step": 170
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.27689873417721517,
|
| 252 |
+
"grad_norm": 0.6863754987716675,
|
| 253 |
+
"learning_rate": 2.985779974228708e-05,
|
| 254 |
+
"loss": 0.978,
|
| 255 |
+
"step": 175
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.2848101265822785,
|
| 259 |
+
"grad_norm": 0.8001527190208435,
|
| 260 |
+
"learning_rate": 2.98392633202697e-05,
|
| 261 |
+
"loss": 0.9764,
|
| 262 |
+
"step": 180
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2927215189873418,
|
| 266 |
+
"grad_norm": 0.6926974654197693,
|
| 267 |
+
"learning_rate": 2.98195980854748e-05,
|
| 268 |
+
"loss": 0.9681,
|
| 269 |
+
"step": 185
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.30063291139240506,
|
| 273 |
+
"grad_norm": 0.8066754341125488,
|
| 274 |
+
"learning_rate": 2.9798805533823555e-05,
|
| 275 |
+
"loss": 1.0118,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.30854430379746833,
|
| 280 |
+
"grad_norm": 0.8412411212921143,
|
| 281 |
+
"learning_rate": 2.977688724699138e-05,
|
| 282 |
+
"loss": 0.9111,
|
| 283 |
+
"step": 195
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.31645569620253167,
|
| 287 |
+
"grad_norm": 0.8765762448310852,
|
| 288 |
+
"learning_rate": 2.9753844892287607e-05,
|
| 289 |
+
"loss": 0.9032,
|
| 290 |
+
"step": 200
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.32436708860759494,
|
| 294 |
+
"grad_norm": 0.8607110977172852,
|
| 295 |
+
"learning_rate": 2.972968022252866e-05,
|
| 296 |
+
"loss": 0.8985,
|
| 297 |
+
"step": 205
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.3322784810126582,
|
| 301 |
+
"grad_norm": 0.9015821218490601,
|
| 302 |
+
"learning_rate": 2.9704395075904712e-05,
|
| 303 |
+
"loss": 0.8853,
|
| 304 |
+
"step": 210
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3401898734177215,
|
| 308 |
+
"grad_norm": 0.9376387000083923,
|
| 309 |
+
"learning_rate": 2.967799137583985e-05,
|
| 310 |
+
"loss": 0.8995,
|
| 311 |
+
"step": 215
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.34810126582278483,
|
| 315 |
+
"grad_norm": 0.8817639946937561,
|
| 316 |
+
"learning_rate": 2.9650471130845774e-05,
|
| 317 |
+
"loss": 0.877,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3560126582278481,
|
| 322 |
+
"grad_norm": 0.8314262628555298,
|
| 323 |
+
"learning_rate": 2.9621836434369016e-05,
|
| 324 |
+
"loss": 0.8613,
|
| 325 |
+
"step": 225
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3639240506329114,
|
| 329 |
+
"grad_norm": 0.8700135350227356,
|
| 330 |
+
"learning_rate": 2.9592089464631664e-05,
|
| 331 |
+
"loss": 0.8429,
|
| 332 |
+
"step": 230
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.37183544303797467,
|
| 336 |
+
"grad_norm": 1.0095256567001343,
|
| 337 |
+
"learning_rate": 2.95612324844657e-05,
|
| 338 |
+
"loss": 0.8467,
|
| 339 |
+
"step": 235
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.379746835443038,
|
| 343 |
+
"grad_norm": 0.9629245400428772,
|
| 344 |
+
"learning_rate": 2.952926784114084e-05,
|
| 345 |
+
"loss": 0.8143,
|
| 346 |
+
"step": 240
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3876582278481013,
|
| 350 |
+
"grad_norm": 0.9581960439682007,
|
| 351 |
+
"learning_rate": 2.9496197966185996e-05,
|
| 352 |
+
"loss": 0.8074,
|
| 353 |
+
"step": 245
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.39556962025316456,
|
| 357 |
+
"grad_norm": 1.0179094076156616,
|
| 358 |
+
"learning_rate": 2.9462025375204315e-05,
|
| 359 |
+
"loss": 0.8255,
|
| 360 |
+
"step": 250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.40348101265822783,
|
| 364 |
+
"grad_norm": 0.8479974865913391,
|
| 365 |
+
"learning_rate": 2.9426752667681798e-05,
|
| 366 |
+
"loss": 0.7955,
|
| 367 |
+
"step": 255
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.41139240506329117,
|
| 371 |
+
"grad_norm": 0.9385902881622314,
|
| 372 |
+
"learning_rate": 2.9390382526789576e-05,
|
| 373 |
+
"loss": 0.8103,
|
| 374 |
+
"step": 260
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.41930379746835444,
|
| 378 |
+
"grad_norm": 0.9720443487167358,
|
| 379 |
+
"learning_rate": 2.9352917719179797e-05,
|
| 380 |
+
"loss": 0.7167,
|
| 381 |
+
"step": 265
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.4272151898734177,
|
| 385 |
+
"grad_norm": 0.9628554582595825,
|
| 386 |
+
"learning_rate": 2.931436109477518e-05,
|
| 387 |
+
"loss": 0.7611,
|
| 388 |
+
"step": 270
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.435126582278481,
|
| 392 |
+
"grad_norm": 0.9764614105224609,
|
| 393 |
+
"learning_rate": 2.9274715586552193e-05,
|
| 394 |
+
"loss": 0.8169,
|
| 395 |
+
"step": 275
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4430379746835443,
|
| 399 |
+
"grad_norm": 1.021030306816101,
|
| 400 |
+
"learning_rate": 2.9233984210317984e-05,
|
| 401 |
+
"loss": 0.8143,
|
| 402 |
+
"step": 280
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4509493670886076,
|
| 406 |
+
"grad_norm": 1.0237860679626465,
|
| 407 |
+
"learning_rate": 2.919217006448093e-05,
|
| 408 |
+
"loss": 0.7984,
|
| 409 |
+
"step": 285
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4588607594936709,
|
| 413 |
+
"grad_norm": 1.0391464233398438,
|
| 414 |
+
"learning_rate": 2.9149276329814985e-05,
|
| 415 |
+
"loss": 0.7593,
|
| 416 |
+
"step": 290
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.46677215189873417,
|
| 420 |
+
"grad_norm": 1.0567541122436523,
|
| 421 |
+
"learning_rate": 2.910530626921768e-05,
|
| 422 |
+
"loss": 0.7064,
|
| 423 |
+
"step": 295
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.47468354430379744,
|
| 427 |
+
"grad_norm": 1.0766088962554932,
|
| 428 |
+
"learning_rate": 2.906026322746195e-05,
|
| 429 |
+
"loss": 0.7244,
|
| 430 |
+
"step": 300
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.4825949367088608,
|
| 434 |
+
"grad_norm": 1.0170453786849976,
|
| 435 |
+
"learning_rate": 2.9014150630941673e-05,
|
| 436 |
+
"loss": 0.7243,
|
| 437 |
+
"step": 305
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.49050632911392406,
|
| 441 |
+
"grad_norm": 0.989050030708313,
|
| 442 |
+
"learning_rate": 2.8966971987411046e-05,
|
| 443 |
+
"loss": 0.7843,
|
| 444 |
+
"step": 310
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.49841772151898733,
|
| 448 |
+
"grad_norm": 1.0374884605407715,
|
| 449 |
+
"learning_rate": 2.8918730885717742e-05,
|
| 450 |
+
"loss": 0.7,
|
| 451 |
+
"step": 315
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5063291139240507,
|
| 455 |
+
"grad_norm": 1.2468342781066895,
|
| 456 |
+
"learning_rate": 2.886943099552991e-05,
|
| 457 |
+
"loss": 0.698,
|
| 458 |
+
"step": 320
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5142405063291139,
|
| 462 |
+
"grad_norm": 1.1388477087020874,
|
| 463 |
+
"learning_rate": 2.881907606705703e-05,
|
| 464 |
+
"loss": 0.7607,
|
| 465 |
+
"step": 325
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5221518987341772,
|
| 469 |
+
"grad_norm": 1.2193673849105835,
|
| 470 |
+
"learning_rate": 2.876766993076464e-05,
|
| 471 |
+
"loss": 0.6712,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5300632911392406,
|
| 476 |
+
"grad_norm": 0.9723927974700928,
|
| 477 |
+
"learning_rate": 2.8715216497082938e-05,
|
| 478 |
+
"loss": 0.6643,
|
| 479 |
+
"step": 335
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5379746835443038,
|
| 483 |
+
"grad_norm": 1.0612977743148804,
|
| 484 |
+
"learning_rate": 2.866171975610934e-05,
|
| 485 |
+
"loss": 0.7147,
|
| 486 |
+
"step": 340
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5458860759493671,
|
| 490 |
+
"grad_norm": 1.221621036529541,
|
| 491 |
+
"learning_rate": 2.8607183777304934e-05,
|
| 492 |
+
"loss": 0.6723,
|
| 493 |
+
"step": 345
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5537974683544303,
|
| 497 |
+
"grad_norm": 1.2126332521438599,
|
| 498 |
+
"learning_rate": 2.8551612709184944e-05,
|
| 499 |
+
"loss": 0.6722,
|
| 500 |
+
"step": 350
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5617088607594937,
|
| 504 |
+
"grad_norm": 1.043410301208496,
|
| 505 |
+
"learning_rate": 2.8495010779003135e-05,
|
| 506 |
+
"loss": 0.6914,
|
| 507 |
+
"step": 355
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.569620253164557,
|
| 511 |
+
"grad_norm": 1.0380438566207886,
|
| 512 |
+
"learning_rate": 2.843738229243026e-05,
|
| 513 |
+
"loss": 0.6426,
|
| 514 |
+
"step": 360
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.5775316455696202,
|
| 518 |
+
"grad_norm": 1.0474764108657837,
|
| 519 |
+
"learning_rate": 2.8378731633226515e-05,
|
| 520 |
+
"loss": 0.6279,
|
| 521 |
+
"step": 365
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.5854430379746836,
|
| 525 |
+
"grad_norm": 1.1480685472488403,
|
| 526 |
+
"learning_rate": 2.8319063262908097e-05,
|
| 527 |
+
"loss": 0.6834,
|
| 528 |
+
"step": 370
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.5933544303797469,
|
| 532 |
+
"grad_norm": 1.0916494131088257,
|
| 533 |
+
"learning_rate": 2.825838172040778e-05,
|
| 534 |
+
"loss": 0.629,
|
| 535 |
+
"step": 375
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6012658227848101,
|
| 539 |
+
"grad_norm": 1.1295806169509888,
|
| 540 |
+
"learning_rate": 2.8196691621729694e-05,
|
| 541 |
+
"loss": 0.5996,
|
| 542 |
+
"step": 380
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6091772151898734,
|
| 546 |
+
"grad_norm": 1.1031495332717896,
|
| 547 |
+
"learning_rate": 2.8133997659598133e-05,
|
| 548 |
+
"loss": 0.5765,
|
| 549 |
+
"step": 385
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6170886075949367,
|
| 553 |
+
"grad_norm": 1.151245355606079,
|
| 554 |
+
"learning_rate": 2.8070304603100624e-05,
|
| 555 |
+
"loss": 0.6304,
|
| 556 |
+
"step": 390
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.625,
|
| 560 |
+
"grad_norm": 1.0607008934020996,
|
| 561 |
+
"learning_rate": 2.8005617297325117e-05,
