Upload folder using huggingface_hub
Browse files- instruct/103_128_e3_3e-5/adapter_config.json +39 -0
- instruct/103_128_e3_3e-5/adapter_model.safetensors +3 -0
- instruct/103_128_e3_3e-5/added_tokens.json +9 -0
- instruct/103_128_e3_3e-5/all_results.json +9 -0
- instruct/103_128_e3_3e-5/chat_template.jinja +62 -0
- instruct/103_128_e3_3e-5/config.json +32 -0
- instruct/103_128_e3_3e-5/merges.txt +0 -0
- instruct/103_128_e3_3e-5/special_tokens_map.json +33 -0
- instruct/103_128_e3_3e-5/tokenizer.json +0 -0
- instruct/103_128_e3_3e-5/tokenizer_config.json +234 -0
- instruct/103_128_e3_3e-5/train_results.json +9 -0
- instruct/103_128_e3_3e-5/trainer_state.json +2486 -0
- instruct/103_128_e3_3e-5/training_args.bin +3 -0
- instruct/103_128_e3_3e-5/vocab.json +0 -0
instruct/103_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 |
+
"gate_proj",
|
| 28 |
+
"k_proj",
|
| 29 |
+
"o_proj",
|
| 30 |
+
"q_proj",
|
| 31 |
+
"v_proj",
|
| 32 |
+
"up_proj",
|
| 33 |
+
"down_proj"
|
| 34 |
+
],
|
| 35 |
+
"task_type": "CAUSAL_LM",
|
| 36 |
+
"trainable_token_indices": null,
|
| 37 |
+
"use_dora": false,
|
| 38 |
+
"use_rslora": false
|
| 39 |
+
}
|
instruct/103_128_e3_3e-5/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ae8a26217c659e7977d3522ffdf6ce1c15e93e4e0463ec333bef9d8fe8eac19
|
| 3 |
+
size 791751704
|
instruct/103_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/103_128_e3_3e-5/all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.174167758776828e+18,
|
| 4 |
+
"train_loss": 0.3808436165718449,
|
| 5 |
+
"train_runtime": 1120.1195,
|
| 6 |
+
"train_samples": 18596,
|
| 7 |
+
"train_samples_per_second": 49.805,
|
| 8 |
+
"train_steps_per_second": 1.559
|
| 9 |
+
}
|
instruct/103_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/103_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/103_128_e3_3e-5/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
instruct/103_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/103_128_e3_3e-5/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
instruct/103_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/103_128_e3_3e-5/train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.174167758776828e+18,
|
| 4 |
+
"train_loss": 0.3808436165718449,
|
| 5 |
+
"train_runtime": 1120.1195,
|
| 6 |
+
"train_samples": 18596,
|
| 7 |
+
"train_samples_per_second": 49.805,
|
| 8 |
+
"train_steps_per_second": 1.559
|
| 9 |
+
}
|
instruct/103_128_e3_3e-5/trainer_state.json
ADDED
|
@@ -0,0 +1,2486 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1746,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008598452278589854,
|
| 14 |
+
"grad_norm": 2.7518341541290283,
|
| 15 |
+
"learning_rate": 1.3636363636363636e-06,
|
| 16 |
+
"loss": 1.71,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.017196904557179708,
|
| 21 |
+
"grad_norm": 2.1277854442596436,
|
| 22 |
+
"learning_rate": 3.0681818181818186e-06,
|
| 23 |
+
"loss": 1.5888,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.025795356835769563,
|
| 28 |
+
"grad_norm": 1.6290963888168335,
|
| 29 |
+
"learning_rate": 4.7727272727272725e-06,
|
| 30 |
+
"loss": 1.5796,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.034393809114359415,
|
| 35 |
+
"grad_norm": 0.7683576941490173,
|
| 36 |
+
"learning_rate": 6.477272727272727e-06,
|
| 37 |
+
"loss": 1.5146,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04299226139294927,
|
| 42 |
+
"grad_norm": 0.5081639289855957,
|
| 43 |
+
"learning_rate": 8.181818181818181e-06,
|
| 44 |
+
"loss": 1.521,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.051590713671539126,
|
| 49 |
+
"grad_norm": 0.47997575998306274,
|
| 50 |
+
"learning_rate": 9.886363636363637e-06,
|
| 51 |
+
"loss": 1.4461,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06018916595012898,
|
| 56 |
+
"grad_norm": 0.44006526470184326,
|
| 57 |
+
"learning_rate": 1.159090909090909e-05,
|
| 58 |
+
"loss": 1.4635,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06878761822871883,
|
| 63 |
+
"grad_norm": 0.42511072754859924,
|
| 64 |
+
"learning_rate": 1.3295454545454546e-05,
|
| 65 |
+
"loss": 1.4626,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.07738607050730868,
|
| 70 |
+
"grad_norm": 0.4504232704639435,
|
| 71 |
+
"learning_rate": 1.5e-05,
|
| 72 |
+
"loss": 1.401,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08598452278589853,
|
| 77 |
+
"grad_norm": 0.37811583280563354,
|
| 78 |
+
"learning_rate": 1.6704545454545454e-05,
|
| 79 |
+
"loss": 1.4211,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09458297506448839,
|
| 84 |
+
"grad_norm": 0.43692904710769653,
|
| 85 |
+
"learning_rate": 1.840909090909091e-05,
|
| 86 |
+
"loss": 1.3876,
|
| 87 |
+
"step": 55
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10318142734307825,
|
| 91 |
+
"grad_norm": 0.46241337060928345,
|
| 92 |
+
"learning_rate": 2.0113636363636362e-05,
|
| 93 |
+
"loss": 1.4128,
|
| 94 |
+
"step": 60
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1117798796216681,
|
| 98 |
+
"grad_norm": 0.44271963834762573,
|
| 99 |
+
"learning_rate": 2.1818181818181818e-05,
|
| 100 |
+
"loss": 1.4411,
|
| 101 |
+
"step": 65
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.12037833190025796,
|
| 105 |
+
"grad_norm": 0.4086330831050873,
|
| 106 |
+
"learning_rate": 2.3522727272727273e-05,
|
| 107 |
+
"loss": 1.4244,
|
| 108 |
+
"step": 70
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.1289767841788478,
|
| 112 |
+
"grad_norm": 0.42371106147766113,
|
| 113 |
+
"learning_rate": 2.522727272727273e-05,
|
| 114 |
+
"loss": 1.3687,
|
| 115 |
+
"step": 75
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.13757523645743766,
|
| 119 |
+
"grad_norm": 0.670913577079773,
|
| 120 |
+
"learning_rate": 2.693181818181818e-05,
|
| 121 |
+
"loss": 1.3653,
|
| 122 |
+
"step": 80
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1461736887360275,
|
| 126 |
+
"grad_norm": 0.46337324380874634,
|
| 127 |
+
"learning_rate": 2.8636363636363637e-05,
|
| 128 |
+
"loss": 1.2804,
|
| 129 |
+
"step": 85
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.15477214101461736,
|
| 133 |
+
"grad_norm": 0.49487408995628357,
|
| 134 |
+
"learning_rate": 2.9999973072760916e-05,
|
| 135 |
+
"loss": 1.3357,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.16337059329320722,
|
| 140 |
+
"grad_norm": 0.5143293738365173,
|
| 141 |
+
"learning_rate": 2.9999030629544e-05,
|
| 142 |
+
"loss": 1.3286,
|
| 143 |
+
"step": 95
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.17196904557179707,
|
| 147 |
+
"grad_norm": 0.5468387603759766,
|
| 148 |
+
"learning_rate": 2.9996741921048156e-05,
|
| 149 |
+
"loss": 1.2657,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.18056749785038692,
|
| 154 |
+
"grad_norm": 0.6305248737335205,
|
| 155 |
+
"learning_rate": 2.9993107152700567e-05,
|
| 156 |
+
"loss": 1.2565,
|
| 157 |
+
"step": 105
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.18916595012897677,
|
| 161 |
+
"grad_norm": 0.5087760090827942,
|
| 162 |
+
"learning_rate": 2.9988126650746485e-05,
|
| 163 |
+
"loss": 1.2685,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.19776440240756663,
|
| 168 |
+
"grad_norm": 0.6118876934051514,
|
| 169 |
+
"learning_rate": 2.998180086221992e-05,
|
| 170 |
+
"loss": 1.247,
|
| 171 |
+
"step": 115
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2063628546861565,
|
| 175 |
+
"grad_norm": 0.5835036635398865,
|
| 176 |
+
"learning_rate": 2.997413035490353e-05,
|
| 177 |
+
"loss": 1.217,
|
| 178 |
+
"step": 120
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.21496130696474636,
|
| 182 |
+
"grad_norm": 0.5640929937362671,
|
| 183 |
+
"learning_rate": 2.9965115817277652e-05,
|
| 184 |
+
"loss": 1.2001,
|
| 185 |
+
"step": 125
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2235597592433362,
|
| 189 |
+
"grad_norm": 0.651319682598114,
|
| 190 |
+
"learning_rate": 2.9954758058458506e-05,
|
| 191 |
+
"loss": 1.2247,
|
| 192 |
+
"step": 130
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.23215821152192606,
|
| 196 |
+
"grad_norm": 0.6857420206069946,
|
| 197 |
+
"learning_rate": 2.9943058008125583e-05,
|
| 198 |
+
"loss": 1.1513,
|
| 199 |
+
"step": 135
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2407566638005159,
|
| 203 |
+
"grad_norm": 0.6124457716941833,
|
| 204 |
+
"learning_rate": 2.993001671643818e-05,
|
| 205 |
+
"loss": 1.1313,
|
| 206 |
+
"step": 140
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.24935511607910577,
|
| 210 |
+
"grad_norm": 0.6747180819511414,
|
| 211 |
+
"learning_rate": 2.9915635353941155e-05,
|
| 212 |
+
"loss": 1.1431,
|
| 213 |
+
"step": 145
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.2579535683576956,
|
| 217 |
+
"grad_norm": 0.6426604986190796,
|
| 218 |
+
"learning_rate": 2.9899915211459864e-05,
|
| 219 |
+
"loss": 1.178,
|
| 220 |
+
"step": 150
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.26655202063628547,
|
| 224 |
+
"grad_norm": 0.6392008066177368,
|
| 225 |
+
"learning_rate": 2.988285769998429e-05,
|
| 226 |
+
"loss": 1.0984,
|
| 227 |
+
"step": 155
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2751504729148753,
|
| 231 |
+
"grad_norm": 0.6633354425430298,
|
| 232 |
+
"learning_rate": 2.98644643505424e-05,
|
| 233 |
+
"loss": 1.0759,
|
| 234 |
+
"step": 160
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2837489251934652,
|
| 238 |
+
"grad_norm": 0.7786049246788025,
|
| 239 |
+
"learning_rate": 2.984473681406274e-05,
|
| 240 |
+
"loss": 1.0722,
|
| 241 |
+
"step": 165
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.292347377472055,
|
| 245 |
+
"grad_norm": 0.7378125786781311,
|
| 246 |
+
"learning_rate": 2.9823676861226236e-05,
|
| 247 |
+
"loss": 1.0569,
|
| 248 |
+
"step": 170
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3009458297506449,
|
| 252 |
+
"grad_norm": 0.7481679916381836,
|
| 253 |
+
"learning_rate": 2.980128638230725e-05,
|
| 254 |
+
"loss": 1.009,
|
| 255 |
+
"step": 175
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.30954428202923473,
|
| 259 |
+
"grad_norm": 0.7314655780792236,
|
| 260 |
+
"learning_rate": 2.9777567387003964e-05,
|
| 261 |
+
"loss": 0.9929,
|
| 262 |
+
"step": 180