|
| 562 |
+
"loss": 0.5824,
|
| 563 |
+
"step": 395
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6329113924050633,
|
| 567 |
+
"grad_norm": 0.9911614656448364,
|
| 568 |
+
"learning_rate": 2.7939940662991447e-05,
|
| 569 |
+
"loss": 0.6231,
|
| 570 |
+
"step": 400
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.6408227848101266,
|
| 574 |
+
"grad_norm": 1.1462503671646118,
|
| 575 |
+
"learning_rate": 2.7873279696077004e-05,
|
| 576 |
+
"loss": 0.5796,
|
| 577 |
+
"step": 405
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6487341772151899,
|
| 581 |
+
"grad_norm": 1.183719515800476,
|
| 582 |
+
"learning_rate": 2.7805639467436694e-05,
|
| 583 |
+
"loss": 0.5602,
|
| 584 |
+
"step": 410
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6566455696202531,
|
| 588 |
+
"grad_norm": 14.371760368347168,
|
| 589 |
+
"learning_rate": 2.773702512241721e-05,
|
| 590 |
+
"loss": 0.5845,
|
| 591 |
+
"step": 415
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.6645569620253164,
|
| 595 |
+
"grad_norm": 1.139249563217163,
|
| 596 |
+
"learning_rate": 2.7667441880465603e-05,
|
| 597 |
+
"loss": 0.5642,
|
| 598 |
+
"step": 420
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.6724683544303798,
|
| 602 |
+
"grad_norm": 1.1947458982467651,
|
| 603 |
+
"learning_rate": 2.7596895034732287e-05,
|
| 604 |
+
"loss": 0.5651,
|
| 605 |
+
"step": 425
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.680379746835443,
|
| 609 |
+
"grad_norm": 1.1384472846984863,
|
| 610 |
+
"learning_rate": 2.7525389951668356e-05,
|
| 611 |
+
"loss": 0.5662,
|
| 612 |
+
"step": 430
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.6882911392405063,
|
| 616 |
+
"grad_norm": 1.2073407173156738,
|
| 617 |
+
"learning_rate": 2.745293207061737e-05,
|
| 618 |
+
"loss": 0.537,
|
| 619 |
+
"step": 435
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.6962025316455697,
|
| 623 |
+
"grad_norm": 1.1086006164550781,
|
| 624 |
+
"learning_rate": 2.737952690340159e-05,
|
| 625 |
+
"loss": 0.5471,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7041139240506329,
|
| 630 |
+
"grad_norm": 1.1883656978607178,
|
| 631 |
+
"learning_rate": 2.7305180033902703e-05,
|
| 632 |
+
"loss": 0.5178,
|
| 633 |
+
"step": 445
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7120253164556962,
|
| 637 |
+
"grad_norm": 1.1278756856918335,
|
| 638 |
+
"learning_rate": 2.7229897117637064e-05,
|
| 639 |
+
"loss": 0.5397,
|
| 640 |
+
"step": 450
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7199367088607594,
|
| 644 |
+
"grad_norm": 1.2886579036712646,
|
| 645 |
+
"learning_rate": 2.7153683881325453e-05,
|
| 646 |
+
"loss": 0.5206,
|
| 647 |
+
"step": 455
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.7278481012658228,
|
| 651 |
+
"grad_norm": 1.0516884326934814,
|
| 652 |
+
"learning_rate": 2.7076546122457494e-05,
|
| 653 |
+
"loss": 0.4953,
|
| 654 |
+
"step": 460
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7357594936708861,
|
| 658 |
+
"grad_norm": 1.3163795471191406,
|
| 659 |
+
"learning_rate": 2.69984897088506e-05,
|
| 660 |
+
"loss": 0.5091,
|
| 661 |
+
"step": 465
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7436708860759493,
|
| 665 |
+
"grad_norm": 1.1612937450408936,
|
| 666 |
+
"learning_rate": 2.6919520578203643e-05,
|
| 667 |
+
"loss": 0.4506,
|
| 668 |
+
"step": 470
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7515822784810127,
|
| 672 |
+
"grad_norm": 1.3387264013290405,
|
| 673 |
+
"learning_rate": 2.6839644737645273e-05,
|
| 674 |
+
"loss": 0.4863,
|
| 675 |
+
"step": 475
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.759493670886076,
|
| 679 |
+
"grad_norm": 1.2922239303588867,
|
| 680 |
+
"learning_rate": 2.6758868263276934e-05,
|
| 681 |
+
"loss": 0.5172,
|
| 682 |
+
"step": 480
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.7674050632911392,
|
| 686 |
+
"grad_norm": 1.4410349130630493,
|
| 687 |
+
"learning_rate": 2.667719729971069e-05,
|
| 688 |
+
"loss": 0.5294,
|
| 689 |
+
"step": 485
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.7753164556962026,
|
| 693 |
+
"grad_norm": 1.0973846912384033,
|
| 694 |
+
"learning_rate": 2.6594638059601797e-05,
|
| 695 |
+
"loss": 0.5064,
|
| 696 |
+
"step": 490
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.7832278481012658,
|
| 700 |
+
"grad_norm": 1.2511711120605469,
|
| 701 |
+
"learning_rate": 2.6511196823176102e-05,
|
| 702 |
+
"loss": 0.5345,
|
| 703 |
+
"step": 495
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.7911392405063291,
|
| 707 |
+
"grad_norm": 1.2438185214996338,
|
| 708 |
+
"learning_rate": 2.6426879937752336e-05,
|
| 709 |
+
"loss": 0.4948,
|
| 710 |
+
"step": 500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.7990506329113924,
|
| 714 |
+
"grad_norm": 1.194139003753662,
|
| 715 |
+
"learning_rate": 2.6341693817259237e-05,
|
| 716 |
+
"loss": 0.4678,
|
| 717 |
+
"step": 505
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8069620253164557,
|
| 721 |
+
"grad_norm": 1.2999294996261597,
|
| 722 |
+
"learning_rate": 2.6255644941747684e-05,
|
| 723 |
+
"loss": 0.4683,
|
| 724 |
+
"step": 510
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.814873417721519,
|
| 728 |
+
"grad_norm": 1.2776968479156494,
|
| 729 |
+
"learning_rate": 2.616873985689775e-05,
|
| 730 |
+
"loss": 0.4891,
|
| 731 |
+
"step": 515
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8227848101265823,
|
| 735 |
+
"grad_norm": 1.1751092672348022,
|
| 736 |
+
"learning_rate": 2.608098517352077e-05,
|
| 737 |
+
"loss": 0.4846,
|
| 738 |
+
"step": 520
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8306962025316456,
|
| 742 |
+
"grad_norm": 1.2429425716400146,
|
| 743 |
+
"learning_rate": 2.5992387567056472e-05,
|
| 744 |
+
"loss": 0.4316,
|
| 745 |
+
"step": 525
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8386075949367089,
|
| 749 |
+
"grad_norm": 1.1492716073989868,
|
| 750 |
+
"learning_rate": 2.5902953777065177e-05,
|
| 751 |
+
"loss": 0.468,
|
| 752 |
+
"step": 530
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8465189873417721,
|
| 756 |
+
"grad_norm": 1.2880111932754517,
|
| 757 |
+
"learning_rate": 2.5812690606715118e-05,
|
| 758 |
+
"loss": 0.4474,
|
| 759 |
+
"step": 535
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.8544303797468354,
|
| 763 |
+
"grad_norm": 1.0557680130004883,
|
| 764 |
+
"learning_rate": 2.5721604922264942e-05,
|
| 765 |
+
"loss": 0.4333,
|
| 766 |
+
"step": 540
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.8623417721518988,
|
| 770 |
+
"grad_norm": 1.2208117246627808,
|
| 771 |
+
"learning_rate": 2.5629703652541383e-05,
|
| 772 |
+
"loss": 0.4154,
|
| 773 |
+
"step": 545
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.870253164556962,
|
| 777 |
+
"grad_norm": 1.0802950859069824,
|
| 778 |
+
"learning_rate": 2.5536993788412206e-05,
|
| 779 |
+
"loss": 0.4324,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.8781645569620253,
|
| 784 |
+
"grad_norm": 1.4286012649536133,
|
| 785 |
+
"learning_rate": 2.5443482382254397e-05,
|
| 786 |
+
"loss": 0.4217,
|
| 787 |
+
"step": 555
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.8860759493670886,
|
| 791 |
+
"grad_norm": 1.2151390314102173,
|
| 792 |
+
"learning_rate": 2.5349176547417718e-05,
|
| 793 |
+
"loss": 0.404,
|
| 794 |
+
"step": 560
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.8939873417721519,
|
| 798 |
+
"grad_norm": 1.2180513143539429,
|
| 799 |
+
"learning_rate": 2.525408345768358e-05,
|
| 800 |
+
"loss": 0.4624,
|
| 801 |
+
"step": 565
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.9018987341772152,
|
| 805 |
+
"grad_norm": 1.2627403736114502,
|
| 806 |
+
"learning_rate": 2.5158210346719344e-05,
|
| 807 |
+
"loss": 0.3965,
|
| 808 |
+
"step": 570
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.9098101265822784,
|
| 812 |
+
"grad_norm": 1.307173490524292,
|
| 813 |
+
"learning_rate": 2.5061564507528053e-05,
|
| 814 |
+
"loss": 0.376,
|
| 815 |
+
"step": 575
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9177215189873418,
|
| 819 |
+
"grad_norm": 1.284371018409729,
|
| 820 |
+
"learning_rate": 2.4964153291893688e-05,
|
| 821 |
+
"loss": 0.4049,
|
| 822 |
+
"step": 580
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.9256329113924051,
|
| 826 |
+
"grad_norm": 1.166039228439331,
|
| 827 |
+
"learning_rate": 2.486598410982188e-05,
|
| 828 |
+
"loss": 0.3901,
|
| 829 |
+
"step": 585
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9335443037974683,
|
| 833 |
+
"grad_norm": 1.1340969800949097,
|
| 834 |
+
"learning_rate": 2.4767064428976265e-05,
|
| 835 |
+
"loss": 0.4218,
|
| 836 |
+
"step": 590
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.9414556962025317,
|
| 840 |
+
"grad_norm": 1.1339762210845947,
|
| 841 |
+
"learning_rate": 2.4667401774110424e-05,
|
| 842 |
+
"loss": 0.4135,
|
| 843 |
+
"step": 595
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.9493670886075949,
|
| 847 |
+
"grad_norm": 1.178755760192871,
|
| 848 |
+
"learning_rate": 2.456700372649545e-05,
|
| 849 |
+
"loss": 0.3973,
|
| 850 |
+
"step": 600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.9572784810126582,
|
| 854 |
+
"grad_norm": 1.2037619352340698,
|
| 855 |
+
"learning_rate": 2.4465877923343287e-05,
|
| 856 |
+
"loss": 0.3724,
|
| 857 |
+
"step": 605
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.9651898734177216,
|
| 861 |
+
"grad_norm": 1.219773292541504,
|
| 862 |
+
"learning_rate": 2.4364032057225756e-05,
|
| 863 |
+
"loss": 0.3758,
|
| 864 |
+
"step": 610
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.9731012658227848,
|
| 868 |
+
"grad_norm": 1.1657541990280151,
|
| 869 |
+
"learning_rate": 2.4261473875489366e-05,
|
| 870 |
+
"loss": 0.4084,
|
| 871 |
+
"step": 615
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.9810126582278481,
|
| 875 |
+
"grad_norm": 1.3129708766937256,
|
| 876 |
+
"learning_rate": 2.4158211179666007e-05,
|
| 877 |
+
"loss": 0.376,
|
| 878 |
+
"step": 620
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.9889240506329114,
|
| 882 |
+
"grad_norm": 1.0885345935821533,
|
| 883 |
+
"learning_rate": 2.405425182487948e-05,
|
| 884 |
+
"loss": 0.3447,
|
| 885 |
+
"step": 625
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.9968354430379747,
|
| 889 |
+
"grad_norm": 1.0967590808868408,
|
| 890 |
+
"learning_rate": 2.394960371924796e-05,
|
| 891 |
+
"loss": 0.3473,
|
| 892 |
+
"step": 630
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.004746835443038,
|
| 896 |
+
"grad_norm": 1.1879879236221313,
|
| 897 |
+
"learning_rate": 2.3844274823282445e-05,
|
| 898 |
+
"loss": 0.3246,
|
| 899 |
+
"step": 635
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.0126582278481013,
|
| 903 |
+
"grad_norm": 1.2285014390945435,
|
| 904 |
+
"learning_rate": 2.3738273149281184e-05,
|
| 905 |
+
"loss": 0.3124,
|
| 906 |
+
"step": 640
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.0205696202531644,
|
| 910 |
+
"grad_norm": 1.1759848594665527,
|
| 911 |
+
"learning_rate": 2.3631606760720214e-05,
|
| 912 |
+
"loss": 0.3169,
|
| 913 |
+
"step": 645
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.0284810126582278,
|
| 917 |
+
"grad_norm": 1.3058689832687378,
|
| 918 |
+
"learning_rate": 2.3524283771639945e-05,
|
| 919 |
+
"loss": 0.2962,
|
| 920 |
+
"step": 650
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.0363924050632911,
|
| 924 |
+
"grad_norm": 1.1710305213928223,
|
| 925 |
+
"learning_rate": 2.3416312346027956e-05,
|
| 926 |
+
"loss": 0.3053,
|
| 927 |
+
"step": 655
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.0443037974683544,
|
| 931 |
+
"grad_norm": 1.251835823059082,
|
| 932 |
+
"learning_rate": 2.330770069719795e-05,
|
| 933 |
+
"loss": 0.2704,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.0522151898734178,
|
| 938 |
+
"grad_norm": 1.0766069889068604,
|
| 939 |
+
"learning_rate": 2.3198457087164993e-05,
|
| 940 |
+
"loss": 0.2962,
|
| 941 |
+
"step": 665
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.0601265822784811,
|
| 945 |
+
"grad_norm": 1.2666971683502197,
|
| 946 |
+
"learning_rate": 2.3088589826016997e-05,
|
| 947 |
+
"loss": 0.2698,
|
| 948 |
+
"step": 670
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.0680379746835442,
|
| 952 |
+
"grad_norm": 1.1509121656417847,
|
| 953 |
+
"learning_rate": 2.297810727128261e-05,
|
| 954 |
+
"loss": 0.2551,
|
| 955 |
+
"step": 675
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.0759493670886076,
|
| 959 |
+
"grad_norm": 1.197072148323059,
|
| 960 |
+
"learning_rate": 2.286701782729544e-05,
|
| 961 |
+
"loss": 0.2793,
|
| 962 |
+
"step": 680
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.0838607594936709,
|
| 966 |
+
"grad_norm": 1.1826645135879517,
|
| 967 |
+
"learning_rate": 2.275532994455475e-05,
|
| 968 |
+
"loss": 0.2753,
|
| 969 |
+
"step": 685
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.0917721518987342,
|
| 973 |
+
"grad_norm": 1.2750444412231445,
|
| 974 |
+
"learning_rate": 2.2643052119082646e-05,
|
| 975 |
+
"loss": 0.2889,
|
| 976 |
+
"step": 690
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.0996835443037976,
|
| 980 |
+
"grad_norm": 1.1425836086273193,
|
| 981 |
+
"learning_rate": 2.2530192891777783e-05,
|
| 982 |
+
"loss": 0.2901,
|
| 983 |
+
"step": 695
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.1075949367088607,
|
| 987 |
+
"grad_norm": 1.225294589996338,
|
| 988 |
+
"learning_rate": 2.2416760847765633e-05,
|
| 989 |
+
"loss": 0.2981,
|
| 990 |
+
"step": 700
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.115506329113924,
|
| 994 |
+
"grad_norm": 1.1347010135650635,
|
| 995 |
+
"learning_rate": 2.23027646157455e-05,
|
| 996 |
+
"loss": 0.2452,
|
| 997 |
+
"step": 705
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.1234177215189873,
|
| 1001 |
+
"grad_norm": 1.017078161239624,
|
| 1002 |
+
"learning_rate": 2.2188212867334038e-05,
|
| 1003 |
+
"loss": 0.2415,
|
| 1004 |
+
"step": 710
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.1313291139240507,
|
| 1008 |
+
"grad_norm": 1.0773953199386597,
|
| 1009 |
+
"learning_rate": 2.2073114316405694e-05,
|
| 1010 |
+
"loss": 0.2623,
|
| 1011 |
+
"step": 715
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.139240506329114,
|
| 1015 |
+
"grad_norm": 1.362718939781189,
|
| 1016 |
+
"learning_rate": 2.19574777184298e-05,
|
| 1017 |
+
"loss": 0.2936,
|
| 1018 |
+
"step": 720
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.1471518987341773,
|
| 1022 |
+
"grad_norm": 1.1041392087936401,
|
| 1023 |
+
"learning_rate": 2.1841311869804575e-05,
|
| 1024 |
+
"loss": 0.2649,
|
| 1025 |
+
"step": 725
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.1550632911392404,
|
| 1029 |
+
"grad_norm": 1.2117396593093872,
|
| 1030 |
+
"learning_rate": 2.1724625607187966e-05,
|
| 1031 |
+
"loss": 0.2295,
|
| 1032 |
+
"step": 730
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.1629746835443038,
|
| 1036 |
+
"grad_norm": 1.1867276430130005,
|
| 1037 |
+
"learning_rate": 2.1607427806825472e-05,
|
| 1038 |
+
"loss": 0.2209,
|
| 1039 |
+
"step": 735
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.1708860759493671,
|
| 1043 |
+
"grad_norm": 1.0909227132797241,
|
| 1044 |
+
"learning_rate": 2.1489727383874914e-05,
|
| 1045 |
+
"loss": 0.2453,
|
| 1046 |
+
"step": 740
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.1787974683544304,
|
| 1050 |
+
"grad_norm": 1.1393877267837524,
|
| 1051 |
+
"learning_rate": 2.137153329172829e-05,
|
| 1052 |
+
"loss": 0.2268,
|
| 1053 |
+
"step": 745
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.1867088607594938,
|
| 1057 |
+
"grad_norm": 1.1258888244628906,
|
| 1058 |
+
"learning_rate": 2.125285452133067e-05,
|
| 1059 |
+
"loss": 0.2558,
|
| 1060 |
+
"step": 750
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.1946202531645569,
|
| 1064 |
+
"grad_norm": 1.1085089445114136,
|
| 1065 |
+
"learning_rate": 2.1133700100496275e-05,
|
| 1066 |
+
"loss": 0.2232,
|
| 1067 |
+
"step": 755
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.2025316455696202,
|
| 1071 |
+
"grad_norm": 1.1286346912384033,
|
| 1072 |
+
"learning_rate": 2.1014079093221744e-05,
|
| 1073 |
+
"loss": 0.2655,
|
| 1074 |
+
"step": 760
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.2104430379746836,
|
| 1078 |
+
"grad_norm": 1.0915958881378174,
|
| 1079 |
+
"learning_rate": 2.0894000598996608e-05,
|
| 1080 |
+
"loss": 0.2577,
|
| 1081 |
+
"step": 765
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.2183544303797469,
|
| 1085 |
+
"grad_norm": 1.1641857624053955,
|
| 1086 |
+
"learning_rate": 2.077347375211115e-05,
|
| 1087 |
+
"loss": 0.2708,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.2262658227848102,
|
| 1092 |
+
"grad_norm": 1.4235131740570068,
|
| 1093 |
+
"learning_rate": 2.0652507720961527e-05,
|
| 1094 |
+
"loss": 0.2649,
|
| 1095 |
+
"step": 775
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.2341772151898733,
|
| 1099 |
+
"grad_norm": 1.0940698385238647,
|
| 1100 |
+
"learning_rate": 2.0531111707352343e-05,
|
| 1101 |
+
"loss": 0.2579,
|
| 1102 |
+
"step": 780
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.2420886075949367,
|
| 1106 |
+
"grad_norm": 1.0031784772872925,
|
| 1107 |
+
"learning_rate": 2.0409294945796685e-05,
|
| 1108 |
+
"loss": 0.239,
|
| 1109 |
+
"step": 785
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.25,
|
| 1113 |
+
"grad_norm": 1.3176120519638062,
|
| 1114 |
+
"learning_rate": 2.0287066702813656e-05,
|
| 1115 |
+
"loss": 0.2292,
|
| 1116 |
+
"step": 790
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.2579113924050633,
|
| 1120 |
+
"grad_norm": 1.1103475093841553,
|
| 1121 |
+
"learning_rate": 2.0164436276223463e-05,
|
| 1122 |
+
"loss": 0.2587,
|
| 1123 |
+
"step": 795
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.2658227848101267,
|
| 1127 |
+
"grad_norm": 1.3331197500228882,
|
| 1128 |
+
"learning_rate": 2.0041412994440157e-05,
|
| 1129 |
+
"loss": 0.2298,
|
| 1130 |
+
"step": 800
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.2737341772151898,
|
| 1134 |
+
"grad_norm": 1.2051442861557007,
|
| 1135 |
+
"learning_rate": 1.9918006215762024e-05,
|
| 1136 |
+
"loss": 0.2155,
|
| 1137 |
+
"step": 805
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.2816455696202531,
|
| 1141 |
+
"grad_norm": 1.064636468887329,
|
| 1142 |
+
"learning_rate": 1.9794225327659686e-05,
|
| 1143 |
+
"loss": 0.225,
|
| 1144 |
+
"step": 810
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.2895569620253164,
|
| 1148 |
+
"grad_norm": 1.0860430002212524,
|
| 1149 |
+
"learning_rate": 1.9670079746062035e-05,
|
| 1150 |
+
"loss": 0.2326,
|
| 1151 |
+
"step": 815
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.2974683544303798,
|
| 1155 |
+
"grad_norm": 1.1983343362808228,
|
| 1156 |
+
"learning_rate": 1.9545578914639937e-05,
|
| 1157 |
+
"loss": 0.2098,
|
| 1158 |
+
"step": 820
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.3053797468354431,
|
| 1162 |
+
"grad_norm": 1.080658197402954,
|
| 1163 |
+
"learning_rate": 1.942073230408788e-05,
|
| 1164 |
+
"loss": 0.2165,
|
| 1165 |
+
"step": 825
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.3132911392405062,
|
| 1169 |
+
"grad_norm": 1.0960582494735718,
|
| 1170 |
+
"learning_rate": 1.9295549411403536e-05,
|
| 1171 |
+
"loss": 0.2236,
|
| 1172 |
+
"step": 830
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.3212025316455696,
|
| 1176 |
+
"grad_norm": 1.2715785503387451,
|
| 1177 |
+
"learning_rate": 1.917003975916534e-05,
|
| 1178 |
+
"loss": 0.2144,
|
| 1179 |
+
"step": 835
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.3291139240506329,
|
| 1183 |
+