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3181427343078246,
|
| 266 |
+
"grad_norm": 0.8287091255187988,
|
| 267 |
+
"learning_rate": 2.975252200425794e-05,
|
| 268 |
+
"loss": 1.0034,
|
| 269 |
+
"step": 185
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.32674118658641443,
|
| 273 |
+
"grad_norm": 0.8267622590065002,
|
| 274 |
+
"learning_rate": 2.9726152482063068e-05,
|
| 275 |
+
"loss": 1.0172,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3353396388650043,
|
| 280 |
+
"grad_norm": 1.2825679779052734,
|
| 281 |
+
"learning_rate": 2.969846118726378e-05,
|
| 282 |
+
"loss": 0.9896,
|
| 283 |
+
"step": 195
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.34393809114359414,
|
| 287 |
+
"grad_norm": 0.9046663641929626,
|
| 288 |
+
"learning_rate": 2.9669450605342617e-05,
|
| 289 |
+
"loss": 0.9656,
|
| 290 |
+
"step": 200
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.352536543422184,
|
| 294 |
+
"grad_norm": 0.8750171065330505,
|
| 295 |
+
"learning_rate": 2.9639123340197138e-05,
|
| 296 |
+
"loss": 1.0018,
|
| 297 |
+
"step": 205
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.36113499570077384,
|
| 301 |
+
"grad_norm": 0.9116073250770569,
|
| 302 |
+
"learning_rate": 2.9607482113906177e-05,
|
| 303 |
+
"loss": 0.9118,
|
| 304 |
+
"step": 210
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3697334479793637,
|
| 308 |
+
"grad_norm": 0.931168794631958,
|
| 309 |
+
"learning_rate": 2.9574529766485578e-05,
|
| 310 |
+
"loss": 0.9558,
|
| 311 |
+
"step": 215
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.37833190025795355,
|
| 315 |
+
"grad_norm": 0.921818733215332,
|
| 316 |
+
"learning_rate": 2.9540269255633213e-05,
|
| 317 |
+
"loss": 0.9917,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3869303525365434,
|
| 322 |
+
"grad_norm": 1.0902138948440552,
|
| 323 |
+
"learning_rate": 2.950470365646357e-05,
|
| 324 |
+
"loss": 0.9289,
|
| 325 |
+
"step": 225
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.39552880481513325,
|
| 329 |
+
"grad_norm": 0.9466609358787537,
|
| 330 |
+
"learning_rate": 2.9467836161231686e-05,
|
| 331 |
+
"loss": 0.8902,
|
| 332 |
+
"step": 230
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.4041272570937231,
|
| 336 |
+
"grad_norm": 0.9133700132369995,
|
| 337 |
+
"learning_rate": 2.9429670079046675e-05,
|
| 338 |
+
"loss": 0.8809,
|
| 339 |
+
"step": 235
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.412725709372313,
|
| 343 |
+
"grad_norm": 0.939562201499939,
|
| 344 |
+
"learning_rate": 2.9390208835574662e-05,
|
| 345 |
+
"loss": 0.9363,
|
| 346 |
+
"step": 240
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.42132416165090286,
|
| 350 |
+
"grad_norm": 1.1226320266723633,
|
| 351 |
+
"learning_rate": 2.9349455972731337e-05,
|
| 352 |
+
"loss": 0.8625,
|
| 353 |
+
"step": 245
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.4299226139294927,
|
| 357 |
+
"grad_norm": 0.9787728786468506,
|
| 358 |
+
"learning_rate": 2.930741514836405e-05,
|
| 359 |
+
"loss": 0.8817,
|
| 360 |
+
"step": 250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.43852106620808257,
|
| 364 |
+
"grad_norm": 0.9790905117988586,
|
| 365 |
+
"learning_rate": 2.9264090135923458e-05,
|
| 366 |
+
"loss": 0.8634,
|
| 367 |
+
"step": 255
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.4471195184866724,
|
| 371 |
+
"grad_norm": 1.0485502481460571,
|
| 372 |
+
"learning_rate": 2.9219484824124867e-05,
|
| 373 |
+
"loss": 0.8556,
|
| 374 |
+
"step": 260
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.45571797076526227,
|
| 378 |
+
"grad_norm": 1.1596089601516724,
|
| 379 |
+
"learning_rate": 2.9173603216599182e-05,
|
| 380 |
+
"loss": 0.8417,
|
| 381 |
+
"step": 265
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.4643164230438521,
|
| 385 |
+
"grad_norm": 1.0699294805526733,
|
| 386 |
+
"learning_rate": 2.9126449431533553e-05,
|
| 387 |
+
"loss": 0.804,
|
| 388 |
+
"step": 270
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.472914875322442,
|
| 392 |
+
"grad_norm": 1.2136189937591553,
|
| 393 |
+
"learning_rate": 2.9078027701301727e-05,
|
| 394 |
+
"loss": 0.7367,
|
| 395 |
+
"step": 275
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4815133276010318,
|
| 399 |
+
"grad_norm": 1.1433069705963135,
|
| 400 |
+
"learning_rate": 2.9028342372084192e-05,
|
| 401 |
+
"loss": 0.7584,
|
| 402 |
+
"step": 280
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4901117798796217,
|
| 406 |
+
"grad_norm": 1.1357698440551758,
|
| 407 |
+
"learning_rate": 2.897739790347805e-05,
|
| 408 |
+
"loss": 0.7659,
|
| 409 |
+
"step": 285
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.49871023215821153,
|
| 413 |
+
"grad_norm": 1.2245454788208008,
|
| 414 |
+
"learning_rate": 2.8925198868096756e-05,
|
| 415 |
+
"loss": 0.7854,
|
| 416 |
+
"step": 290
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5073086844368013,
|
| 420 |
+
"grad_norm": 1.1545631885528564,
|
| 421 |
+
"learning_rate": 2.887174995115969e-05,
|
| 422 |
+
"loss": 0.7222,
|
| 423 |
+
"step": 295
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5159071367153912,
|
| 427 |
+
"grad_norm": 1.081814169883728,
|
| 428 |
+
"learning_rate": 2.8817055950071618e-05,
|
| 429 |
+
"loss": 0.7577,
|
| 430 |
+
"step": 300
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.524505588993981,
|
| 434 |
+
"grad_norm": 1.0623806715011597,
|
| 435 |
+
"learning_rate": 2.8761121773992096e-05,
|
| 436 |
+
"loss": 0.7703,
|
| 437 |
+
"step": 305
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5331040412725709,
|
| 441 |
+
"grad_norm": 1.2113078832626343,
|
| 442 |
+
"learning_rate": 2.8703952443394857e-05,
|
| 443 |
+
"loss": 0.782,
|
| 444 |
+
"step": 310
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5417024935511608,
|
| 448 |
+
"grad_norm": 1.0700323581695557,
|
| 449 |
+
"learning_rate": 2.8645553089617164e-05,
|
| 450 |
+
"loss": 0.7744,
|
| 451 |
+
"step": 315
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5503009458297506,
|
| 455 |
+
"grad_norm": 1.2286615371704102,
|
| 456 |
+
"learning_rate": 2.8585928954399242e-05,
|
| 457 |
+
"loss": 0.7513,
|
| 458 |
+
"step": 320
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5588993981083406,
|
| 462 |
+
"grad_norm": 1.182822346687317,
|
| 463 |
+
"learning_rate": 2.8525085389413817e-05,
|
| 464 |
+
"loss": 0.6453,
|
| 465 |
+
"step": 325
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5674978503869303,
|
| 469 |
+
"grad_norm": 1.2485918998718262,
|
| 470 |
+
"learning_rate": 2.8463027855785744e-05,
|
| 471 |
+
"loss": 0.6909,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5760963026655203,
|
| 476 |
+
"grad_norm": 1.1847907304763794,
|
| 477 |
+
"learning_rate": 2.839976192360184e-05,
|
| 478 |
+
"loss": 0.6993,
|
| 479 |
+
"step": 335
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.58469475494411,
|
| 483 |
+
"grad_norm": 1.3093066215515137,
|
| 484 |
+
"learning_rate": 2.8335293271410943e-05,
|
| 485 |
+
"loss": 0.6759,
|
| 486 |
+
"step": 340
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5932932072227,
|
| 490 |
+
"grad_norm": 1.2519084215164185,
|
| 491 |
+
"learning_rate": 2.8269627685714196e-05,
|
| 492 |
+
"loss": 0.6446,
|
| 493 |
+
"step": 345
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6018916595012898,
|
| 497 |
+
"grad_norm": 1.260975956916809,
|
| 498 |
+
"learning_rate": 2.8202771060445705e-05,
|
| 499 |
+
"loss": 0.6935,
|
| 500 |
+
"step": 350
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6104901117798797,
|
| 504 |
+
"grad_norm": 1.1231117248535156,
|
| 505 |
+
"learning_rate": 2.8134729396443477e-05,
|
| 506 |
+
"loss": 0.7034,
|
| 507 |
+
"step": 355
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6190885640584695,
|
| 511 |
+
"grad_norm": 1.095755934715271,
|
| 512 |
+
"learning_rate": 2.806550880091085e-05,
|
| 513 |
+
"loss": 0.637,
|
| 514 |
+
"step": 360
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6276870163370594,
|
| 518 |
+
"grad_norm": 1.4837367534637451,
|
| 519 |
+
"learning_rate": 2.7995115486868297e-05,
|
| 520 |
+
"loss": 0.6622,
|
| 521 |
+
"step": 365
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.6362854686156492,
|
| 525 |
+
"grad_norm": 1.0324267148971558,
|
| 526 |
+
"learning_rate": 2.792355577259576e-05,
|
| 527 |
+
"loss": 0.7279,
|
| 528 |
+
"step": 370
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6448839208942391,
|
| 532 |
+
"grad_norm": 1.3489702939987183,
|
| 533 |
+
"learning_rate": 2.7850836081065586e-05,
|
| 534 |
+
"loss": 0.6443,
|
| 535 |
+
"step": 375
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6534823731728289,
|
| 539 |
+
"grad_norm": 1.3159942626953125,
|
| 540 |
+
"learning_rate": 2.7776962939365957e-05,
|
| 541 |
+
"loss": 0.6529,
|
| 542 |
+
"step": 380
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6620808254514188,
|
| 546 |
+
"grad_norm": 1.210880160331726,
|
| 547 |
+
"learning_rate": 2.7701942978115092e-05,
|
| 548 |
+
"loss": 0.6855,
|
| 549 |
+
"step": 385
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6706792777300086,
|
| 553 |
+
"grad_norm": 1.093973994255066,
|
| 554 |
+
"learning_rate": 2.7625782930866088e-05,
|
| 555 |
+
"loss": 0.6599,
|
| 556 |
+
"step": 390
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6792777300085985,
|
| 560 |
+
"grad_norm": 1.1897146701812744,
|
| 561 |
+
"learning_rate": 2.754848963350253e-05,
|
| 562 |
+
"loss": 0.5954,
|
| 563 |
+
"step": 395
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6878761822871883,
|
| 567 |
+
"grad_norm": 1.4455071687698364,
|
| 568 |
+
"learning_rate": 2.7470070023624922e-05,
|
| 569 |
+
"loss": 0.5969,
|
| 570 |
+
"step": 400
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.6964746345657782,
|
| 574 |
+
"grad_norm": 1.1741461753845215,
|
| 575 |
+
"learning_rate": 2.7390531139928003e-05,
|
| 576 |
+
"loss": 0.5734,
|
| 577 |
+
"step": 405
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.705073086844368,
|
| 581 |
+
"grad_norm": 1.3174519538879395,
|
| 582 |
+
"learning_rate": 2.7309880121568967e-05,
|
| 583 |
+
"loss": 0.6007,
|
| 584 |
+
"step": 410
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7136715391229579,
|
| 588 |
+
"grad_norm": 1.2455363273620605,