"grad_norm": 1.1728013753890991,
|
| 1184 |
+
"learning_rate": 1.9044212894808095e-05,
|
| 1185 |
+
"loss": 0.2144,
|
| 1186 |
+
"step": 840
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.3370253164556962,
|
| 1190 |
+
"grad_norm": 1.3844877481460571,
|
| 1191 |
+
"learning_rate": 1.891807838989672e-05,
|
| 1192 |
+
"loss": 0.2217,
|
| 1193 |
+
"step": 845
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.3449367088607596,
|
| 1197 |
+
"grad_norm": 1.1015386581420898,
|
| 1198 |
+
"learning_rate": 1.8791645839398146e-05,
|
| 1199 |
+
"loss": 0.1967,
|
| 1200 |
+
"step": 850
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.3528481012658227,
|
| 1204 |
+
"grad_norm": 1.0521535873413086,
|
| 1205 |
+
"learning_rate": 1.866492486095144e-05,
|
| 1206 |
+
"loss": 0.1866,
|
| 1207 |
+
"step": 855
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.360759493670886,
|
| 1211 |
+
"grad_norm": 1.3538264036178589,
|
| 1212 |
+
"learning_rate": 1.8537925094136168e-05,
|
| 1213 |
+
"loss": 0.1861,
|
| 1214 |
+
"step": 860
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.3686708860759493,
|
| 1218 |
+
"grad_norm": 1.1505900621414185,
|
| 1219 |
+
"learning_rate": 1.8410656199739154e-05,
|
| 1220 |
+
"loss": 0.2243,
|
| 1221 |
+
"step": 865
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.3765822784810127,
|
| 1225 |
+
"grad_norm": 1.1845885515213013,
|
| 1226 |
+
"learning_rate": 1.828312785901958e-05,
|
| 1227 |
+
"loss": 0.1958,
|
| 1228 |
+
"step": 870
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.384493670886076,
|
| 1232 |
+
"grad_norm": 1.110446572303772,
|
| 1233 |
+
"learning_rate": 1.8155349772972523e-05,
|
| 1234 |
+
"loss": 0.2149,
|
| 1235 |
+
"step": 875
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.3924050632911391,
|
| 1239 |
+
"grad_norm": 0.9138773083686829,
|
| 1240 |
+
"learning_rate": 1.8027331661591022e-05,
|
| 1241 |
+
"loss": 0.2124,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.4003164556962027,
|
| 1246 |
+
"grad_norm": 1.0536731481552124,
|
| 1247 |
+
"learning_rate": 1.7899083263126695e-05,
|
| 1248 |
+
"loss": 0.2106,
|
| 1249 |
+
"step": 885
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.4082278481012658,
|
| 1253 |
+
"grad_norm": 1.18684720993042,
|
| 1254 |
+
"learning_rate": 1.777061433334892e-05,
|
| 1255 |
+
"loss": 0.1925,
|
| 1256 |
+
"step": 890
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.4161392405063291,
|
| 1260 |
+
"grad_norm": 1.0312713384628296,
|
| 1261 |
+
"learning_rate": 1.764193464480276e-05,
|
| 1262 |
+
"loss": 0.2043,
|
| 1263 |
+
"step": 895
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.4240506329113924,
|
| 1267 |
+
"grad_norm": 1.134125828742981,
|
| 1268 |
+
"learning_rate": 1.7513053986065564e-05,
|
| 1269 |
+
"loss": 0.181,
|
| 1270 |
+
"step": 900
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.4319620253164558,
|
| 1274 |
+
"grad_norm": 1.2018568515777588,
|
| 1275 |
+
"learning_rate": 1.738398216100232e-05,
|
| 1276 |
+
"loss": 0.1961,
|
| 1277 |
+
"step": 905
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.439873417721519,
|
| 1281 |
+
"grad_norm": 1.0200194120407104,
|
| 1282 |
+
"learning_rate": 1.7254728988019937e-05,
|
| 1283 |
+
"loss": 0.2006,
|
| 1284 |
+
"step": 910
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.4477848101265822,
|
| 1288 |
+
"grad_norm": 1.2849904298782349,
|
| 1289 |
+
"learning_rate": 1.7125304299320326e-05,
|
| 1290 |
+
"loss": 0.1978,
|
| 1291 |
+
"step": 915
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.4556962025316456,
|
| 1295 |
+
"grad_norm": 0.9776687622070312,
|
| 1296 |
+
"learning_rate": 1.699571794015248e-05,
|
| 1297 |
+
"loss": 0.1867,
|
| 1298 |
+
"step": 920
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.4636075949367089,
|
| 1302 |
+
"grad_norm": 1.0699931383132935,
|
| 1303 |
+
"learning_rate": 1.6865979768063564e-05,
|
| 1304 |
+
"loss": 0.1593,
|
| 1305 |
+
"step": 925
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.4715189873417722,
|
| 1309 |
+
"grad_norm": 1.171876072883606,
|
| 1310 |
+
"learning_rate": 1.6736099652149048e-05,
|
| 1311 |
+
"loss": 0.2134,
|
| 1312 |
+
"step": 930
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.4794303797468356,
|
| 1316 |
+
"grad_norm": 1.076789140701294,
|
| 1317 |
+
"learning_rate": 1.6606087472301968e-05,
|
| 1318 |
+
"loss": 0.1757,
|
| 1319 |
+
"step": 935
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.4873417721518987,
|
| 1323 |
+
"grad_norm": 1.2642792463302612,
|
| 1324 |
+
"learning_rate": 1.6475953118461384e-05,
|
| 1325 |
+
"loss": 0.1871,
|
| 1326 |
+
"step": 940
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.495253164556962,
|
| 1330 |
+
"grad_norm": 1.043723464012146,
|
| 1331 |
+
"learning_rate": 1.6345706489860033e-05,
|
| 1332 |
+
"loss": 0.1926,
|
| 1333 |
+
"step": 945
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.5031645569620253,
|
| 1337 |
+
"grad_norm": 1.1900668144226074,
|
| 1338 |
+
"learning_rate": 1.6215357494271334e-05,
|
| 1339 |
+
"loss": 0.1927,
|
| 1340 |
+
"step": 950
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.5110759493670884,
|
| 1344 |
+
"grad_norm": 1.327351689338684,
|
| 1345 |
+
"learning_rate": 1.6084916047255693e-05,
|
| 1346 |
+
"loss": 0.1903,
|
| 1347 |
+
"step": 955
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.518987341772152,
|
| 1351 |
+
"grad_norm": 1.1277024745941162,
|
| 1352 |
+
"learning_rate": 1.595439207140622e-05,
|
| 1353 |
+
"loss": 0.1959,
|
| 1354 |
+
"step": 960
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.5268987341772151,
|
| 1358 |
+
"grad_norm": 0.9847940802574158,
|
| 1359 |
+
"learning_rate": 1.5823795495593955e-05,
|
| 1360 |
+
"loss": 0.1639,
|
| 1361 |
+
"step": 965
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.5348101265822784,
|
| 1365 |
+
"grad_norm": 1.1352672576904297,
|
| 1366 |
+
"learning_rate": 1.5693136254212557e-05,
|
| 1367 |
+
"loss": 0.1509,
|
| 1368 |
+
"step": 970
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.5427215189873418,
|
| 1372 |
+
"grad_norm": 1.0179147720336914,
|
| 1373 |
+
"learning_rate": 1.556242428642261e-05,
|
| 1374 |
+
"loss": 0.175,
|
| 1375 |
+
"step": 975
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.5506329113924051,
|
| 1379 |
+
"grad_norm": 0.9815003871917725,
|
| 1380 |
+
"learning_rate": 1.5431669535395576e-05,
|
| 1381 |
+
"loss": 0.1641,
|
| 1382 |
+
"step": 980
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.5585443037974684,
|
| 1386 |
+
"grad_norm": 0.983579695224762,
|
| 1387 |
+
"learning_rate": 1.530088194755739e-05,
|
| 1388 |
+
"loss": 0.1563,
|
| 1389 |
+
"step": 985
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.5664556962025316,
|
| 1393 |
+
"grad_norm": 1.1190534830093384,
|
| 1394 |
+
"learning_rate": 1.5170071471831875e-05,
|
| 1395 |
+
"loss": 0.1704,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.5743670886075949,
|
| 1400 |
+
"grad_norm": 1.1313245296478271,
|
| 1401 |
+
"learning_rate": 1.5039248058883916e-05,
|
| 1402 |
+
"loss": 0.1937,
|
| 1403 |
+
"step": 995
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.5822784810126582,
|
| 1407 |
+
"grad_norm": 1.093643069267273,
|
| 1408 |
+
"learning_rate": 1.490842166036252e-05,
|
| 1409 |
+
"loss": 0.1596,
|
| 1410 |
+
"step": 1000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.5901898734177216,
|
| 1414 |
+
"grad_norm": 1.124353289604187,
|
| 1415 |
+
"learning_rate": 1.477760222814381e-05,
|
| 1416 |
+
"loss": 0.1556,
|
| 1417 |
+
"step": 1005
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.5981012658227849,
|
| 1421 |
+
"grad_norm": 0.9633249044418335,
|
| 1422 |
+
"learning_rate": 1.464679971357399e-05,
|
| 1423 |
+
"loss": 0.1622,
|
| 1424 |
+
"step": 1010
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.606012658227848,
|
| 1428 |
+
"grad_norm": 0.8686103224754333,
|
| 1429 |
+
"learning_rate": 1.4516024066712336e-05,
|
| 1430 |
+
"loss": 0.1315,
|
| 1431 |
+
"step": 1015
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.6139240506329116,
|
| 1435 |
+
"grad_norm": 1.0397095680236816,
|
| 1436 |
+
"learning_rate": 1.4385285235574325e-05,
|
| 1437 |
+
"loss": 0.1502,
|
| 1438 |
+
"step": 1020
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.6218354430379747,
|
| 1442 |
+
"grad_norm": 1.152609944343567,
|
| 1443 |
+
"learning_rate": 1.4254593165374895e-05,
|
| 1444 |
+
"loss": 0.1826,
|
| 1445 |
+
"step": 1025
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.629746835443038,
|
| 1449 |
+
"grad_norm": 0.9158191680908203,
|
| 1450 |
+
"learning_rate": 1.4123957797771891e-05,
|
| 1451 |
+
"loss": 0.144,
|
| 1452 |
+
"step": 1030
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.6376582278481013,
|
| 1456 |
+
"grad_norm": 1.0531560182571411,
|
| 1457 |
+
"learning_rate": 1.3993389070109848e-05,
|
| 1458 |
+
"loss": 0.1405,
|
| 1459 |
+
"step": 1035
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.6455696202531644,
|
| 1463 |
+
"grad_norm": 1.0512011051177979,
|
| 1464 |
+
"learning_rate": 1.3862896914664037e-05,
|
| 1465 |
+
"loss": 0.1637,
|
| 1466 |
+
"step": 1040
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.653481012658228,
|
| 1470 |
+
"grad_norm": 1.1094757318496704,
|
| 1471 |
+
"learning_rate": 1.3732491257884927e-05,
|
| 1472 |
+
"loss": 0.1422,
|
| 1473 |
+
"step": 1045
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.6613924050632911,
|
| 1477 |
+
"grad_norm": 0.8569801449775696,
|
| 1478 |
+
"learning_rate": 1.3602182019643103e-05,
|
| 1479 |
+
"loss": 0.1411,
|
| 1480 |
+
"step": 1050
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.6693037974683544,
|
| 1484 |
+
"grad_norm": 1.129196047782898,
|
| 1485 |
+
"learning_rate": 1.3471979112474652e-05,
|
| 1486 |
+
"loss": 0.1625,
|
| 1487 |
+
"step": 1055
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.6772151898734178,
|
| 1491 |
+
"grad_norm": 0.9935137033462524,
|
| 1492 |
+
"learning_rate": 1.3341892440827123e-05,
|
| 1493 |
+
"loss": 0.1371,
|
| 1494 |
+
"step": 1060
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.685126582278481,
|
| 1498 |
+
"grad_norm": 1.135963797569275,
|
| 1499 |
+
"learning_rate": 1.3211931900306118e-05,
|
| 1500 |
+
"loss": 0.1685,
|
| 1501 |
+
"step": 1065
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.6930379746835444,
|
| 1505 |
+
"grad_norm": 1.0298207998275757,
|
| 1506 |
+
"learning_rate": 1.3082107376922513e-05,
|
| 1507 |
+
"loss": 0.1445,
|
| 1508 |
+
"step": 1070
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.7009493670886076,
|
| 1512 |
+
"grad_norm": 1.242588996887207,
|
| 1513 |
+
"learning_rate": 1.2952428746340467e-05,
|
| 1514 |
+
"loss": 0.1395,
|
| 1515 |
+
"step": 1075
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.7088607594936709,
|
| 1519 |
+
"grad_norm": 0.946211040019989,
|
| 1520 |
+
"learning_rate": 1.2822905873126169e-05,
|
| 1521 |
+
"loss": 0.1439,
|
| 1522 |
+
"step": 1080
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.7167721518987342,
|
| 1526 |
+
"grad_norm": 1.136708378791809,
|
| 1527 |
+
"learning_rate": 1.2693548609997439e-05,
|
| 1528 |
+
"loss": 0.1506,
|
| 1529 |
+
"step": 1085
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.7246835443037973,
|
| 1533 |
+
"grad_norm": 0.8927178382873535,
|
| 1534 |
+
"learning_rate": 1.256436679707426e-05,
|
| 1535 |
+
"loss": 0.1336,
|
| 1536 |
+
"step": 1090
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.7325949367088609,
|
| 1540 |
+
"grad_norm": 0.9015725255012512,
|
| 1541 |
+
"learning_rate": 1.2435370261130236e-05,
|
| 1542 |
+
"loss": 0.1213,
|
| 1543 |
+
"step": 1095
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.740506329113924,
|
| 1547 |
+
"grad_norm": 1.0154457092285156,
|
| 1548 |
+
"learning_rate": 1.2306568814845068e-05,
|
| 1549 |
+
"loss": 0.119,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.7484177215189873,
|
| 1554 |
+
"grad_norm": 0.8864551782608032,
|
| 1555 |
+
"learning_rate": 1.2177972256058125e-05,
|
| 1556 |
+
"loss": 0.1398,
|
| 1557 |
+
"step": 1105
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.7563291139240507,
|
| 1561 |
+
"grad_norm": 0.9897282719612122,
|
| 1562 |
+
"learning_rate": 1.2049590367023116e-05,
|
| 1563 |
+
"loss": 0.1219,
|
| 1564 |
+
"step": 1110
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.7642405063291138,
|
| 1568 |
+
"grad_norm": 1.0648620128631592,
|
| 1569 |
+
"learning_rate": 1.1921432913663976e-05,
|
| 1570 |
+
"loss": 0.1171,
|
| 1571 |
+
"step": 1115
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.7721518987341773,
|
| 1575 |
+
"grad_norm": 0.927117645740509,
|
| 1576 |
+
"learning_rate": 1.1793509644831964e-05,
|
| 1577 |
+
"loss": 0.1255,
|
| 1578 |
+
"step": 1120
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.7800632911392404,
|
| 1582 |
+
"grad_norm": 0.9205576181411743,
|
| 1583 |
+
"learning_rate": 1.166583029156407e-05,
|
| 1584 |
+
"loss": 0.1298,
|
| 1585 |
+
"step": 1125
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.7879746835443038,
|
| 1589 |
+
"grad_norm": 0.9967123866081238,
|
| 1590 |
+
"learning_rate": 1.1538404566342807e-05,
|
| 1591 |
+
"loss": 0.1399,
|
| 1592 |
+
"step": 1130
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.7958860759493671,
|
| 1596 |
+
"grad_norm": 1.3310614824295044,
|
| 1597 |
+
"learning_rate": 1.1411242162357363e-05,
|
| 1598 |
+
"loss": 0.133,
|
| 1599 |
+
"step": 1135
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.8037974683544302,
|
| 1603 |
+
"grad_norm": 0.822165846824646,
|
| 1604 |
+
"learning_rate": 1.128435275276626e-05,
|
| 1605 |
+
"loss": 0.136,
|
| 1606 |
+
"step": 1140
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.8117088607594938,
|
| 1610 |
+
"grad_norm": 1.0309679508209229,
|
| 1611 |
+
"learning_rate": 1.1157745989961517e-05,
|
| 1612 |
+
"loss": 0.1271,
|
| 1613 |
+
"step": 1145
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.8196202531645569,
|
| 1617 |
+
"grad_norm": 0.8331815004348755,
|
| 1618 |
+
"learning_rate": 1.1031431504834409e-05,
|
| 1619 |
+
"loss": 0.1149,
|
| 1620 |
+
"step": 1150
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.8275316455696202,
|
| 1624 |
+
"grad_norm": 0.9345864653587341,
|
| 1625 |
+
"learning_rate": 1.0905418906042836e-05,
|
| 1626 |
+
"loss": 0.1329,
|
| 1627 |
+
"step": 1155
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.8354430379746836,
|
| 1631 |
+
"grad_norm": 0.9099761843681335,
|
| 1632 |
+
"learning_rate": 1.0779717779280418e-05,
|
| 1633 |
+
"loss": 0.1245,
|
| 1634 |
+
"step": 1160
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.8433544303797469,
|
| 1638 |
+
"grad_norm": 0.9058949947357178,
|
| 1639 |
+
"learning_rate": 1.0654337686547294e-05,
|
| 1640 |
+
"loss": 0.1215,
|
| 1641 |
+
"step": 1165
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.8512658227848102,
|
| 1645 |
+
"grad_norm": 1.0564491748809814,
|
| 1646 |
+
"learning_rate": 1.052928816542277e-05,
|
| 1647 |
+
"loss": 0.1283,
|
| 1648 |
+
"step": 1170
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.8591772151898733,
|
| 1652 |
+
"grad_norm": 0.9570179581642151,
|
| 1653 |
+
"learning_rate": 1.0404578728339796e-05,
|
| 1654 |
+
"loss": 0.1151,
|
| 1655 |
+
"step": 1175
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.8670886075949367,
|
| 1659 |
+
"grad_norm": 0.9448406100273132,
|
| 1660 |
+
"learning_rate": 1.0280218861861343e-05,
|
| 1661 |
+
"loss": 0.1094,
|
| 1662 |
+
"step": 1180
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.875,
|
| 1666 |
+
"grad_norm": 0.899459719657898,
|
| 1667 |
+
"learning_rate": 1.0156218025958792e-05,
|
| 1668 |
+
"loss": 0.1109,
|
| 1669 |
+
"step": 1185
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.8829113924050633,
|
| 1673 |
+
"grad_norm": 0.9216104745864868,
|
| 1674 |
+
"learning_rate": 1.0032585653292313e-05,
|
| 1675 |
+
"loss": 0.1098,
|
| 1676 |
+
"step": 1190
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.8908227848101267,
|
| 1680 |
+
"grad_norm": 0.8242044448852539,
|
| 1681 |
+
"learning_rate": 9.909331148493314e-06,
|
| 1682 |
+
"loss": 0.1318,
|
| 1683 |
+
"step": 1195
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.8987341772151898,
|
| 1687 |
+
"grad_norm": 0.7547638416290283,
|
| 1688 |
+
"learning_rate": 9.786463887449052e-06,
|
| 1689 |
+
"loss": 0.1098,
|
| 1690 |
+
"step": 1200
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.9066455696202531,
|
| 1694 |
+
"grad_norm": 0.9817409515380859,
|
| 1695 |
+
"learning_rate": 9.663993216589425e-06,
|
| 1696 |
+
"loss": 0.0951,
|
| 1697 |
+
"step": 1205
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.9145569620253164,
|
| 1701 |
+
"grad_norm": 0.8632509112358093,
|
| 1702 |
+
"learning_rate": 9.541928452175963e-06,
|
| 1703 |
+
"loss": 0.1175,
|
| 1704 |
+
"step": 1210
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.9224683544303798,
|
| 1708 |
+
"grad_norm": 0.9336681365966797,
|
| 1709 |
+
"learning_rate": 9.420278879593175e-06,
|
| 1710 |
+
"loss": 0.1144,
|
| 1711 |
+
"step": 1215
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.9303797468354431,
|
| 1715 |
+
"grad_norm": 0.9144463539123535,
|
| 1716 |
+
"learning_rate": 9.299053752642215e-06,
|
| 1717 |
+
"loss": 0.1191,
|
| 1718 |
+
"step": 1220
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.9382911392405062,
|
| 1722 |
+
"grad_norm": 0.8816940784454346,
|
| 1723 |
+
"learning_rate": 9.178262292836926e-06,
|
| 1724 |
+
"loss": 0.1119,
|
| 1725 |
+
"step": 1225
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.9462025316455698,
|
| 1729 |
+
"grad_norm": 0.9447281360626221,
|
| 1730 |
+
"learning_rate": 9.057913688702395e-06,
|
| 1731 |
+
"loss": 0.1202,
|
| 1732 |
+
"step": 1230
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.9541139240506329,
|
| 1736 |
+
"grad_norm": 0.9763627052307129,
|
| 1737 |
+
"learning_rate": 8.93801709507597e-06,
|
| 1738 |
+
"loss": 0.1063,
|
| 1739 |
+
"step": 1235
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.9620253164556962,
|
| 1743 |
+
"grad_norm": 1.1537237167358398,
|
| 1744 |
+
"learning_rate": 8.81858163241087e-06,
|
| 1745 |
+
"loss": 0.1212,
|
| 1746 |
+
"step": 1240
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.9699367088607596,
|
| 1750 |
+
"grad_norm": 1.01103675365448,
|
| 1751 |
+
"learning_rate": 8.699616386082389e-06,
|
| 1752 |
+
"loss": 0.1062,
|
| 1753 |
+
"step": 1245
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.9778481012658227,
|
| 1757 |
+
"grad_norm": 1.0035678148269653,
|
| 1758 |
+
"learning_rate": 8.581130405696783e-06,
|
| 1759 |
+
"loss": 0.0981,
|
| 1760 |
+
"step": 1250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.9857594936708862,
|
| 1764 |
+
"grad_norm": 0.7947705984115601,
|
| 1765 |
+
"learning_rate": 8.463132704402863e-06,
|
| 1766 |
+
"loss": 0.0935,
|
| 1767 |
+
"step": 1255
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.9936708860759493,
|
| 1771 |
+
"grad_norm": 1.1136794090270996,
|
| 1772 |
+
"learning_rate": 8.345632258206377e-06,
|
| 1773 |
+
"loss": 0.1194,
|
| 1774 |
+
"step": 1260
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.0015822784810124,
|
| 1778 |
+
"grad_norm": 0.6469590067863464,
|
| 1779 |
+
"learning_rate": 8.228638005287227e-06,
|
| 1780 |
+
"loss": 0.0922,
|
| 1781 |
+
"step": 1265
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.009493670886076,
|
| 1785 |
+
"grad_norm": 0.8095146417617798,
|
| 1786 |
+
"learning_rate": 8.112158845319525e-06,
|
| 1787 |
+
"loss": 0.09,
|
| 1788 |
+
"step": 1270
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.017405063291139,
|
| 1792 |
+
"grad_norm": 0.8335313200950623,
|
| 1793 |
+
"learning_rate": 7.996203638794596e-06,