|
| 589 |
+
"learning_rate": 2.7228124207526677e-05,
|
| 590 |
+
"loss": 0.5871,
|
| 591 |
+
"step": 415
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7222699914015477,
|
| 595 |
+
"grad_norm": 1.2164477109909058,
|
| 596 |
+
"learning_rate": 2.7145270735951932e-05,
|
| 597 |
+
"loss": 0.5815,
|
| 598 |
+
"step": 420
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7308684436801376,
|
| 602 |
+
"grad_norm": 1.1618539094924927,
|
| 603 |
+
"learning_rate": 2.706132714350878e-05,
|
| 604 |
+
"loss": 0.5607,
|
| 605 |
+
"step": 425
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.7394668959587274,
|
| 609 |
+
"grad_norm": 1.2314118146896362,
|
| 610 |
+
"learning_rate": 2.6976300964707072e-05,
|
| 611 |
+
"loss": 0.5508,
|
| 612 |
+
"step": 430
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7480653482373173,
|
| 616 |
+
"grad_norm": 1.1935690641403198,
|
| 617 |
+
"learning_rate": 2.689019983122615e-05,
|
| 618 |
+
"loss": 0.586,
|
| 619 |
+
"step": 435
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7566638005159071,
|
| 623 |
+
"grad_norm": 1.3903708457946777,
|
| 624 |
+
"learning_rate": 2.6803031471229876e-05,
|
| 625 |
+
"loss": 0.5908,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.765262252794497,
|
| 630 |
+
"grad_norm": 1.2954262495040894,
|
| 631 |
+
"learning_rate": 2.6714803708672977e-05,
|
| 632 |
+
"loss": 0.535,
|
| 633 |
+
"step": 445
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7738607050730868,
|
| 637 |
+
"grad_norm": 1.610106110572815,
|
| 638 |
+
"learning_rate": 2.6625524462598777e-05,
|
| 639 |
+
"loss": 0.5821,
|
| 640 |
+
"step": 450
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7824591573516767,
|
| 644 |
+
"grad_norm": 1.1699998378753662,
|
| 645 |
+
"learning_rate": 2.6535201746428407e-05,
|
| 646 |
+
"loss": 0.5448,
|
| 647 |
+
"step": 455
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.7910576096302665,
|
| 651 |
+
"grad_norm": 1.149044156074524,
|
| 652 |
+
"learning_rate": 2.6443843667241568e-05,
|
| 653 |
+
"loss": 0.4974,
|
| 654 |
+
"step": 460
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7996560619088564,
|
| 658 |
+
"grad_norm": 1.1841697692871094,
|
| 659 |
+
"learning_rate": 2.6351458425048843e-05,
|
| 660 |
+
"loss": 0.502,
|
| 661 |
+
"step": 465
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.8082545141874462,
|
| 665 |
+
"grad_norm": 1.2326538562774658,
|
| 666 |
+
"learning_rate": 2.6258054312055712e-05,
|
| 667 |
+
"loss": 0.5012,
|
| 668 |
+
"step": 470
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.8168529664660361,
|
| 672 |
+
"grad_norm": 1.2268785238265991,
|
| 673 |
+
"learning_rate": 2.616363971191824e-05,
|
| 674 |
+
"loss": 0.5134,
|
| 675 |
+
"step": 475
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.825451418744626,
|
| 679 |
+
"grad_norm": 1.2168312072753906,
|
| 680 |
+
"learning_rate": 2.6068223098990618e-05,
|
| 681 |
+
"loss": 0.4778,
|
| 682 |
+
"step": 480
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8340498710232158,
|
| 686 |
+
"grad_norm": 1.2268548011779785,
|
| 687 |
+
"learning_rate": 2.5971813037564512e-05,
|
| 688 |
+
"loss": 0.5018,
|
| 689 |
+
"step": 485
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8426483233018057,
|
| 693 |
+
"grad_norm": 1.2626659870147705,
|
| 694 |
+
"learning_rate": 2.5874418181100357e-05,
|
| 695 |
+
"loss": 0.4962,
|
| 696 |
+
"step": 490
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.8512467755803955,
|
| 700 |
+
"grad_norm": 1.2760711908340454,
|
| 701 |
+
"learning_rate": 2.5776047271450672e-05,
|
| 702 |
+
"loss": 0.4834,
|
| 703 |
+
"step": 495
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.8598452278589854,
|
| 707 |
+
"grad_norm": 1.2341679334640503,
|
| 708 |
+
"learning_rate": 2.5676709138075405e-05,
|
| 709 |
+
"loss": 0.4653,
|
| 710 |
+
"step": 500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8684436801375752,
|
| 714 |
+
"grad_norm": 1.5237632989883423,
|
| 715 |
+
"learning_rate": 2.5576412697249415e-05,
|
| 716 |
+
"loss": 0.4807,
|
| 717 |
+
"step": 505
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8770421324161651,
|
| 721 |
+
"grad_norm": 1.4803695678710938,
|
| 722 |
+
"learning_rate": 2.54751669512622e-05,
|
| 723 |
+
"loss": 0.4631,
|
| 724 |
+
"step": 510
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.8856405846947549,
|
| 728 |
+
"grad_norm": 1.3156218528747559,
|
| 729 |
+
"learning_rate": 2.5372980987609867e-05,
|
| 730 |
+
"loss": 0.4563,
|
| 731 |
+
"step": 515
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8942390369733448,
|
| 735 |
+
"grad_norm": 1.5108305215835571,
|
| 736 |
+
"learning_rate": 2.5269863978179474e-05,
|
| 737 |
+
"loss": 0.4301,
|
| 738 |
+
"step": 520
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.9028374892519346,
|
| 742 |
+
"grad_norm": 1.2623145580291748,
|
| 743 |
+
"learning_rate": 2.516582517842578e-05,
|
| 744 |
+
"loss": 0.4293,
|
| 745 |
+
"step": 525
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.9114359415305245,
|
| 749 |
+
"grad_norm": 1.3140374422073364,
|
| 750 |
+
"learning_rate": 2.5060873926540513e-05,
|
| 751 |
+
"loss": 0.4462,
|
| 752 |
+
"step": 530
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.9200343938091143,
|
| 756 |
+
"grad_norm": 1.2793163061141968,
|
| 757 |
+
"learning_rate": 2.4955019642614206e-05,
|
| 758 |
+
"loss": 0.4143,
|
| 759 |
+
"step": 535
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.9286328460877042,
|
| 763 |
+
"grad_norm": 1.4150524139404297,
|
| 764 |
+
"learning_rate": 2.484827182779067e-05,
|
| 765 |
+
"loss": 0.451,
|
| 766 |
+
"step": 540
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.937231298366294,
|
| 770 |
+
"grad_norm": 1.2748008966445923,
|
| 771 |
+
"learning_rate": 2.474064006341421e-05,
|
| 772 |
+
"loss": 0.4571,
|
| 773 |
+
"step": 545
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.945829750644884,
|
| 777 |
+
"grad_norm": 1.36494779586792,
|
| 778 |
+
"learning_rate": 2.4632134010169634e-05,
|
| 779 |
+
"loss": 0.4507,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9544282029234737,
|
| 784 |
+
"grad_norm": 1.3021773099899292,
|
| 785 |
+
"learning_rate": 2.4522763407215135e-05,
|
| 786 |
+
"loss": 0.4166,
|
| 787 |
+
"step": 555
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.9630266552020637,
|
| 791 |
+
"grad_norm": 1.2684458494186401,
|
| 792 |
+
"learning_rate": 2.4412538071308134e-05,
|
| 793 |
+
"loss": 0.4124,
|
| 794 |
+
"step": 560
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.9716251074806534,
|
| 798 |
+
"grad_norm": 1.2295825481414795,
|
| 799 |
+
"learning_rate": 2.430146789592416e-05,
|
| 800 |
+
"loss": 0.4289,
|
| 801 |
+
"step": 565
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.9802235597592434,
|
| 805 |
+
"grad_norm": 1.3683432340621948,
|
| 806 |
+
"learning_rate": 2.4189562850368867e-05,
|
| 807 |
+
"loss": 0.4327,
|
| 808 |
+
"step": 570
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.9888220120378332,
|
| 812 |
+
"grad_norm": 1.1714926958084106,
|
| 813 |
+
"learning_rate": 2.4076832978883177e-05,
|
| 814 |
+
"loss": 0.4196,
|
| 815 |
+
"step": 575
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9974204643164231,
|
| 819 |
+
"grad_norm": 1.3588829040527344,
|
| 820 |
+
"learning_rate": 2.3963288399741775e-05,
|
| 821 |
+
"loss": 0.4008,
|
| 822 |
+
"step": 580
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.0051590713671539,
|
| 826 |
+
"grad_norm": 1.3453835248947144,
|
| 827 |
+
"learning_rate": 2.384893930434491e-05,
|
| 828 |
+
"loss": 0.4003,
|
| 829 |
+
"step": 585
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.0137575236457437,
|
| 833 |
+
"grad_norm": 1.4790918827056885,
|
| 834 |
+
"learning_rate": 2.3733795956303656e-05,
|
| 835 |
+
"loss": 0.3063,
|
| 836 |
+
"step": 590
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.0223559759243337,
|
| 840 |
+
"grad_norm": 1.133102536201477,
|
| 841 |
+
"learning_rate": 2.3617868690518666e-05,
|
| 842 |
+
"loss": 0.3172,
|
| 843 |
+
"step": 595
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.0309544282029235,
|
| 847 |
+
"grad_norm": 1.2257696390151978,
|
| 848 |
+
"learning_rate": 2.3501167912252567e-05,
|
| 849 |
+
"loss": 0.3472,
|
| 850 |
+
"step": 600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.0395528804815133,
|
| 854 |
+
"grad_norm": 1.143408179283142,
|
| 855 |
+
"learning_rate": 2.338370409619601e-05,
|
| 856 |
+
"loss": 0.3179,
|
| 857 |
+
"step": 605
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.048151332760103,
|
| 861 |
+
"grad_norm": 1.2887088060379028,
|
| 862 |
+
"learning_rate": 2.326548778552748e-05,
|
| 863 |
+
"loss": 0.3429,
|
| 864 |
+
"step": 610
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.056749785038693,
|
| 868 |
+
"grad_norm": 1.1695152521133423,
|
| 869 |
+
"learning_rate": 2.314652959096698e-05,
|
| 870 |
+
"loss": 0.3234,
|
| 871 |
+
"step": 615
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.065348237317283,
|
| 875 |
+
"grad_norm": 1.2436615228652954,
|
| 876 |
+
"learning_rate": 2.3026840189823672e-05,
|
| 877 |
+
"loss": 0.3429,
|
| 878 |
+
"step": 620
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.0739466895958727,
|
| 882 |
+
"grad_norm": 1.2530831098556519,
|
| 883 |
+
"learning_rate": 2.290643032503748e-05,
|
| 884 |
+
"loss": 0.3418,
|
| 885 |
+
"step": 625
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.0825451418744625,
|
| 889 |
+
"grad_norm": 1.6981089115142822,
|
| 890 |
+
"learning_rate": 2.278531080421485e-05,
|
| 891 |
+
"loss": 0.3197,
|
| 892 |
+
"step": 630
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.0911435941530525,
|
| 896 |
+
"grad_norm": 1.3294607400894165,
|
| 897 |
+
"learning_rate": 2.2663492498658707e-05,
|
| 898 |
+
"loss": 0.299,
|
| 899 |
+
"step": 635
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.0997420464316423,
|
| 903 |
+
"grad_norm": 1.186793327331543,
|
| 904 |
+
"learning_rate": 2.2540986342392667e-05,
|
| 905 |
+
"loss": 0.3074,
|
| 906 |
+
"step": 640
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.108340498710232,
|
| 910 |
+
"grad_norm": 1.3619786500930786,
|
| 911 |
+
"learning_rate": 2.2417803331179636e-05,