|
| 1794 |
+
"loss": 0.0805,
|
| 1795 |
+
"step": 1275
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.0253164556962027,
|
| 1799 |
+
"grad_norm": 0.8516942858695984,
|
| 1800 |
+
"learning_rate": 7.880781206347001e-06,
|
| 1801 |
+
"loss": 0.0869,
|
| 1802 |
+
"step": 1280
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.0332278481012658,
|
| 1806 |
+
"grad_norm": 0.9114158749580383,
|
| 1807 |
+
"learning_rate": 7.765900328083524e-06,
|
| 1808 |
+
"loss": 0.0881,
|
| 1809 |
+
"step": 1285
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.041139240506329,
|
| 1813 |
+
"grad_norm": 0.8066091537475586,
|
| 1814 |
+
"learning_rate": 7.651569742915276e-06,
|
| 1815 |
+
"loss": 0.0811,
|
| 1816 |
+
"step": 1290
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.0490506329113924,
|
| 1820 |
+
"grad_norm": 0.8260509967803955,
|
| 1821 |
+
"learning_rate": 7.53779814789297e-06,
|
| 1822 |
+
"loss": 0.0765,
|
| 1823 |
+
"step": 1295
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.0569620253164556,
|
| 1827 |
+
"grad_norm": 0.9944354295730591,
|
| 1828 |
+
"learning_rate": 7.424594197545289e-06,
|
| 1829 |
+
"loss": 0.0911,
|
| 1830 |
+
"step": 1300
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.064873417721519,
|
| 1834 |
+
"grad_norm": 0.7893029451370239,
|
| 1835 |
+
"learning_rate": 7.311966503220574e-06,
|
| 1836 |
+
"loss": 0.0775,
|
| 1837 |
+
"step": 1305
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.0727848101265822,
|
| 1841 |
+
"grad_norm": 0.7460876703262329,
|
| 1842 |
+
"learning_rate": 7.199923632431772e-06,
|
| 1843 |
+
"loss": 0.0835,
|
| 1844 |
+
"step": 1310
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.0806962025316458,
|
| 1848 |
+
"grad_norm": 0.6351117491722107,
|
| 1849 |
+
"learning_rate": 7.088474108204679e-06,
|
| 1850 |
+
"loss": 0.083,
|
| 1851 |
+
"step": 1315
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.088607594936709,
|
| 1855 |
+
"grad_norm": 0.8060542345046997,
|
| 1856 |
+
"learning_rate": 6.97762640842962e-06,
|
| 1857 |
+
"loss": 0.0848,
|
| 1858 |
+
"step": 1320
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.096518987341772,
|
| 1862 |
+
"grad_norm": 0.7928087711334229,
|
| 1863 |
+
"learning_rate": 6.8673889652165535e-06,
|
| 1864 |
+
"loss": 0.0795,
|
| 1865 |
+
"step": 1325
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.1044303797468356,
|
| 1869 |
+
"grad_norm": 0.8051559329032898,
|
| 1870 |
+
"learning_rate": 6.757770164253616e-06,
|
| 1871 |
+
"loss": 0.0808,
|
| 1872 |
+
"step": 1330
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.1123417721518987,
|
| 1876 |
+
"grad_norm": 0.6816345453262329,
|
| 1877 |
+
"learning_rate": 6.648778344169246e-06,
|
| 1878 |
+
"loss": 0.0811,
|
| 1879 |
+
"step": 1335
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.1202531645569622,
|
| 1883 |
+
"grad_norm": 0.7813927531242371,
|
| 1884 |
+
"learning_rate": 6.540421795897873e-06,
|
| 1885 |
+
"loss": 0.0866,
|
| 1886 |
+
"step": 1340
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.1281645569620253,
|
| 1890 |
+
"grad_norm": 0.7709259390830994,
|
| 1891 |
+
"learning_rate": 6.432708762049237e-06,
|
| 1892 |
+
"loss": 0.0846,
|
| 1893 |
+
"step": 1345
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.1360759493670884,
|
| 1897 |
+
"grad_norm": 0.8114832043647766,
|
| 1898 |
+
"learning_rate": 6.32564743628135e-06,
|
| 1899 |
+
"loss": 0.0793,
|
| 1900 |
+
"step": 1350
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.143987341772152,
|
| 1904 |
+
"grad_norm": 0.728865921497345,
|
| 1905 |
+
"learning_rate": 6.21924596267723e-06,
|
| 1906 |
+
"loss": 0.0871,
|
| 1907 |
+
"step": 1355
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.151898734177215,
|
| 1911 |
+
"grad_norm": 0.6699401140213013,
|
| 1912 |
+
"learning_rate": 6.113512435125406e-06,
|
| 1913 |
+
"loss": 0.0818,
|
| 1914 |
+
"step": 1360
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.1598101265822787,
|
| 1918 |
+
"grad_norm": 0.8308571577072144,
|
| 1919 |
+
"learning_rate": 6.008454896704177e-06,
|
| 1920 |
+
"loss": 0.0795,
|
| 1921 |
+
"step": 1365
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.1677215189873418,
|
| 1925 |
+
"grad_norm": 0.6745468378067017,
|
| 1926 |
+
"learning_rate": 5.9040813390698075e-06,
|
| 1927 |
+
"loss": 0.0824,
|
| 1928 |
+
"step": 1370
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.175632911392405,
|
| 1932 |
+
"grad_norm": 0.7378197312355042,
|
| 1933 |
+
"learning_rate": 5.8003997018486195e-06,
|
| 1934 |
+
"loss": 0.0733,
|
| 1935 |
+
"step": 1375
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.1835443037974684,
|
| 1939 |
+
"grad_norm": 0.6287952065467834,
|
| 1940 |
+
"learning_rate": 5.697417872033e-06,
|
| 1941 |
+
"loss": 0.0787,
|
| 1942 |
+
"step": 1380
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.1914556962025316,
|
| 1946 |
+
"grad_norm": 0.5483230948448181,
|
| 1947 |
+
"learning_rate": 5.595143683381453e-06,
|
| 1948 |
+
"loss": 0.0696,
|
| 1949 |
+
"step": 1385
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.199367088607595,
|
| 1953 |
+
"grad_norm": 0.7113633751869202,
|
| 1954 |
+
"learning_rate": 5.493584915822716e-06,
|
| 1955 |
+
"loss": 0.0756,
|
| 1956 |
+
"step": 1390
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.2072784810126582,
|
| 1960 |
+
"grad_norm": 0.7312512397766113,
|
| 1961 |
+
"learning_rate": 5.392749294863906e-06,
|
| 1962 |
+
"loss": 0.0855,
|
| 1963 |
+
"step": 1395
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.2151898734177213,
|
| 1967 |
+
"grad_norm": 0.8007563352584839,
|
| 1968 |
+
"learning_rate": 5.292644491002868e-06,
|
| 1969 |
+
"loss": 0.0852,
|
| 1970 |
+
"step": 1400
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.223101265822785,
|
| 1974 |
+
"grad_norm": 0.6663233041763306,
|
| 1975 |
+
"learning_rate": 5.193278119144693e-06,
|
| 1976 |
+
"loss": 0.0704,
|
| 1977 |
+
"step": 1405
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.231012658227848,
|
| 1981 |
+
"grad_norm": 0.875812828540802,
|
| 1982 |
+
"learning_rate": 5.094657738022436e-06,
|
| 1983 |
+
"loss": 0.0743,
|
| 1984 |
+
"step": 1410
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.2389240506329116,
|
| 1988 |
+
"grad_norm": 0.7085732817649841,
|
| 1989 |
+
"learning_rate": 4.996790849622136e-06,
|
| 1990 |
+
"loss": 0.0816,
|
| 1991 |
+
"step": 1415
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.2468354430379747,
|
| 1995 |
+
"grad_norm": 0.8398520946502686,
|
| 1996 |
+
"learning_rate": 4.899684898612159e-06,
|
| 1997 |
+
"loss": 0.0688,
|
| 1998 |
+
"step": 1420
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.254746835443038,
|
| 2002 |
+
"grad_norm": 0.8422523140907288,
|
| 2003 |
+
"learning_rate": 4.803347271776869e-06,
|
| 2004 |
+
"loss": 0.0755,
|
| 2005 |
+
"step": 1425
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.2626582278481013,
|
| 2009 |
+
"grad_norm": 0.9676128625869751,
|
| 2010 |
+
"learning_rate": 4.707785297454721e-06,
|
| 2011 |
+
"loss": 0.0753,
|
| 2012 |
+
"step": 1430
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.2705696202531644,
|
| 2016 |
+
"grad_norm": 0.7076261639595032,
|
| 2017 |
+
"learning_rate": 4.613006244980821e-06,
|
| 2018 |
+
"loss": 0.0693,
|
| 2019 |
+
"step": 1435
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.278481012658228,
|
| 2023 |
+
"grad_norm": 0.6433950066566467,
|
| 2024 |
+
"learning_rate": 4.519017324133921e-06,
|
| 2025 |
+
"loss": 0.0755,
|
| 2026 |
+
"step": 1440
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.286392405063291,
|
| 2030 |
+
"grad_norm": 0.5876809358596802,
|
| 2031 |
+
"learning_rate": 4.4258256845880006e-06,
|
| 2032 |
+
"loss": 0.0825,
|
| 2033 |
+
"step": 1445
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.2943037974683547,
|
| 2037 |
+
"grad_norm": 0.6657724380493164,
|
| 2038 |
+
"learning_rate": 4.333438415368378e-06,
|
| 2039 |
+
"loss": 0.0701,
|
| 2040 |
+
"step": 1450
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.3022151898734178,
|
| 2044 |
+
"grad_norm": 0.9290494322776794,
|
| 2045 |
+
"learning_rate": 4.2418625443124786e-06,
|
| 2046 |
+
"loss": 0.0786,
|
| 2047 |
+
"step": 1455
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.310126582278481,
|
| 2051 |
+
"grad_norm": 0.6781948804855347,
|
| 2052 |
+
"learning_rate": 4.151105037535201e-06,
|
| 2053 |
+
"loss": 0.0765,
|
| 2054 |
+
"step": 1460
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.3180379746835444,
|
| 2058 |
+
"grad_norm": 0.6075677275657654,
|
| 2059 |
+
"learning_rate": 4.061172798899022e-06,
|
| 2060 |
+
"loss": 0.0695,
|
| 2061 |
+
"step": 1465
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.3259493670886076,
|
| 2065 |
+
"grad_norm": 0.640425980091095,
|
| 2066 |
+
"learning_rate": 3.97207266948884e-06,
|
| 2067 |
+
"loss": 0.0689,
|
| 2068 |
+
"step": 1470
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.333860759493671,
|
| 2072 |
+
"grad_norm": 0.5523871779441833,
|
| 2073 |
+
"learning_rate": 3.883811427091555e-06,
|
| 2074 |
+
"loss": 0.0684,
|
| 2075 |
+
"step": 1475
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.3417721518987342,
|
| 2079 |
+
"grad_norm": 0.6521956324577332,
|
| 2080 |
+
"learning_rate": 3.7963957856804928e-06,
|
| 2081 |
+
"loss": 0.0718,
|
| 2082 |
+
"step": 1480
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.3496835443037973,
|
| 2086 |
+
"grad_norm": 0.6425510048866272,
|
| 2087 |
+
"learning_rate": 3.70983239490469e-06,
|
| 2088 |
+
"loss": 0.0704,
|
| 2089 |
+
"step": 1485
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.357594936708861,
|
| 2093 |
+
"grad_norm": 0.740044116973877,
|
| 2094 |
+
"learning_rate": 3.6241278395830484e-06,
|
| 2095 |
+
"loss": 0.0643,
|
| 2096 |
+
"step": 1490
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.365506329113924,
|
| 2100 |
+
"grad_norm": 0.7017863988876343,
|
| 2101 |
+
"learning_rate": 3.5392886392034252e-06,
|
| 2102 |
+
"loss": 0.0761,
|
| 2103 |
+
"step": 1495
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.3734177215189876,
|
| 2107 |
+
"grad_norm": 0.8681387305259705,
|
| 2108 |
+
"learning_rate": 3.4553212474267245e-06,
|
| 2109 |
+
"loss": 0.0647,
|
| 2110 |
+
"step": 1500
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.3813291139240507,
|
| 2114 |
+
"grad_norm": 0.7004177570343018,
|
| 2115 |
+
"learning_rate": 3.372232051595947e-06,
|
| 2116 |
+
"loss": 0.0654,
|
| 2117 |
+
"step": 1505
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.3892405063291138,
|
| 2121 |
+
"grad_norm": 0.7884745001792908,
|
| 2122 |
+
"learning_rate": 3.2900273722503173e-06,
|
| 2123 |
+
"loss": 0.0704,
|
| 2124 |
+
"step": 1510
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.3971518987341773,
|
| 2128 |
+
"grad_norm": 0.7360832691192627,
|
| 2129 |
+
"learning_rate": 3.2087134626444936e-06,
|
| 2130 |
+
"loss": 0.0709,
|
| 2131 |
+
"step": 1515
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.4050632911392404,
|
| 2135 |
+
"grad_norm": 0.6051527857780457,
|
| 2136 |
+
"learning_rate": 3.128296508272875e-06,
|
| 2137 |
+
"loss": 0.072,
|
| 2138 |
+
"step": 1520
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.412974683544304,
|
| 2142 |
+
"grad_norm": 0.6216411590576172,
|
| 2143 |
+
"learning_rate": 3.0487826263990713e-06,
|
| 2144 |
+
"loss": 0.0755,
|
| 2145 |
+
"step": 1525
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.420886075949367,
|
| 2149 |
+
"grad_norm": 0.697020947933197,
|
| 2150 |
+
"learning_rate": 2.9701778655905954e-06,
|
| 2151 |
+
"loss": 0.0767,
|
| 2152 |
+
"step": 1530
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.4287974683544302,
|
| 2156 |
+
"grad_norm": 0.655642569065094,
|
| 2157 |
+
"learning_rate": 2.892488205258711e-06,
|
| 2158 |
+
"loss": 0.0631,
|
| 2159 |
+
"step": 1535
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.4367088607594938,
|
| 2163 |
+
"grad_norm": 0.7078518271446228,
|
| 2164 |
+
"learning_rate": 2.8157195552036143e-06,
|
| 2165 |
+
"loss": 0.0691,
|
| 2166 |
+
"step": 1540
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.444620253164557,
|
| 2170 |
+
"grad_norm": 0.6857931017875671,
|
| 2171 |
+
"learning_rate": 2.7398777551648717e-06,
|
| 2172 |
+
"loss": 0.0723,
|
| 2173 |
+
"step": 1545
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.4525316455696204,
|
| 2177 |
+
"grad_norm": 0.6030269861221313,
|
| 2178 |
+
"learning_rate": 2.664968574377189e-06,
|
| 2179 |
+
"loss": 0.0637,
|
| 2180 |
+
"step": 1550
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.4604430379746836,
|
| 2184 |
+
"grad_norm": 0.6308946013450623,
|
| 2185 |
+
"learning_rate": 2.5909977111315448e-06,
|
| 2186 |
+
"loss": 0.0733,
|
| 2187 |
+
"step": 1555
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.4683544303797467,
|
| 2191 |
+
"grad_norm": 0.6207740306854248,
|
| 2192 |
+
"learning_rate": 2.5179707923417488e-06,
|
| 2193 |
+
"loss": 0.0663,
|
| 2194 |
+
"step": 1560
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.4762658227848102,
|
| 2198 |
+
"grad_norm": 0.550593912601471,
|
| 2199 |
+
"learning_rate": 2.4458933731163863e-06,
|
| 2200 |
+
"loss": 0.0712,
|
| 2201 |
+
"step": 1565
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.4841772151898733,
|
| 2205 |
+
"grad_norm": 0.5750855803489685,
|
| 2206 |
+
"learning_rate": 2.3747709363362447e-06,
|
| 2207 |
+
"loss": 0.064,
|
| 2208 |
+
"step": 1570
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.492088607594937,
|
| 2212 |
+
"grad_norm": 0.618537425994873,
|
| 2213 |
+
"learning_rate": 2.304608892237235e-06,
|
| 2214 |
+
"loss": 0.066,
|
| 2215 |
+
"step": 1575
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.5,
|
| 2219 |
+
"grad_norm": 0.6581313610076904,
|
| 2220 |
+
"learning_rate": 2.235412577998859e-06,
|
| 2221 |
+
"loss": 0.0656,
|
| 2222 |
+
"step": 1580
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.507911392405063,
|
| 2226 |
+
"grad_norm": 0.6456623673439026,
|
| 2227 |
+
"learning_rate": 2.1671872573381834e-06,
|
| 2228 |
+
"loss": 0.0667,
|
| 2229 |
+
"step": 1585
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.5158227848101267,
|
| 2233 |
+
"grad_norm": 0.606492280960083,
|
| 2234 |
+
"learning_rate": 2.0999381201094504e-06,
|
| 2235 |
+
"loss": 0.0641,
|
| 2236 |
+
"step": 1590
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.5237341772151898,
|
| 2240 |
+
"grad_norm": 0.7273917198181152,
|
| 2241 |
+
"learning_rate": 2.033670281909292e-06,
|
| 2242 |
+
"loss": 0.072,
|
| 2243 |
+
"step": 1595
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.5316455696202533,
|
| 2247 |
+
"grad_norm": 0.5226804614067078,
|
| 2248 |
+
"learning_rate": 1.9683887836875736e-06,
|
| 2249 |
+
"loss": 0.0648,
|
| 2250 |
+
"step": 1600
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.5395569620253164,
|
| 2254 |
+
"grad_norm": 0.5995578169822693,
|
| 2255 |
+
"learning_rate": 1.9040985913639446e-06,
|
| 2256 |
+
"loss": 0.0666,
|
| 2257 |
+
"step": 1605
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.5474683544303796,
|
| 2261 |
+
"grad_norm": 0.6878185868263245,
|
| 2262 |
+
"learning_rate": 1.8408045954500897e-06,
|
| 2263 |
+
"loss": 0.0743,
|
| 2264 |
+
"step": 1610
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.555379746835443,
|
| 2268 |
+
"grad_norm": 0.7470523118972778,
|
| 2269 |
+
"learning_rate": 1.7785116106776916e-06,
|
| 2270 |
+
"loss": 0.0726,
|
| 2271 |
+
"step": 1615
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.5632911392405062,
|
| 2275 |
+
"grad_norm": 0.6122822761535645,
|
| 2276 |
+
"learning_rate": 1.717224375632193e-06,
|
| 2277 |
+
"loss": 0.0675,
|
| 2278 |
+
"step": 1620
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 2.5712025316455698,
|
| 2282 |
+
"grad_norm": 0.5168817043304443,
|
| 2283 |
+
"learning_rate": 1.6569475523923356e-06,
|
| 2284 |
+
"loss": 0.0661,
|
| 2285 |
+
"step": 1625
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.579113924050633,
|
| 2289 |
+
"grad_norm": 0.6188611388206482,
|
| 2290 |
+
"learning_rate": 1.5976857261755112e-06,
|
| 2291 |
+
"loss": 0.0714,
|
| 2292 |
+
"step": 1630
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.587025316455696,
|
| 2296 |
+
"grad_norm": 0.6280317306518555,
|
| 2297 |
+
"learning_rate": 1.5394434049889662e-06,
|
| 2298 |
+
"loss": 0.0748,
|
| 2299 |
+
"step": 1635
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.5949367088607596,
|
| 2303 |
+
"grad_norm": 0.5373277068138123,
|
| 2304 |
+
"learning_rate": 1.4822250192868968e-06,
|
| 2305 |
+
"loss": 0.0688,
|
| 2306 |
+
"step": 1640
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 2.6028481012658227,
|
| 2310 |
+
"grad_norm": 0.679050087928772,
|
| 2311 |
+
"learning_rate": 1.4260349216334063e-06,
|
| 2312 |
+
"loss": 0.0664,
|
| 2313 |
+
"step": 1645
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.6107594936708862,
|
| 2317 |
+
"grad_norm": 0.5583394169807434,
|
| 2318 |
+
"learning_rate": 1.370877386371417e-06,
|
| 2319 |
+
"loss": 0.0653,
|
| 2320 |
+
"step": 1650
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.6186708860759493,
|
| 2324 |
+
"grad_norm": 0.5613365173339844,
|
| 2325 |
+
"learning_rate": 1.3167566092975359e-06,
|
| 2326 |
+
"loss": 0.0649,
|
| 2327 |
+
"step": 1655
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.6265822784810124,
|
| 2331 |
+
"grad_norm": 0.6472129821777344,
|
| 2332 |
+
"learning_rate": 1.2636767073428624e-06,
|
| 2333 |
+
"loss": 0.0739,
|
| 2334 |
+
"step": 1660
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.634493670886076,
|
| 2338 |
+
"grad_norm": 0.5926021337509155,
|
| 2339 |
+
"learning_rate": 1.2116417182598239e-06,
|
| 2340 |
+
"loss": 0.0633,
|
| 2341 |
+
"step": 1665
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.642405063291139,
|
| 2345 |
+
"grad_norm": 0.5478129386901855,
|
| 2346 |
+
"learning_rate": 1.160655600315038e-06,
|
| 2347 |
+
"loss": 0.0656,
|
| 2348 |
+
"step": 1670
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.6503164556962027,
|
| 2352 |
+
"grad_norm": 0.636114239692688,
|
| 2353 |
+
"learning_rate": 1.1107222319881942e-06,
|
| 2354 |
+
"loss": 0.0607,
|
| 2355 |
+
"step": 1675
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.6582278481012658,
|
| 2359 |
+
"grad_norm": 0.6562848687171936,
|
| 2360 |
+
"learning_rate": 1.0618454116770237e-06,
|
| 2361 |
+
"loss": 0.0612,
|
| 2362 |
+
"step": 1680
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.666139240506329,
|
| 2366 |
+
"grad_norm": 0.47253885865211487,
|
| 2367 |
+
"learning_rate": 1.0140288574083727e-06,
|
| 2368 |
+
"loss": 0.0548,
|
| 2369 |
+
"step": 1685
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.6740506329113924,
|
| 2373 |
+
"grad_norm": 0.5989126563072205,
|
| 2374 |
+
"learning_rate": 9.67276206555351e-07,
|
| 2375 |
+
"loss": 0.0722,
|
| 2376 |
+
"step": 1690
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.6819620253164556,
|
| 2380 |
+
"grad_norm": 0.6085955500602722,
|
| 2381 |
+
"learning_rate": 9.215910155606532e-07,
|
| 2382 |
+
"loss": 0.0662,
|
| 2383 |
+
"step": 1695
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.689873417721519,
|
| 2387 |
+
"grad_norm": 0.5660440325737,
|
| 2388 |
+
"learning_rate": 8.769767596660294e-07,
|
| 2389 |
+
"loss": 0.0611,
|
| 2390 |
+
"step": 1700
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.6977848101265822,
|
| 2394 |
+
"grad_norm": 0.54960036277771,
|
| 2395 |
+
"learning_rate": 8.334368326479075e-07,
|
| 2396 |
+
"loss": 0.0639,
|
| 2397 |
+
"step": 1705
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.7056962025316453,
|
| 2401 |
+
"grad_norm": 0.4610978364944458,
|
| 2402 |
+