|
| 912 |
+
"loss": 0.339,
|
| 913 |
+
"step": 645
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.1169389509888221,
|
| 917 |
+
"grad_norm": 1.7010291814804077,
|
| 918 |
+
"learning_rate": 2.2293954521534867e-05,
|
| 919 |
+
"loss": 0.3236,
|
| 920 |
+
"step": 650
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.125537403267412,
|
| 924 |
+
"grad_norm": 1.4136074781417847,
|
| 925 |
+
"learning_rate": 2.2169451029733565e-05,
|
| 926 |
+
"loss": 0.3181,
|
| 927 |
+
"step": 655
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.1341358555460017,
|
| 931 |
+
"grad_norm": 1.1302882432937622,
|
| 932 |
+
"learning_rate": 2.204430403081312e-05,
|
| 933 |
+
"loss": 0.2954,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.1427343078245915,
|
| 938 |
+
"grad_norm": 1.261544108390808,
|
| 939 |
+
"learning_rate": 2.1918524757570072e-05,
|
| 940 |
+
"loss": 0.2975,
|
| 941 |
+
"step": 665
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.1513327601031813,
|
| 945 |
+
"grad_norm": 1.2275208234786987,
|
| 946 |
+
"learning_rate": 2.1792124499551904e-05,
|
| 947 |
+
"loss": 0.3235,
|
| 948 |
+
"step": 670
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.1599312123817713,
|
| 952 |
+
"grad_norm": 1.208698034286499,
|
| 953 |
+
"learning_rate": 2.1665114602043706e-05,
|
| 954 |
+
"loss": 0.2661,
|
| 955 |
+
"step": 675
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.1685296646603611,
|
| 959 |
+
"grad_norm": 0.990852415561676,
|
| 960 |
+
"learning_rate": 2.153750646504987e-05,
|
| 961 |
+
"loss": 0.2775,
|
| 962 |
+
"step": 680
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.177128116938951,
|
| 966 |
+
"grad_norm": 1.2367091178894043,
|
| 967 |
+
"learning_rate": 2.140931154227086e-05,
|
| 968 |
+
"loss": 0.2708,
|
| 969 |
+
"step": 685
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.185726569217541,
|
| 973 |
+
"grad_norm": 1.1172142028808594,
|
| 974 |
+
"learning_rate": 2.128054134007519e-05,
|
| 975 |
+
"loss": 0.2925,
|
| 976 |
+
"step": 690
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.1943250214961307,
|
| 980 |
+
"grad_norm": 1.0324984788894653,
|
| 981 |
+
"learning_rate": 2.115120741646658e-05,
|
| 982 |
+
"loss": 0.292,
|
| 983 |
+
"step": 695
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.2029234737747205,
|
| 987 |
+
"grad_norm": 1.201161503791809,
|
| 988 |
+
"learning_rate": 2.1021321380046633e-05,
|
| 989 |
+
"loss": 0.3108,
|
| 990 |
+
"step": 700
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.2115219260533103,
|
| 994 |
+
"grad_norm": 1.4454787969589233,
|
| 995 |
+
"learning_rate": 2.0890894888972815e-05,
|
| 996 |
+
"loss": 0.2734,
|
| 997 |
+
"step": 705
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.2201203783319003,
|
| 1001 |
+
"grad_norm": 1.240338921546936,
|
| 1002 |
+
"learning_rate": 2.0759939649912088e-05,
|
| 1003 |
+
"loss": 0.248,
|
| 1004 |
+
"step": 710
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.2287188306104901,
|
| 1008 |
+
"grad_norm": 1.3253576755523682,
|
| 1009 |
+
"learning_rate": 2.0628467416990167e-05,
|
| 1010 |
+
"loss": 0.2461,
|
| 1011 |
+
"step": 715
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.23731728288908,
|
| 1015 |
+
"grad_norm": 1.35067880153656,
|
| 1016 |
+
"learning_rate": 2.0496489990736467e-05,
|
| 1017 |
+
"loss": 0.2646,
|
| 1018 |
+
"step": 720
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.2459157351676697,
|
| 1022 |
+
"grad_norm": 1.3040696382522583,
|
| 1023 |
+
"learning_rate": 2.0364019217024975e-05,
|
| 1024 |
+
"loss": 0.2639,
|
| 1025 |
+
"step": 725
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.2545141874462598,
|
| 1029 |
+
"grad_norm": 1.2524213790893555,
|
| 1030 |
+
"learning_rate": 2.023106698601095e-05,
|
| 1031 |
+
"loss": 0.2759,
|
| 1032 |
+
"step": 730
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.2631126397248496,
|
| 1036 |
+
"grad_norm": 1.479215145111084,
|
| 1037 |
+
"learning_rate": 2.009764523106375e-05,
|
| 1038 |
+
"loss": 0.2296,
|
| 1039 |
+
"step": 735
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.2717110920034393,
|
| 1043 |
+
"grad_norm": 1.1554856300354004,
|
| 1044 |
+
"learning_rate": 1.996376592769569e-05,
|
| 1045 |
+
"loss": 0.2655,
|
| 1046 |
+
"step": 740
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.2803095442820291,
|
| 1050 |
+
"grad_norm": 1.4139991998672485,
|
| 1051 |
+
"learning_rate": 1.9829441092487195e-05,
|
| 1052 |
+
"loss": 0.2822,
|
| 1053 |
+
"step": 745
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.2889079965606192,
|
| 1057 |
+
"grad_norm": 1.2377347946166992,
|
| 1058 |
+
"learning_rate": 1.969468278200819e-05,
|
| 1059 |
+
"loss": 0.2257,
|
| 1060 |
+
"step": 750
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.297506448839209,
|
| 1064 |
+
"grad_norm": 1.2477161884307861,
|
| 1065 |
+
"learning_rate": 1.9559503091735975e-05,
|
| 1066 |
+
"loss": 0.2532,
|
| 1067 |
+
"step": 755
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.3061049011177988,
|
| 1071 |
+
"grad_norm": 1.3383243083953857,
|
| 1072 |
+
"learning_rate": 1.9423914154969557e-05,
|
| 1073 |
+
"loss": 0.2563,
|
| 1074 |
+
"step": 760
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.3147033533963888,
|
| 1078 |
+
"grad_norm": 1.4661064147949219,
|
| 1079 |
+
"learning_rate": 1.9287928141740605e-05,
|
| 1080 |
+
"loss": 0.2269,
|
| 1081 |
+
"step": 765
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.3233018056749786,
|
| 1085 |
+
"grad_norm": 1.173181414604187,
|
| 1086 |
+
"learning_rate": 1.9151557257721108e-05,
|
| 1087 |
+
"loss": 0.2124,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.3319002579535684,
|
| 1092 |
+
"grad_norm": 1.1441689729690552,
|
| 1093 |
+
"learning_rate": 1.9014813743127834e-05,
|
| 1094 |
+
"loss": 0.2525,
|
| 1095 |
+
"step": 775
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.3404987102321582,
|
| 1099 |
+
"grad_norm": 1.2079240083694458,
|
| 1100 |
+
"learning_rate": 1.887770987162369e-05,
|
| 1101 |
+
"loss": 0.2161,
|
| 1102 |
+
"step": 780
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.349097162510748,
|
| 1106 |
+
"grad_norm": 1.2237778902053833,
|
| 1107 |
+
"learning_rate": 1.8740257949216058e-05,
|
| 1108 |
+
"loss": 0.2282,
|
| 1109 |
+
"step": 785
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.357695614789338,
|
| 1113 |
+
"grad_norm": 1.1923264265060425,
|
| 1114 |
+
"learning_rate": 1.8602470313152287e-05,
|
| 1115 |
+
"loss": 0.2398,
|
| 1116 |
+
"step": 790
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.3662940670679278,
|
| 1120 |
+
"grad_norm": 1.518862009048462,
|
| 1121 |
+
"learning_rate": 1.846435933081229e-05,
|
| 1122 |
+
"loss": 0.2535,
|
| 1123 |
+
"step": 795
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.3748925193465176,
|
| 1127 |
+
"grad_norm": 1.3634533882141113,
|
| 1128 |
+
"learning_rate": 1.8325937398598534e-05,
|
| 1129 |
+
"loss": 0.25,
|
| 1130 |
+
"step": 800
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.3834909716251076,
|
| 1134 |
+
"grad_norm": 1.2422845363616943,
|
| 1135 |
+
"learning_rate": 1.8187216940823338e-05,
|
| 1136 |
+
"loss": 0.2329,
|
| 1137 |
+
"step": 805
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.3920894239036974,
|
| 1141 |
+
"grad_norm": 1.1868435144424438,
|
| 1142 |
+
"learning_rate": 1.8048210408593754e-05,
|
| 1143 |
+
"loss": 0.2149,
|
| 1144 |
+
"step": 810
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.4006878761822872,
|
| 1148 |
+
"grad_norm": 1.0920038223266602,
|
| 1149 |
+
"learning_rate": 1.7908930278693927e-05,
|
| 1150 |
+
"loss": 0.2189,
|
| 1151 |
+
"step": 815
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.409286328460877,
|
| 1155 |
+
"grad_norm": 1.2655036449432373,
|
| 1156 |
+
"learning_rate": 1.7769389052465298e-05,
|
| 1157 |
+
"loss": 0.2274,
|
| 1158 |
+
"step": 820
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.4178847807394668,
|
| 1162 |
+
"grad_norm": 1.3738645315170288,
|
| 1163 |
+
"learning_rate": 1.7629599254684463e-05,
|
| 1164 |
+
"loss": 0.2092,
|
| 1165 |
+
"step": 825
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.4264832330180568,
|
| 1169 |
+
"grad_norm": 1.1948843002319336,
|
| 1170 |
+
"learning_rate": 1.7489573432439006e-05,
|
| 1171 |
+
"loss": 0.2179,
|
| 1172 |
+
"step": 830
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.4350816852966466,
|
| 1176 |
+
"grad_norm": 1.2676000595092773,
|
| 1177 |
+
"learning_rate": 1.7349324154001336e-05,
|
| 1178 |
+
"loss": 0.2104,
|
| 1179 |
+
"step": 835
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.4436801375752364,
|
| 1183 |
+
"grad_norm": 1.3298603296279907,
|
| 1184 |
+
"learning_rate": 1.7208864007700555e-05,
|
| 1185 |
+
"loss": 0.219,
|
| 1186 |
+
"step": 840
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.4522785898538264,
|
| 1190 |
+
"grad_norm": 1.1727118492126465,
|
| 1191 |
+
"learning_rate": 1.706820560079262e-05,
|
| 1192 |
+
"loss": 0.2208,
|
| 1193 |
+
"step": 845
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.4608770421324162,
|
| 1197 |
+
"grad_norm": 1.1638908386230469,
|
| 1198 |
+
"learning_rate": 1.6927361558328703e-05,
|
| 1199 |
+
"loss": 0.2081,
|
| 1200 |
+
"step": 850
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.469475494411006,
|
| 1204 |
+
"grad_norm": 1.290232539176941,
|
| 1205 |
+
"learning_rate": 1.678634452202205e-05,
|
| 1206 |
+
"loss": 0.1884,
|
| 1207 |
+
"step": 855
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.4780739466895958,
|
| 1211 |
+
"grad_norm": 1.4613267183303833,
|
| 1212 |
+
"learning_rate": 1.6645167149113277e-05,
|
| 1213 |
+
"loss": 0.2042,
|
| 1214 |
+
"step": 860
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.4866723989681856,
|
| 1218 |
+
"grad_norm": 1.1067838668823242,
|
| 1219 |
+
"learning_rate": 1.6503842111234306e-05,
|
| 1220 |
+
"loss": 0.1894,
|
| 1221 |
+
"step": 865
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.4952708512467756,
|
| 1225 |
+
"grad_norm": 1.3368257284164429,
|
| 1226 |
+
"learning_rate": 1.6362382093271004e-05,
|
| 1227 |
+
"loss": 0.2104,
|
| 1228 |
+
"step": 870
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.5038693035253654,
|
| 1232 |
+
"grad_norm": 1.2354159355163574,
|
| 1233 |
+
"learning_rate": 1.62207997922246e-05,
|
| 1234 |
+
"loss": 0.2157,
|
| 1235 |
+
"step": 875
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.5124677558039554,
|
| 1239 |
+
"grad_norm": 1.1606972217559814,
|
| 1240 |
+
"learning_rate": 1.607910791607209e-05,
|
| 1241 |
+
"loss": 0.1938,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.5210662080825452,
|
| 1246 |
+
"grad_norm": 1.2570160627365112,
|
| 1247 |
+
"learning_rate": 1.5937319182625558e-05,
|
| 1248 |
+
"loss": 0.217,
|
| 1249 |
+
"step": 885
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.529664660361135,
|
| 1253 |
+
"grad_norm": 1.2953159809112549,
|
| 1254 |
+
"learning_rate": 1.579544631839071e-05,
|
| 1255 |
+
"loss": 0.1804,
|
| 1256 |
+
"step": 890
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.5382631126397248,
|
| 1260 |
+
"grad_norm": 1.0525399446487427,
|
| 1261 |
+
"learning_rate": 1.5653502057424562e-05,
|
| 1262 |
+
"loss": 0.182,
|
| 1263 |
+
"step": 895
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.5468615649183146,
|
| 1267 |
+
"grad_norm": 1.0252060890197754,
|
| 1268 |
+
"learning_rate": 1.5511499140192463e-05,
|
| 1269 |
+
"loss": 0.1753,
|
| 1270 |
+
"step": 900
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.5554600171969044,
|
| 1274 |
+
"grad_norm": 1.3365756273269653,
|
| 1275 |
+
"learning_rate": 1.536945031242456e-05,
|
| 1276 |
+
"loss": 0.1738,
|
| 1277 |
+
"step": 905
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.5640584694754944,
|
| 1281 |
+
"grad_norm": 1.293882966041565,
|
| 1282 |
+
"learning_rate": 1.5227368323971801e-05,
|
| 1283 |
+
"loss": 0.1783,
|
| 1284 |
+
"step": 910
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.5726569217540842,
|
| 1288 |
+
"grad_norm": 1.1192034482955933,
|
| 1289 |
+
"learning_rate": 1.5085265927661519e-05,
|
| 1290 |
+
"loss": 0.2006,
|
| 1291 |
+
"step": 915
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.5812553740326742,
|
| 1295 |
+
"grad_norm": 1.4057538509368896,
|
| 1296 |
+
"learning_rate": 1.4943155878152792e-05,
|
| 1297 |
+
"loss": 0.1866,
|
| 1298 |
+
"step": 920
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.589853826311264,
|
| 1302 |
+
"grad_norm": 1.2273590564727783,
|
| 1303 |
+
"learning_rate": 1.4801050930791624e-05,
|
| 1304 |
+
"loss": 0.1929,
|
| 1305 |
+
"step": 925
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.5984522785898538,
|
| 1309 |
+
"grad_norm": 1.1663767099380493,
|
| 1310 |
+
"learning_rate": 1.465896384046607e-05,
|
| 1311 |
+
"loss": 0.1639,
|
| 1312 |
+
"step": 930
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.6070507308684436,
|
| 1316 |
+
"grad_norm": 1.2579584121704102,
|
| 1317 |
+
"learning_rate": 1.4516907360461395e-05,
|
| 1318 |
+
"loss": 0.1771,
|
| 1319 |
+
"step": 935
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.6156491831470334,
|
| 1323 |
+
"grad_norm": 1.1554639339447021,
|
| 1324 |
+
"learning_rate": 1.4374894241315383e-05,
|
| 1325 |
+
"loss": 0.1465,
|
| 1326 |
+
"step": 940
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.6242476354256232,
|
| 1330 |
+
"grad_norm": 1.2014966011047363,
|
| 1331 |
+
"learning_rate": 1.423293722967387e-05,
|
| 1332 |
+
"loss": 0.1491,
|
| 1333 |
+
"step": 945
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.6328460877042132,
|
| 1337 |
+
"grad_norm": 1.0337907075881958,
|
| 1338 |
+
"learning_rate": 1.4091049067146674e-05,
|
| 1339 |
+
"loss": 0.1764,
|
| 1340 |
+
"step": 950
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.641444539982803,
|
| 1344 |
+
"grad_norm": 1.244678258895874,
|
| 1345 |
+
"learning_rate": 1.3949242489163928e-05,
|
| 1346 |
+
"loss": 0.1521,
|
| 1347 |
+
"step": 955
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.650042992261393,
|
| 1351 |
+
"grad_norm": 1.2274010181427002,
|
| 1352 |
+
"learning_rate": 1.380753022383301e-05,
|
| 1353 |
+
"loss": 0.1751,
|
| 1354 |
+
"step": 960
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.6586414445399829,
|
| 1358 |
+
"grad_norm": 1.1629021167755127,
|
| 1359 |
+
"learning_rate": 1.3665924990796073e-05,
|
| 1360 |
+
"loss": 0.1948,
|
| 1361 |
+
"step": 965
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.6672398968185727,
|
| 1365 |
+
"grad_norm": 1.03988778591156,
|
| 1366 |
+
"learning_rate": 1.3524439500088403e-05,
|
| 1367 |
+
"loss": 0.1608,
|
| 1368 |
+
"step": 970
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.6758383490971624,
|
| 1372 |
+
"grad_norm": 1.1245485544204712,
|
| 1373 |
+
"learning_rate": 1.3383086450997608e-05,
|
| 1374 |
+
"loss": 0.1729,
|
| 1375 |
+
"step": 975
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.6844368013757522,
|
| 1379 |
+
"grad_norm": 1.1119444370269775,
|
| 1380 |
+
"learning_rate": 1.324187853092373e-05,
|
| 1381 |
+
"loss": 0.1627,
|
| 1382 |
+
"step": 980
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.6930352536543423,
|
| 1386 |
+
"grad_norm": 1.0288985967636108,
|
| 1387 |
+
"learning_rate": 1.310082841424051e-05,
|
| 1388 |
+
"loss": 0.1524,
|
| 1389 |
+
"step": 985
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.701633705932932,
|
| 1393 |
+
"grad_norm": 1.2200804948806763,
|
| 1394 |
+
"learning_rate": 1.2959948761157763e-05,
|
| 1395 |
+
"loss": 0.1702,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.710232158211522,
|
| 1400 |
+
"grad_norm": 1.0861482620239258,
|
| 1401 |
+
"learning_rate": 1.2819252216585023e-05,
|
| 1402 |
+
"loss": 0.158,
|
| 1403 |
+
"step": 995
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.7188306104901119,
|
| 1407 |
+
"grad_norm": 1.1331754922866821,
|
| 1408 |
+
"learning_rate": 1.2678751408996586e-05,
|
| 1409 |
+
"loss": 0.1571,
|
| 1410 |
+
"step": 1000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.7274290627687017,
|
| 1414 |
+
"grad_norm": 1.0072189569473267,
|
| 1415 |
+
"learning_rate": 1.253845894929802e-05,
|
| 1416 |
+
"loss": 0.1531,
|
| 1417 |
+
"step": 1005
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.7360275150472915,
|
| 1421 |
+
"grad_norm": 1.072929859161377,
|
| 1422 |
+
"learning_rate": 1.239838742969426e-05,
|
| 1423 |
+
"loss": 0.1473,
|
| 1424 |
+
"step": 1010
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.7446259673258813,
|
| 1428 |
+
"grad_norm": 1.1905951499938965,
|
| 1429 |
+
"learning_rate": 1.2258549422559352e-05,
|
| 1430 |
+
"loss": 0.1513,
|
| 1431 |
+
"step": 1015
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.753224419604471,
|
| 1435 |
+
"grad_norm": 1.090506672859192,
|
| 1436 |
+
"learning_rate": 1.2118957479307997e-05,
|
| 1437 |
+
"loss": 0.1428,
|
| 1438 |
+
"step": 1020
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.761822871883061,
|
| 1442 |
+
"grad_norm": 1.002059817314148,
|
| 1443 |
+
"learning_rate": 1.1979624129269002e-05,
|
| 1444 |
+
"loss": 0.1398,
|
| 1445 |
+
"step": 1025
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.7704213241616509,
|
| 1449 |
+
"grad_norm": 0.9589114189147949,
|
| 1450 |
+
"learning_rate": 1.184056187856066e-05,
|
| 1451 |
+
"loss": 0.1547,
|
| 1452 |
+
"step": 1030
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.779019776440241,
|
| 1456 |
+
"grad_norm": 1.2421773672103882,
|
| 1457 |
+
"learning_rate": 1.1701783208968248e-05,
|
| 1458 |
+
"loss": 0.1488,
|
| 1459 |
+
"step": 1035
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.7876182287188307,
|
| 1463 |
+
"grad_norm": 0.9530978202819824,
|
| 1464 |
+
"learning_rate": 1.1563300576823713e-05,
|
| 1465 |
+
"loss": 0.1424,
|
| 1466 |
+
"step": 1040
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.7962166809974205,
|
| 1470 |
+
"grad_norm": 1.0903871059417725,
|
| 1471 |
+
"learning_rate": 1.1425126411887614e-05,
|
| 1472 |
+
"loss": 0.1336,
|
| 1473 |
+
"step": 1045
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.8048151332760103,
|
| 1477 |
+
"grad_norm": 0.9657735824584961,
|
| 1478 |
+
"learning_rate": 1.1287273116233482e-05,
|
| 1479 |
+
"loss": 0.1355,
|
| 1480 |
+
"step": 1050
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.8134135855546,
|
| 1484 |
+
"grad_norm": 1.1656955480575562,
|
| 1485 |
+
"learning_rate": 1.1149753063134628e-05,
|
| 1486 |
+
"loss": 0.1382,
|
| 1487 |
+
"step": 1055
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.8220120378331899,
|
| 1491 |
+
"grad_norm": 1.20346200466156,
|
| 1492 |
+
"learning_rate": 1.1012578595953593e-05,
|
| 1493 |
+
"loss": 0.1394,
|
| 1494 |
+
"step": 1060
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.83061049011178,
|
| 1498 |
+
"grad_norm": 1.202398657798767,
|
| 1499 |
+
"learning_rate": 1.0875762027034203e-05,
|
| 1500 |
+
"loss": 0.1417,
|
| 1501 |
+
"step": 1065
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.8392089423903697,
|
| 1505 |
+
"grad_norm": 0.9019559621810913,
|
| 1506 |
+
"learning_rate": 1.0739315636596497e-05,
|
| 1507 |
+
"loss": 0.1349,
|
| 1508 |
+
"step": 1070
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.8478073946689597,
|
| 1512 |
+
"grad_norm": 1.0478671789169312,
|
| 1513 |
+
"learning_rate": 1.0603251671634447e-05,
|
| 1514 |
+
"loss": 0.1372,
|
| 1515 |
+
"step": 1075
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.8564058469475495,
|
| 1519 |
+
"grad_norm": 0.9594427347183228,
|
| 1520 |
+
"learning_rate": 1.0467582344816745e-05,
|
| 1521 |
+
"loss": 0.1262,
|
| 1522 |
+
"step": 1080
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.8650042992261393,
|
| 1526 |
+
"grad_norm": 1.2927871942520142,
|
| 1527 |
+
"learning_rate": 1.0332319833390626e-05,
|
| 1528 |
+
"loss": 0.1468,
|
| 1529 |
+
"step": 1085
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.873602751504729,
|
| 1533 |
+
"grad_norm": 1.110992193222046,
|
| 1534 |
+
"learning_rate": 1.0197476278088844e-05,
|
| 1535 |
+
"loss": 0.1526,
|
| 1536 |
+
"step": 1090
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.882201203783319,
|
| 1540 |
+
"grad_norm": 1.0229934453964233,
|
| 1541 |
+
"learning_rate": 1.006306378204001e-05,
|
| 1542 |
+
"loss": 0.1217,
|
| 1543 |
+
"step": 1095
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.8907996560619087,
|
| 1547 |
+
"grad_norm": 1.1181706190109253,
|
| 1548 |
+