"learning_rate": 7.909745465592444e-07,
|
| 2403 |
+
"loss": 0.054,
|
| 2404 |
+
"step": 1710
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 2.713607594936709,
|
| 2408 |
+
"grad_norm": 0.5121480822563171,
|
| 2409 |
+
"learning_rate": 7.495931314775723e-07,
|
| 2410 |
+
"loss": 0.0547,
|
| 2411 |
+
"step": 1715
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 2.721518987341772,
|
| 2415 |
+
"grad_norm": 0.6044265031814575,
|
| 2416 |
+
"learning_rate": 7.092957352593033e-07,
|
| 2417 |
+
"loss": 0.0779,
|
| 2418 |
+
"step": 1720
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 2.7294303797468356,
|
| 2422 |
+
"grad_norm": 0.6387259364128113,
|
| 2423 |
+
"learning_rate": 6.700854233002546e-07,
|
| 2424 |
+
"loss": 0.0628,
|
| 2425 |
+
"step": 1725
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.7373417721518987,
|
| 2429 |
+
"grad_norm": 0.5435252785682678,
|
| 2430 |
+
"learning_rate": 6.319651783024733e-07,
|
| 2431 |
+
"loss": 0.0666,
|
| 2432 |
+
"step": 1730
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 2.745253164556962,
|
| 2436 |
+
"grad_norm": 0.5887588262557983,
|
| 2437 |
+
"learning_rate": 5.949379000473593e-07,
|
| 2438 |
+
"loss": 0.0675,
|
| 2439 |
+
"step": 1735
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 2.7531645569620253,
|
| 2443 |
+
"grad_norm": 0.5682973265647888,
|
| 2444 |
+
"learning_rate": 5.590064051750521e-07,
|
| 2445 |
+
"loss": 0.0676,
|
| 2446 |
+
"step": 1740
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 2.7610759493670884,
|
| 2450 |
+
"grad_norm": 0.5745964050292969,
|
| 2451 |
+
"learning_rate": 5.241734269701953e-07,
|
| 2452 |
+
"loss": 0.0641,
|
| 2453 |
+
"step": 1745
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 2.768987341772152,
|
| 2457 |
+
"grad_norm": 0.6089118123054504,
|
| 2458 |
+
"learning_rate": 4.904416151540081e-07,
|
| 2459 |
+
"loss": 0.0689,
|
| 2460 |
+
"step": 1750
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 2.776898734177215,
|
| 2464 |
+
"grad_norm": 0.4897151291370392,
|
| 2465 |
+
"learning_rate": 4.578135356827212e-07,
|
| 2466 |
+
"loss": 0.0616,
|
| 2467 |
+
"step": 1755
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 2.7848101265822782,
|
| 2471 |
+
"grad_norm": 0.6580669283866882,
|
| 2472 |
+
"learning_rate": 4.2629167055238836e-07,
|
| 2473 |
+
"loss": 0.0699,
|
| 2474 |
+
"step": 1760
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 2.7927215189873418,
|
| 2478 |
+
"grad_norm": 0.6968865990638733,
|
| 2479 |
+
"learning_rate": 3.958784176100855e-07,
|
| 2480 |
+
"loss": 0.0672,
|
| 2481 |
+
"step": 1765
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 2.8006329113924053,
|
| 2485 |
+
"grad_norm": 0.5710925459861755,
|
| 2486 |
+
"learning_rate": 3.6657609037149986e-07,
|
| 2487 |
+
"loss": 0.066,
|
| 2488 |
+
"step": 1770
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 2.8085443037974684,
|
| 2492 |
+
"grad_norm": 0.624765932559967,
|
| 2493 |
+
"learning_rate": 3.383869178449528e-07,
|
| 2494 |
+
"loss": 0.0628,
|
| 2495 |
+
"step": 1775
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 2.8164556962025316,
|
| 2499 |
+
"grad_norm": 0.5765222311019897,
|
| 2500 |
+
"learning_rate": 3.1131304436183504e-07,
|
| 2501 |
+
"loss": 0.0632,
|
| 2502 |
+
"step": 1780
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 2.8243670886075947,
|
| 2506 |
+
"grad_norm": 0.49525973200798035,
|
| 2507 |
+
"learning_rate": 2.8535652941348746e-07,
|
| 2508 |
+
"loss": 0.0637,
|
| 2509 |
+
"step": 1785
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 2.8322784810126582,
|
| 2513 |
+
"grad_norm": 0.4840955436229706,
|
| 2514 |
+
"learning_rate": 2.605193474945394e-07,
|
| 2515 |
+
"loss": 0.0629,
|
| 2516 |
+
"step": 1790
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 2.8401898734177218,
|
| 2520 |
+
"grad_norm": 0.6750727891921997,
|
| 2521 |
+
"learning_rate": 2.368033879527093e-07,
|
| 2522 |
+
"loss": 0.0643,
|
| 2523 |
+
"step": 1795
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 2.848101265822785,
|
| 2527 |
+
"grad_norm": 0.5535368323326111,
|
| 2528 |
+
"learning_rate": 2.1421045484508571e-07,
|
| 2529 |
+
"loss": 0.0642,
|
| 2530 |
+
"step": 1800
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 2.856012658227848,
|
| 2534 |
+
"grad_norm": 0.5374280214309692,
|
| 2535 |
+
"learning_rate": 1.9274226680088935e-07,
|
| 2536 |
+
"loss": 0.0638,
|
| 2537 |
+
"step": 1805
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 2.8639240506329116,
|
| 2541 |
+
"grad_norm": 0.594165563583374,
|
| 2542 |
+
"learning_rate": 1.7240045689074392e-07,
|
| 2543 |
+
"loss": 0.0684,
|
| 2544 |
+
"step": 1810
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 2.8718354430379747,
|
| 2548 |
+
"grad_norm": 0.7364162802696228,
|
| 2549 |
+
"learning_rate": 1.531865725024406e-07,
|
| 2550 |
+
"loss": 0.075,
|
| 2551 |
+
"step": 1815
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 2.879746835443038,
|
| 2555 |
+
"grad_norm": 0.6080699563026428,
|
| 2556 |
+
"learning_rate": 1.3510207522324236e-07,
|
| 2557 |
+
"loss": 0.0526,
|
| 2558 |
+
"step": 1820
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 2.8876582278481013,
|
| 2562 |
+
"grad_norm": 0.5654861330986023,
|
| 2563 |
+
"learning_rate": 1.181483407286893e-07,
|
| 2564 |
+
"loss": 0.0651,
|
| 2565 |
+
"step": 1825
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 2.8955696202531644,
|
| 2569 |
+
"grad_norm": 0.5697436928749084,
|
| 2570 |
+
"learning_rate": 1.0232665867795754e-07,
|
| 2571 |
+
"loss": 0.0629,
|
| 2572 |
+
"step": 1830
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 2.903481012658228,
|
| 2576 |
+
"grad_norm": 0.7414392232894897,
|
| 2577 |
+
"learning_rate": 8.763823261575766e-08,
|
| 2578 |
+
"loss": 0.0622,
|
| 2579 |
+
"step": 1835
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 2.911392405063291,
|
| 2583 |
+
"grad_norm": 0.6146828532218933,
|
| 2584 |
+
"learning_rate": 7.40841798807762e-08,
|
| 2585 |
+
"loss": 0.0612,
|
| 2586 |
+
"step": 1840
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 2.9193037974683547,
|
| 2590 |
+
"grad_norm": 0.5820028781890869,
|
| 2591 |
+
"learning_rate": 6.166553152068877e-08,
|
| 2592 |
+
"loss": 0.0763,
|
| 2593 |
+
"step": 1845
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 2.9272151898734178,
|
| 2597 |
+
"grad_norm": 0.6345282793045044,
|
| 2598 |
+
"learning_rate": 5.0383232213722655e-08,
|
| 2599 |
+
"loss": 0.0697,
|
| 2600 |
+
"step": 1850
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 2.935126582278481,
|
| 2604 |
+
"grad_norm": 0.5352232456207275,
|
| 2605 |
+
"learning_rate": 4.023814019679106e-08,
|
| 2606 |
+
"loss": 0.0649,
|
| 2607 |
+
"step": 1855
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 2.9430379746835444,
|
| 2611 |
+
"grad_norm": 0.6531502604484558,
|
| 2612 |
+
"learning_rate": 3.1231027200221975e-08,
|
| 2613 |
+
"loss": 0.064,
|
| 2614 |
+
"step": 1860
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 2.9509493670886076,
|
| 2618 |
+
"grad_norm": 0.7060708403587341,
|
| 2619 |
+
"learning_rate": 2.3362578389041766e-08,
|
| 2620 |
+
"loss": 0.064,
|
| 2621 |
+
"step": 1865
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 2.958860759493671,
|
| 2625 |
+
"grad_norm": 0.5448899269104004,
|
| 2626 |
+
"learning_rate": 1.6633392310855255e-08,
|
| 2627 |
+
"loss": 0.0644,
|
| 2628 |
+
"step": 1870
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 2.9667721518987342,
|
| 2632 |
+
"grad_norm": 0.6815257668495178,
|
| 2633 |
+
"learning_rate": 1.1043980850323764e-08,
|
| 2634 |
+
"loss": 0.0697,
|
| 2635 |
+
"step": 1875
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 2.9746835443037973,
|
| 2639 |
+
"grad_norm": 0.5188614726066589,
|
| 2640 |
+
"learning_rate": 6.594769190212957e-09,
|
| 2641 |
+
"loss": 0.0585,
|
| 2642 |
+
"step": 1880
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 2.982594936708861,
|
| 2646 |
+
"grad_norm": 0.5929737687110901,
|
| 2647 |
+
"learning_rate": 3.2860957790636914e-09,
|
| 2648 |
+
"loss": 0.0713,
|
| 2649 |
+
"step": 1885
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 2.990506329113924,
|
| 2653 |
+
"grad_norm": 0.5531390905380249,
|
| 2654 |
+
"learning_rate": 1.1182123054342963e-09,
|
| 2655 |
+
"loss": 0.0549,
|
| 2656 |
+
"step": 1890
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 2.9984177215189876,
|
| 2660 |
+
"grad_norm": 0.4587094187736511,
|
| 2661 |
+
"learning_rate": 9.128367876420907e-11,
|
| 2662 |
+
"loss": 0.0591,
|
| 2663 |
+
"step": 1895
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 3.0,
|
| 2667 |
+
"step": 1896,
|
| 2668 |
+
"total_flos": 2.47550395020424e+18,
|
| 2669 |
+
"train_loss": 0.3481526232446359,
|
| 2670 |
+
"train_runtime": 1204.0905,
|
| 2671 |
+
"train_samples_per_second": 50.363,
|
| 2672 |
+
"train_steps_per_second": 1.575
|
| 2673 |
+
}
|
| 2674 |
+
],
|
| 2675 |
+
"logging_steps": 5,
|
| 2676 |
+
"max_steps": 1896,
|
| 2677 |
+
"num_input_tokens_seen": 0,
|
| 2678 |
+
"num_train_epochs": 3,
|
| 2679 |
+
"save_steps": 20000,
|
| 2680 |
+
"stateful_callbacks": {
|
| 2681 |
+
"TrainerControl": {
|
| 2682 |
+
"args": {
|
| 2683 |
+
"should_epoch_stop": false,
|
| 2684 |
+
"should_evaluate": false,
|
| 2685 |
+
"should_log": false,
|
| 2686 |
+
"should_save": false,
|
| 2687 |
+
"should_training_stop": false
|
| 2688 |
+
},
|
| 2689 |
+
"attributes": {}
|
| 2690 |
+
}
|
| 2691 |
+
},
|
| 2692 |
+
"total_flos": 2.47550395020424e+18,
|
| 2693 |
+
"train_batch_size": 2,
|
| 2694 |
+
"trial_name": null,
|
| 2695 |
+
"trial_params": null
|
| 2696 |
+
}
|
instruct/13_128_e3_3e-5/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:493d1079427bf68e4d878f1f9730c009da3d30cfca2648aab17c0b0322bef320
|
| 3 |
+
size 8273
|
instruct/13_128_e3_3e-5/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|