"learning_rate": 9.92909440968221e-06,
|
| 1549 |
+
"loss": 0.1215,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.8993981083404987,
|
| 1554 |
+
"grad_norm": 0.9190809726715088,
|
| 1555 |
+
"learning_rate": 9.795580185680173e-06,
|
| 1556 |
+
"loss": 0.1277,
|
| 1557 |
+
"step": 1105
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.9079965606190885,
|
| 1561 |
+
"grad_norm": 1.1508560180664062,
|
| 1562 |
+
"learning_rate": 9.662533093845944e-06,
|
| 1563 |
+
"loss": 0.1182,
|
| 1564 |
+
"step": 1110
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.9165950128976785,
|
| 1568 |
+
"grad_norm": 1.0510079860687256,
|
| 1569 |
+
"learning_rate": 9.529965076063274e-06,
|
| 1570 |
+
"loss": 0.1172,
|
| 1571 |
+
"step": 1115
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.9251934651762683,
|
| 1575 |
+
"grad_norm": 1.0487672090530396,
|
| 1576 |
+
"learning_rate": 9.397888031215764e-06,
|
| 1577 |
+
"loss": 0.1289,
|
| 1578 |
+
"step": 1120
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.9337919174548581,
|
| 1582 |
+
"grad_norm": 0.9364534616470337,
|
| 1583 |
+
"learning_rate": 9.26631381411883e-06,
|
| 1584 |
+
"loss": 0.1441,
|
| 1585 |
+
"step": 1125
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.942390369733448,
|
| 1589 |
+
"grad_norm": 1.0469499826431274,
|
| 1590 |
+
"learning_rate": 9.135254234455673e-06,
|
| 1591 |
+
"loss": 0.1179,
|
| 1592 |
+
"step": 1130
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.9509888220120377,
|
| 1596 |
+
"grad_norm": 0.9588321447372437,
|
| 1597 |
+
"learning_rate": 9.004721055717276e-06,
|
| 1598 |
+
"loss": 0.1333,
|
| 1599 |
+
"step": 1135
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.9595872742906277,
|
| 1603 |
+
"grad_norm": 0.9027939438819885,
|
| 1604 |
+
"learning_rate": 8.874725994146554e-06,
|
| 1605 |
+
"loss": 0.1405,
|
| 1606 |
+
"step": 1140
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.9681857265692175,
|
| 1610 |
+
"grad_norm": 1.0032628774642944,
|
| 1611 |
+
"learning_rate": 8.745280717686731e-06,
|
| 1612 |
+
"loss": 0.1241,
|
| 1613 |
+
"step": 1145
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.9767841788478075,
|
| 1617 |
+
"grad_norm": 0.9550046920776367,
|
| 1618 |
+
"learning_rate": 8.616396844934071e-06,
|
| 1619 |
+
"loss": 0.1141,
|
| 1620 |
+
"step": 1150
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.9853826311263973,
|
| 1624 |
+
"grad_norm": 1.0686699151992798,
|
| 1625 |
+
"learning_rate": 8.488085944095023e-06,
|
| 1626 |
+
"loss": 0.1098,
|
| 1627 |
+
"step": 1155
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.9939810834049871,
|
| 1631 |
+
"grad_norm": 1.0226243734359741,
|
| 1632 |
+
"learning_rate": 8.360359531947906e-06,
|
| 1633 |
+
"loss": 0.1171,
|
| 1634 |
+
"step": 1160
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.001719690455718,
|
| 1638 |
+
"grad_norm": 0.7105857729911804,
|
| 1639 |
+
"learning_rate": 8.233229072809187e-06,
|
| 1640 |
+
"loss": 0.1112,
|
| 1641 |
+
"step": 1165
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 2.0103181427343078,
|
| 1645 |
+
"grad_norm": 1.111761212348938,
|
| 1646 |
+
"learning_rate": 8.106705977504493e-06,
|
| 1647 |
+
"loss": 0.0919,
|
| 1648 |
+
"step": 1170
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 2.0189165950128976,
|
| 1652 |
+
"grad_norm": 0.9137134552001953,
|
| 1653 |
+
"learning_rate": 7.980801602344396e-06,
|
| 1654 |
+
"loss": 0.0901,
|
| 1655 |
+
"step": 1175
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 2.0275150472914873,
|
| 1659 |
+
"grad_norm": 0.8778184056282043,
|
| 1660 |
+
"learning_rate": 7.855527248105127e-06,
|
| 1661 |
+
"loss": 0.0897,
|
| 1662 |
+
"step": 1180
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 2.0361134995700776,
|
| 1666 |
+
"grad_norm": 0.8700590133666992,
|
| 1667 |
+
"learning_rate": 7.730894159014233e-06,
|
| 1668 |
+
"loss": 0.1008,
|
| 1669 |
+
"step": 1185
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 2.0447119518486674,
|
| 1673 |
+
"grad_norm": 0.8803809881210327,
|
| 1674 |
+
"learning_rate": 7.606913521741351e-06,
|
| 1675 |
+
"loss": 0.0871,
|
| 1676 |
+
"step": 1190
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.053310404127257,
|
| 1680 |
+
"grad_norm": 0.7277244925498962,
|
| 1681 |
+
"learning_rate": 7.483596464394108e-06,
|
| 1682 |
+
"loss": 0.0944,
|
| 1683 |
+
"step": 1195
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.061908856405847,
|
| 1687 |
+
"grad_norm": 0.7915072441101074,
|
| 1688 |
+
"learning_rate": 7.360954055519314e-06,
|
| 1689 |
+
"loss": 0.0977,
|
| 1690 |
+
"step": 1200
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 2.0705073086844368,
|
| 1694 |
+
"grad_norm": 0.7213804125785828,
|
| 1695 |
+
"learning_rate": 7.238997303109478e-06,
|
| 1696 |
+
"loss": 0.0758,
|
| 1697 |
+
"step": 1205
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.0791057609630266,
|
| 1701 |
+
"grad_norm": 0.8061484098434448,
|
| 1702 |
+
"learning_rate": 7.117737153614765e-06,
|
| 1703 |
+
"loss": 0.0898,
|
| 1704 |
+
"step": 1210
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.0877042132416164,
|
| 1708 |
+
"grad_norm": 0.8220699429512024,
|
| 1709 |
+
"learning_rate": 6.997184490960476e-06,
|
| 1710 |
+
"loss": 0.0875,
|
| 1711 |
+
"step": 1215
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.096302665520206,
|
| 1715 |
+
"grad_norm": 0.8346285223960876,
|
| 1716 |
+
"learning_rate": 6.877350135570144e-06,
|
| 1717 |
+
"loss": 0.0923,
|
| 1718 |
+
"step": 1220
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.1049011177987964,
|
| 1722 |
+
"grad_norm": 0.90418940782547,
|
| 1723 |
+
"learning_rate": 6.758244843394331e-06,
|
| 1724 |
+
"loss": 0.0872,
|
| 1725 |
+
"step": 1225
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.113499570077386,
|
| 1729 |
+
"grad_norm": 0.8478304147720337,
|
| 1730 |
+
"learning_rate": 6.6398793049452e-06,
|
| 1731 |
+
"loss": 0.0982,
|
| 1732 |
+
"step": 1230
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.122098022355976,
|
| 1736 |
+
"grad_norm": 1.000467300415039,
|
| 1737 |
+
"learning_rate": 6.5222641443369715e-06,
|
| 1738 |
+
"loss": 0.0983,
|
| 1739 |
+
"step": 1235
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.130696474634566,
|
| 1743 |
+
"grad_norm": 0.6949626803398132,
|
| 1744 |
+
"learning_rate": 6.405409918332344e-06,
|
| 1745 |
+
"loss": 0.0858,
|
| 1746 |
+
"step": 1240
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 2.1392949269131556,
|
| 1750 |
+
"grad_norm": 0.7565341591835022,
|
| 1751 |
+
"learning_rate": 6.289327115394939e-06,
|
| 1752 |
+
"loss": 0.0907,
|
| 1753 |
+
"step": 1245
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 2.1478933791917454,
|
| 1757 |
+
"grad_norm": 0.9924172163009644,
|
| 1758 |
+
"learning_rate": 6.1740261547479025e-06,
|
| 1759 |
+
"loss": 0.0818,
|
| 1760 |
+
"step": 1250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.156491831470335,
|
| 1764 |
+
"grad_norm": 0.6371341347694397,
|
| 1765 |
+
"learning_rate": 6.0595173854386996e-06,
|
| 1766 |
+
"loss": 0.077,
|
| 1767 |
+
"step": 1255
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 2.165090283748925,
|
| 1771 |
+
"grad_norm": 0.8110918402671814,
|
| 1772 |
+
"learning_rate": 5.945811085410225e-06,
|
| 1773 |
+
"loss": 0.092,
|
| 1774 |
+
"step": 1260
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.1736887360275152,
|
| 1778 |
+
"grad_norm": 0.7571492195129395,
|
| 1779 |
+
"learning_rate": 5.832917460578276e-06,
|
| 1780 |
+
"loss": 0.0867,
|
| 1781 |
+
"step": 1265
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.182287188306105,
|
| 1785 |
+
"grad_norm": 0.8446589112281799,
|
| 1786 |
+
"learning_rate": 5.7208466439155285e-06,
|
| 1787 |
+
"loss": 0.0843,
|
| 1788 |
+
"step": 1270
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.190885640584695,
|
| 1792 |
+
"grad_norm": 0.7499189376831055,
|
| 1793 |
+
"learning_rate": 5.609608694541988e-06,
|
| 1794 |
+
"loss": 0.0846,
|
| 1795 |
+
"step": 1275
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.1994840928632846,
|
| 1799 |
+
"grad_norm": 0.7476209998130798,
|
| 1800 |
+
"learning_rate": 5.4992135968221535e-06,
|
| 1801 |
+
"loss": 0.0827,
|
| 1802 |
+
"step": 1280
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.2080825451418744,
|
| 1806 |
+
"grad_norm": 0.822342038154602,
|
| 1807 |
+
"learning_rate": 5.389671259468841e-06,
|
| 1808 |
+
"loss": 0.0807,
|
| 1809 |
+
"step": 1285
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.216680997420464,
|
| 1813 |
+
"grad_norm": 0.8006495833396912,
|
| 1814 |
+
"learning_rate": 5.280991514653821e-06,
|
| 1815 |
+
"loss": 0.0835,
|
| 1816 |
+
"step": 1290
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.225279449699054,
|
| 1820 |
+
"grad_norm": 0.9013881087303162,
|
| 1821 |
+
"learning_rate": 5.173184117125276e-06,
|
| 1822 |
+
"loss": 0.0824,
|
| 1823 |
+
"step": 1295
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.2338779019776442,
|
| 1827 |
+
"grad_norm": 0.7320542335510254,
|
| 1828 |
+
"learning_rate": 5.066258743332282e-06,
|
| 1829 |
+
"loss": 0.0905,
|
| 1830 |
+
"step": 1300
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.242476354256234,
|
| 1834 |
+
"grad_norm": 0.8053211569786072,
|
| 1835 |
+
"learning_rate": 4.960224990556286e-06,
|
| 1836 |
+
"loss": 0.0827,
|
| 1837 |
+
"step": 1305
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.251074806534824,
|
| 1841 |
+
"grad_norm": 0.6791473627090454,
|
| 1842 |
+
"learning_rate": 4.855092376049641e-06,
|
| 1843 |
+
"loss": 0.0896,
|
| 1844 |
+
"step": 1310
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.2596732588134136,
|
| 1848 |
+
"grad_norm": 0.7488510012626648,
|
| 1849 |
+
"learning_rate": 4.750870336181416e-06,
|
| 1850 |
+
"loss": 0.0869,
|
| 1851 |
+
"step": 1315
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.2682717110920034,
|
| 1855 |
+
"grad_norm": 0.7841968536376953,
|
| 1856 |
+
"learning_rate": 4.647568225590383e-06,
|
| 1857 |
+
"loss": 0.0638,
|
| 1858 |
+
"step": 1320
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.2768701633705932,
|
| 1862 |
+
"grad_norm": 0.8047657012939453,
|
| 1863 |
+
"learning_rate": 4.545195316345411e-06,
|
| 1864 |
+
"loss": 0.0855,
|
| 1865 |
+
"step": 1325
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.285468615649183,
|
| 1869 |
+
"grad_norm": 0.7320202589035034,
|
| 1870 |
+
"learning_rate": 4.44376079711318e-06,
|
| 1871 |
+
"loss": 0.0763,
|
| 1872 |
+
"step": 1330
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.294067067927773,
|
| 1876 |
+
"grad_norm": 0.7897084951400757,
|
| 1877 |
+
"learning_rate": 4.343273772333485e-06,
|
| 1878 |
+
"loss": 0.0837,
|
| 1879 |
+
"step": 1335
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.3026655202063626,
|
| 1883 |
+
"grad_norm": 0.6113478541374207,
|
| 1884 |
+
"learning_rate": 4.24374326140204e-06,
|
| 1885 |
+
"loss": 0.0774,
|
| 1886 |
+
"step": 1340
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.311263972484953,
|
| 1890 |
+
"grad_norm": 0.6849018335342407,
|
| 1891 |
+
"learning_rate": 4.145178197860911e-06,
|
| 1892 |
+
"loss": 0.0756,
|
| 1893 |
+
"step": 1345
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.3198624247635427,
|
| 1897 |
+
"grad_norm": 0.8156599402427673,
|
| 1898 |
+
"learning_rate": 4.04758742859666e-06,
|
| 1899 |
+
"loss": 0.0905,
|
| 1900 |
+
"step": 1350
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.3284608770421324,
|
| 1904 |
+
"grad_norm": 0.8464699983596802,
|
| 1905 |
+
"learning_rate": 3.950979713046326e-06,
|
| 1906 |
+
"loss": 0.0814,
|
| 1907 |
+
"step": 1355
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.3370593293207222,
|
| 1911 |
+
"grad_norm": 0.9155684113502502,
|
| 1912 |
+
"learning_rate": 3.855363722411151e-06,
|
| 1913 |
+
"loss": 0.0798,
|
| 1914 |
+
"step": 1360
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.345657781599312,
|
| 1918 |
+
"grad_norm": 0.7674224376678467,
|
| 1919 |
+
"learning_rate": 3.760748038878317e-06,
|
| 1920 |
+
"loss": 0.0851,
|
| 1921 |
+
"step": 1365
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.354256233877902,
|
| 1925 |
+
"grad_norm": 0.5711732506752014,
|
| 1926 |
+
"learning_rate": 3.667141154850599e-06,
|
| 1927 |
+
"loss": 0.0734,
|
| 1928 |
+
"step": 1370
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.362854686156492,
|
| 1932 |
+
"grad_norm": 0.8187562823295593,
|
| 1933 |
+
"learning_rate": 3.5745514721841693e-06,
|
| 1934 |
+
"loss": 0.0719,
|
| 1935 |
+
"step": 1375
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.371453138435082,
|
| 1939 |
+
"grad_norm": 0.8430051207542419,
|
| 1940 |
+
"learning_rate": 3.4829873014344236e-06,
|
| 1941 |
+
"loss": 0.0815,
|
| 1942 |
+
"step": 1380
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.3800515907136717,
|
| 1946 |
+
"grad_norm": 0.9995542764663696,
|
| 1947 |
+
"learning_rate": 3.3924568611100595e-06,
|
| 1948 |
+
"loss": 0.0682,
|
| 1949 |
+
"step": 1385
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.3886500429922615,
|
| 1953 |
+
"grad_norm": 0.7107648253440857,
|
| 1954 |
+
"learning_rate": 3.3029682769354476e-06,
|
| 1955 |
+
"loss": 0.0768,
|
| 1956 |
+
"step": 1390
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.3972484952708513,
|
| 1960 |
+
"grad_norm": 0.8143495917320251,
|
| 1961 |
+
"learning_rate": 3.2145295811212417e-06,
|
| 1962 |
+
"loss": 0.0733,
|
| 1963 |
+
"step": 1395
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.405846947549441,
|
| 1967 |
+
"grad_norm": 0.6573176980018616,
|
| 1968 |
+
"learning_rate": 3.12714871164347e-06,
|
| 1969 |
+
"loss": 0.0716,
|
| 1970 |
+
"step": 1400
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.414445399828031,
|
| 1974 |
+
"grad_norm": 0.7246174216270447,
|
| 1975 |
+
"learning_rate": 3.0408335115310154e-06,
|
| 1976 |
+
"loss": 0.0813,
|
| 1977 |
+
"step": 1405
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.4230438521066207,
|
| 1981 |
+
"grad_norm": 0.7900003790855408,
|
| 1982 |
+
"learning_rate": 2.9555917281616907e-06,
|
| 1983 |
+
"loss": 0.0744,
|
| 1984 |
+
"step": 1410
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.4316423043852105,
|
| 1988 |
+
"grad_norm": 0.5694577693939209,
|
| 1989 |
+
"learning_rate": 2.8714310125668226e-06,
|
| 1990 |
+
"loss": 0.0802,
|
| 1991 |
+
"step": 1415
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.4402407566638007,
|
| 1995 |
+
"grad_norm": 0.5761277079582214,
|
| 1996 |
+
"learning_rate": 2.788358918744536e-06,
|
| 1997 |
+
"loss": 0.0952,
|
| 1998 |
+
"step": 1420
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.4488392089423905,
|
| 2002 |
+
"grad_norm": 0.7433187365531921,
|
| 2003 |
+
"learning_rate": 2.70638290298173e-06,
|
| 2004 |
+
"loss": 0.0686,
|
| 2005 |
+
"step": 1425
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.4574376612209803,
|
| 2009 |
+
"grad_norm": 0.7149344682693481,
|
| 2010 |
+
"learning_rate": 2.625510323184821e-06,
|
| 2011 |
+
"loss": 0.0793,
|
| 2012 |
+
"step": 1430
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.46603611349957,
|
| 2016 |
+
"grad_norm": 0.7665093541145325,
|
| 2017 |
+
"learning_rate": 2.5457484382193246e-06,
|
| 2018 |
+
"loss": 0.0838,
|
| 2019 |
+
"step": 1435
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.47463456577816,
|
| 2023 |
+
"grad_norm": 0.7998213171958923,
|
| 2024 |
+
"learning_rate": 2.4671044072583205e-06,
|
| 2025 |
+
"loss": 0.0712,
|
| 2026 |
+
"step": 1440
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.4832330180567497,
|
| 2030 |
+
"grad_norm": 0.625607430934906,
|
| 2031 |
+
"learning_rate": 2.389585289139869e-06,
|
| 2032 |
+
"loss": 0.0759,
|
| 2033 |
+
"step": 1445
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.4918314703353395,
|
| 2037 |
+
"grad_norm": 0.5757770538330078,
|
| 2038 |
+
"learning_rate": 2.3131980417334324e-06,
|
| 2039 |
+
"loss": 0.0782,
|
| 2040 |
+
"step": 1450
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.5004299226139297,
|
| 2044 |
+
"grad_norm": 0.7464799284934998,
|
| 2045 |
+
"learning_rate": 2.2379495213153577e-06,
|
| 2046 |
+
"loss": 0.0781,
|
| 2047 |
+
"step": 1455
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.5090283748925195,
|
| 2051 |
+
"grad_norm": 0.588624894618988,
|
| 2052 |
+
"learning_rate": 2.1638464819534816e-06,
|
| 2053 |
+
"loss": 0.0749,
|
| 2054 |
+
"step": 1460
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.5176268271711093,
|
| 2058 |
+
"grad_norm": 0.5571017861366272,
|
| 2059 |
+
"learning_rate": 2.0908955749009e-06,
|
| 2060 |
+
"loss": 0.0719,
|
| 2061 |
+
"step": 1465
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.526225279449699,
|
| 2065 |
+
"grad_norm": 0.7140423655509949,
|
| 2066 |
+
"learning_rate": 2.019103347998983e-06,
|
| 2067 |
+
"loss": 0.0719,
|
| 2068 |
+
"step": 1470
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.534823731728289,
|
| 2072 |
+
"grad_norm": 0.7077119946479797,
|
| 2073 |
+
"learning_rate": 1.948476245089653e-06,
|
| 2074 |
+
"loss": 0.0721,
|
| 2075 |
+
"step": 1475
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.5434221840068787,
|
| 2079 |
+
"grad_norm": 0.607108473777771,
|
| 2080 |
+
"learning_rate": 1.8790206054370095e-06,
|
| 2081 |
+
"loss": 0.0749,
|
| 2082 |
+
"step": 1480
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.5520206362854685,
|
| 2086 |
+
"grad_norm": 0.750508189201355,
|
| 2087 |
+
"learning_rate": 1.8107426631583362e-06,
|
| 2088 |
+
"loss": 0.0702,
|
| 2089 |
+
"step": 1485
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.5606190885640583,
|
| 2093 |
+
"grad_norm": 0.5918256640434265,
|
| 2094 |
+
"learning_rate": 1.743648546664547e-06,
|
| 2095 |
+
"loss": 0.071,
|
| 2096 |
+
"step": 1490
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.569217540842648,
|
| 2100 |
+
"grad_norm": 0.6715498566627502,
|
| 2101 |
+
"learning_rate": 1.67774427811012e-06,
|
| 2102 |
+
"loss": 0.0688,
|
| 2103 |
+
"step": 1495
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.5778159931212383,
|
| 2107 |
+
"grad_norm": 0.5247796177864075,
|
| 2108 |
+
"learning_rate": 1.613035772852564e-06,
|
| 2109 |
+
"loss": 0.0703,
|
| 2110 |
+
"step": 1500
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.586414445399828,
|
| 2114 |
+
"grad_norm": 0.7841174602508545,
|
| 2115 |
+
"learning_rate": 1.54952883892148e-06,
|
| 2116 |
+
"loss": 0.0705,
|
| 2117 |
+
"step": 1505
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.595012897678418,
|
| 2121 |
+
"grad_norm": 0.5807991027832031,
|
| 2122 |
+
"learning_rate": 1.4872291764972506e-06,
|
| 2123 |
+
"loss": 0.0812,
|
| 2124 |
+
"step": 1510
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.6036113499570077,
|
| 2128 |
+
"grad_norm": 0.6700941920280457,
|
| 2129 |
+
"learning_rate": 1.4261423773994047e-06,
|
| 2130 |
+
"loss": 0.0796,
|
| 2131 |
+
"step": 1515
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.6122098022355975,
|
| 2135 |
+
"grad_norm": 0.6654531955718994,
|
| 2136 |
+
"learning_rate": 1.3662739245847193e-06,
|
| 2137 |
+
"loss": 0.0636,
|
| 2138 |
+
"step": 1520
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.6208082545141873,
|
| 2142 |
+
"grad_norm": 0.6881716847419739,
|
| 2143 |
+
"learning_rate": 1.3076291916550797e-06,
|
| 2144 |
+
"loss": 0.0721,
|
| 2145 |
+
"step": 1525
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.6294067067927775,
|
| 2149 |
+
"grad_norm": 0.6220394372940063,
|
| 2150 |
+
"learning_rate": 1.2502134423751816e-06,
|
| 2151 |
+
"loss": 0.0684,
|
| 2152 |
+
"step": 1530
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.6380051590713673,
|
| 2156 |
+
"grad_norm": 0.5421865582466125,
|
| 2157 |
+
"learning_rate": 1.1940318302000342e-06,
|
| 2158 |
+
"loss": 0.0662,
|
| 2159 |
+
"step": 1535
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.646603611349957,
|
| 2163 |
+
"grad_norm": 0.6841817498207092,
|
| 2164 |
+
"learning_rate": 1.1390893978124456e-06,
|
| 2165 |
+
"loss": 0.0722,
|
| 2166 |
+
"step": 1540
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.655202063628547,
|
| 2170 |
+
"grad_norm": 0.6115626692771912,
|
| 2171 |
+
"learning_rate": 1.085391076670378e-06,
|
| 2172 |
+
"loss": 0.0729,
|
| 2173 |
+
"step": 1545
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.6638005159071367,
|
| 2177 |
+
"grad_norm": 0.6052408814430237,
|
| 2178 |
+
"learning_rate": 1.0329416865643276e-06,
|
| 2179 |
+
"loss": 0.0788,
|
| 2180 |
+
"step": 1550
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.6723989681857265,
|
| 2184 |
+
"grad_norm": 0.5840135812759399,
|
| 2185 |
+
"learning_rate": 9.817459351847175e-07,
|
| 2186 |
+
"loss": 0.0603,
|
| 2187 |
+
"step": 1555
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.6809974204643163,
|
| 2191 |
+
"grad_norm": 0.6318619251251221,
|
| 2192 |
+
"learning_rate": 9.318084176993385e-07,
|
| 2193 |
+
"loss": 0.0711,
|
| 2194 |
+
"step": 1560
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.689595872742906,
|
| 2198 |
+
"grad_norm": 0.766850471496582,
|
| 2199 |
+
"learning_rate": 8.83133616340921e-07,
|
| 2200 |
+
"loss": 0.0739,
|
| 2201 |
+
"step": 1565
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.698194325021496,
|
| 2205 |
+
"grad_norm": 0.651466965675354,
|
| 2206 |
+
"learning_rate": 8.357259000048001e-07,
|
| 2207 |
+
"loss": 0.073,
|
| 2208 |
+
"step": 1570
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.7067927773000857,
|
| 2212 |
+
"grad_norm": 0.6638359427452087,
|
| 2213 |
+
"learning_rate": 7.895895238567924e-07,
|
| 2214 |
+
"loss": 0.072,
|
| 2215 |
+
"step": 1575
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.715391229578676,
|
| 2219 |
+
"grad_norm": 0.7421326637268066,
|
| 2220 |
+
"learning_rate": 7.447286289512722e-07,
|
| 2221 |
+
"loss": 0.0638,
|
| 2222 |
+
"step": 1580
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.7239896818572658,
|
| 2226 |
+
"grad_norm": 0.79677414894104,
|
| 2227 |
+
"learning_rate": 7.011472418594678e-07,
|
| 2228 |
+
"loss": 0.0768,
|
| 2229 |
+
"step": 1585
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.7325881341358556,
|
| 2233 |
+
"grad_norm": 0.6642888188362122,
|
| 2234 |
+
"learning_rate": 6.588492743080532e-07,
|
| 2235 |
+
"loss": 0.0814,
|
| 2236 |
+
"step": 1590
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.7411865864144453,
|
| 2240 |
+
"grad_norm": 0.5246862769126892,
|
| 2241 |
+
"learning_rate": 6.178385228280481e-07,
|
| 2242 |
+
"loss": 0.0673,
|
| 2243 |
+
"step": 1595
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.749785038693035,
|
| 2247 |
+
"grad_norm": 0.5967271327972412,
|
| 2248 |
+
"learning_rate": 5.781186684140632e-07,
|
| 2249 |
+
"loss": 0.0717,
|
| 2250 |
+
"step": 1600
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.7583834909716254,
|
| 2254 |
+
"grad_norm": 0.675962507724762,
|
| 2255 |
+
"learning_rate": 5.396932761938783e-07,
|
| 2256 |
+
"loss": 0.0768,
|
| 2257 |
+
"step": 1605
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.766981943250215,
|
| 2261 |
+
"grad_norm": 0.5288321375846863,
|
| 2262 |
+
"learning_rate": 5.025657951084728e-07,
|
| 2263 |
+
"loss": 0.0726,
|
| 2264 |
+
"step": 1610
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.775580395528805,
|
| 2268 |
+
"grad_norm": 0.7039914131164551,
|
| 2269 |
+
"learning_rate": 4.6673955760245355e-07,
|
| 2270 |
+
"loss": 0.0685,
|
| 2271 |
+
"step": 1615
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.7841788478073948,
|
| 2275 |
+
"grad_norm": 0.5980196595191956,
|
| 2276 |
+
"learning_rate": 4.3221777932493765e-07,
|
| 2277 |
+
"loss": 0.071,
|
| 2278 |
+
"step": 1620
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 2.7927773000859846,
|
| 2282 |
+
"grad_norm": 0.49303629994392395,
|
| 2283 |
+
"learning_rate": 3.990035588409313e-07,
|
| 2284 |
+
"loss": 0.0603,
|
| 2285 |
+
"step": 1625
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.8013757523645744,
|
| 2289 |
+
"grad_norm": 0.7239757180213928,
|
| 2290 |
+
"learning_rate": 3.670998773532175e-07,
|
| 2291 |
+
"loss": 0.0794,
|
| 2292 |
+
"step": 1630
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.809974204643164,
|
| 2296 |
+
"grad_norm": 0.4862307608127594,
|
| 2297 |
+
"learning_rate": 3.3650959843476323e-07,
|
| 2298 |
+
"loss": 0.0723,
|
| 2299 |
+
"step": 1635
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.818572656921754,
|
| 2303 |
+
"grad_norm": 0.4852212071418762,
|
| 2304 |
+
"learning_rate": 3.0723546777169875e-07,
|
| 2305 |
+
"loss": 0.0625,
|
| 2306 |
+
"step": 1640
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 2.8271711092003438,
|
| 2310 |
+
"grad_norm": 0.5736478567123413,
|
| 2311 |
+
"learning_rate": 2.792801129168726e-07,
|
| 2312 |
+
"loss": 0.0682,
|
| 2313 |
+
"step": 1645
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.8357695614789336,
|
| 2317 |
+
"grad_norm": 0.6324381232261658,
|
| 2318 |
+
"learning_rate": 2.5264604305401727e-07,
|
| 2319 |
+
"loss": 0.0723,
|
| 2320 |
+
"step": 1650
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.844368013757524,
|
| 2324 |
+
"grad_norm": 0.5429121851921082,
|
| 2325 |
+
"learning_rate": 2.2733564877252254e-07,
|
| 2326 |
+
"loss": 0.0659,
|
| 2327 |
+
"step": 1655
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.8529664660361136,
|
| 2331 |
+
"grad_norm": 0.7229666113853455,
|
| 2332 |
+
"learning_rate": 2.0335120185286714e-07,
|
| 2333 |
+
"loss": 0.0651,
|
| 2334 |
+
"step": 1660
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.8615649183147034,
|
| 2338 |
+
"grad_norm": 0.6239649653434753,
|
| 2339 |
+
"learning_rate": 1.8069485506271688e-07,
|
| 2340 |
+
"loss": 0.0703,
|
| 2341 |
+
"step": 1665
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.870163370593293,
|
| 2345 |
+
"grad_norm": 0.557466983795166,
|
| 2346 |
+
"learning_rate": 1.5936864196368915e-07,
|
| 2347 |
+
"loss": 0.064,
|
| 2348 |
+
"step": 1670
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.878761822871883,
|
| 2352 |
+
"grad_norm": 0.5991723537445068,
|
| 2353 |
+
"learning_rate": 1.3937447672884074e-07,
|
| 2354 |
+
"loss": 0.0723,
|
| 2355 |
+
"step": 1675
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.8873602751504728,
|
| 2359 |
+
"grad_norm": 0.6043932437896729,
|
| 2360 |
+
"learning_rate": 1.2071415397083674e-07,
|
| 2361 |
+
"loss": 0.0701,
|
| 2362 |
+
"step": 1680
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.895958727429063,
|
| 2366 |
+
"grad_norm": 0.6801122426986694,
|
| 2367 |
+
"learning_rate": 1.0338934858089622e-07,
|
| 2368 |
+
"loss": 0.0732,
|
| 2369 |
+
"step": 1685
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.904557179707653,
|
| 2373 |
+
"grad_norm": 0.737115204334259,
|
| 2374 |
+
"learning_rate": 8.740161557844073e-08,
|
| 2375 |
+
"loss": 0.062,
|
| 2376 |
+
"step": 1690
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.9131556319862426,
|
| 2380 |
+
"grad_norm": 0.6453583240509033,
|
| 2381 |
+
"learning_rate": 7.2752389971531e-08,
|
| 2382 |
+
"loss": 0.0668,
|
| 2383 |
+
"step": 1695
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.9217540842648324,
|
| 2387 |
+
"grad_norm": 0.669319212436676,
|
| 2388 |
+
"learning_rate": 5.944298662805825e-08,
|
| 2389 |
+
"loss": 0.0738,
|
| 2390 |
+
"step": 1700
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.930352536543422,
|
| 2394 |
+
"grad_norm": 0.679260790348053,
|
| 2395 |
+
"learning_rate": 4.747460015773031e-08,
|
| 2396 |
+
"loss": 0.0699,
|
| 2397 |
+
"step": 1705
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.938950988822012,
|
| 2401 |
+
"grad_norm": 0.5769382119178772,
|
| 2402 |
+
"learning_rate": 3.6848304804847356e-08,
|
| 2403 |
+
"loss": 0.0676,
|
| 2404 |
+
"step": 1710
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 2.947549441100602,
|
| 2408 |
+
"grad_norm": 0.5470510721206665,
|
| 2409 |
+
"learning_rate": 2.756505435187906e-08,
|
| 2410 |
+
"loss": 0.0761,
|
| 2411 |
+
"step": 1715
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 2.9561478933791916,
|
| 2415 |
+
"grad_norm": 0.6554555892944336,
|
| 2416 |
+
"learning_rate": 1.9625682033856394e-08,
|
| 2417 |
+
"loss": 0.0754,
|
| 2418 |
+
"step": 1720
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 2.9647463456577814,
|
| 2422 |
+
"grad_norm": 0.5629612803459167,
|
| 2423 |
+
"learning_rate": 1.3030900463584794e-08,
|
| 2424 |
+
"loss": 0.0772,
|
| 2425 |
+
"step": 1725
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.973344797936371,
|
| 2429 |
+
"grad_norm": 0.6302787661552429,
|
| 2430 |
+
"learning_rate": 7.781301567681976e-09,
|
| 2431 |
+
"loss": 0.0695,
|
| 2432 |
+
"step": 1730
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 2.9819432502149614,
|
| 2436 |
+
"grad_norm": 0.553763747215271,
|
| 2437 |
+
"learning_rate": 3.877356533445453e-09,
|
| 2438 |
+
"loss": 0.0719,
|
| 2439 |
+
"step": 1735
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 2.9905417024935512,
|
| 2443 |
+
"grad_norm": 0.7303978204727173,
|
| 2444 |
+
"learning_rate": 1.3194157665613605e-09,
|
| 2445 |
+
"loss": 0.0697,
|
| 2446 |
+
"step": 1740
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 2.999140154772141,
|
| 2450 |
+
"grad_norm": 0.5900354385375977,
|
| 2451 |
+
"learning_rate": 1.0770885965960542e-10,
|
| 2452 |
+
"loss": 0.0631,
|
| 2453 |
+
"step": 1745
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 3.0,
|
| 2457 |
+
"step": 1746,
|
| 2458 |
+
"total_flos": 2.174167758776828e+18,
|
| 2459 |
+
"train_loss": 0.3808436165718449,
|
| 2460 |
+
"train_runtime": 1120.1195,
|
| 2461 |
+
"train_samples_per_second": 49.805,
|
| 2462 |
+
"train_steps_per_second": 1.559
|
| 2463 |
+
}
|
| 2464 |
+
],
|
| 2465 |
+
"logging_steps": 5,
|
| 2466 |
+
"max_steps": 1746,
|
| 2467 |
+
"num_input_tokens_seen": 0,
|
| 2468 |
+
"num_train_epochs": 3,
|
| 2469 |
+
"save_steps": 20000,
|
| 2470 |
+
"stateful_callbacks": {
|
| 2471 |
+
"TrainerControl": {
|
| 2472 |
+
"args": {
|
| 2473 |
+
"should_epoch_stop": false,
|
| 2474 |
+
"should_evaluate": false,
|
| 2475 |
+
"should_log": false,
|
| 2476 |
+
"should_save": false,
|
| 2477 |
+
"should_training_stop": false
|
| 2478 |
+
},
|
| 2479 |
+
"attributes": {}
|
| 2480 |
+
}
|
| 2481 |
+
},
|
| 2482 |
+
"total_flos": 2.174167758776828e+18,
|
| 2483 |
+
"train_batch_size": 2,
|
| 2484 |
+
"trial_name": null,
|
| 2485 |
+
"trial_params": null
|
| 2486 |
+
}
|
instruct/103_128_e3_3e-5/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f5e96d29258269e6b2981bbe061b0e9657f9af08ab8e807623e38917c428e20
|
| 3 |
+
size 8273
|
instruct/103_128_e3_3e-5/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|