Upload folder using huggingface_hub
Browse files- 139_128_e3_3e-5/adapter_config.json +39 -0
- 139_128_e3_3e-5/adapter_model.safetensors +3 -0
- 139_128_e3_3e-5/added_tokens.json +9 -0
- 139_128_e3_3e-5/all_results.json +9 -0
- 139_128_e3_3e-5/chat_template.jinja +62 -0
- 139_128_e3_3e-5/config.json +32 -0
- 139_128_e3_3e-5/merges.txt +0 -0
- 139_128_e3_3e-5/special_tokens_map.json +33 -0
- 139_128_e3_3e-5/tokenizer.json +0 -0
- 139_128_e3_3e-5/tokenizer_config.json +234 -0
- 139_128_e3_3e-5/train_results.json +9 -0
- 139_128_e3_3e-5/trainer_state.json +2248 -0
- 139_128_e3_3e-5/training_args.bin +3 -0
- 139_128_e3_3e-5/vocab.json +0 -0
139_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 |
+
"k_proj",
|
| 28 |
+
"v_proj",
|
| 29 |
+
"up_proj",
|
| 30 |
+
"q_proj",
|
| 31 |
+
"o_proj",
|
| 32 |
+
"down_proj",
|
| 33 |
+
"gate_proj"
|
| 34 |
+
],
|
| 35 |
+
"task_type": "CAUSAL_LM",
|
| 36 |
+
"trainable_token_indices": null,
|
| 37 |
+
"use_dora": false,
|
| 38 |
+
"use_rslora": false
|
| 39 |
+
}
|
139_128_e3_3e-5/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e72f3660351b5d7a9c937645ba8a9384c1f83213abfcde000e7dd438df01f47
|
| 3 |
+
size 791751704
|
139_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 |
+
}
|
139_128_e3_3e-5/all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.066537064018477e+18,
|
| 4 |
+
"train_loss": 0.3455555534464778,
|
| 5 |
+
"train_runtime": 975.374,
|
| 6 |
+
"train_samples": 16809,
|
| 7 |
+
"train_samples_per_second": 51.7,
|
| 8 |
+
"train_steps_per_second": 1.618
|
| 9 |
+
}
|
139_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 %}
|
139_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 |
+
}
|
139_128_e3_3e-5/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
139_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 |
+
}
|
139_128_e3_3e-5/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
139_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 |
+
}
|
139_128_e3_3e-5/train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.066537064018477e+18,
|
| 4 |
+
"train_loss": 0.3455555534464778,
|
| 5 |
+
"train_runtime": 975.374,
|
| 6 |
+
"train_samples": 16809,
|
| 7 |
+
"train_samples_per_second": 51.7,
|
| 8 |
+
"train_steps_per_second": 1.618
|
| 9 |
+
}
|
139_128_e3_3e-5/trainer_state.json
ADDED
|
@@ -0,0 +1,2248 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1578,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.009514747859181731,
|
| 14 |
+
"grad_norm": 2.375884771347046,
|
| 15 |
+
"learning_rate": 1.518987341772152e-06,
|
| 16 |
+
"loss": 1.4851,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.019029495718363463,
|
| 21 |
+
"grad_norm": 1.9834611415863037,
|
| 22 |
+
"learning_rate": 3.4177215189873417e-06,
|
| 23 |
+
"loss": 1.5367,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.028544243577545196,
|
| 28 |
+
"grad_norm": 1.1209793090820312,
|
| 29 |
+
"learning_rate": 5.3164556962025316e-06,
|
| 30 |
+
"loss": 1.476,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.038058991436726926,
|
| 35 |
+
"grad_norm": 0.6614314317703247,
|
| 36 |
+
"learning_rate": 7.215189873417722e-06,
|
| 37 |
+
"loss": 1.4183,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.047573739295908656,
|
| 42 |
+
"grad_norm": 0.5300923585891724,
|
| 43 |
+
"learning_rate": 9.113924050632912e-06,
|
| 44 |
+
"loss": 1.3637,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05708848715509039,
|
| 49 |
+
"grad_norm": 0.4909587800502777,
|
| 50 |
+
"learning_rate": 1.1012658227848103e-05,
|
| 51 |
+
"loss": 1.3328,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06660323501427212,
|
| 56 |
+
"grad_norm": 0.6743994951248169,
|
| 57 |
+
"learning_rate": 1.2911392405063291e-05,
|
| 58 |
+
"loss": 1.3209,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07611798287345385,
|
| 63 |
+
"grad_norm": 0.4343916177749634,
|
| 64 |
+
"learning_rate": 1.4810126582278482e-05,
|
| 65 |
+
"loss": 1.3521,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.08563273073263558,
|
| 70 |
+
"grad_norm": 0.44983208179473877,
|
| 71 |
+
"learning_rate": 1.670886075949367e-05,
|
| 72 |
+
"loss": 1.3475,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.09514747859181731,
|
| 77 |
+
"grad_norm": 0.40172743797302246,
|
| 78 |
+
"learning_rate": 1.860759493670886e-05,
|
| 79 |
+
"loss": 1.3371,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.10466222645099905,
|
| 84 |
+
"grad_norm": 0.37942150235176086,
|
| 85 |
+
"learning_rate": 2.050632911392405e-05,
|
| 86 |
+
"loss": 1.247,
|
| 87 |
+
"step": 55
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.11417697431018078,
|
| 91 |
+
"grad_norm": 0.40726006031036377,
|
| 92 |
+
"learning_rate": 2.240506329113924e-05,
|
| 93 |
+
"loss": 1.2731,
|
| 94 |
+
"step": 60
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.12369172216936251,
|
| 98 |
+
"grad_norm": 0.4973227083683014,
|
| 99 |
+
"learning_rate": 2.430379746835443e-05,
|
| 100 |
+
"loss": 1.2838,
|
| 101 |
+
"step": 65
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.13320647002854424,
|
| 105 |
+
"grad_norm": 0.49249252676963806,
|
| 106 |
+
"learning_rate": 2.620253164556962e-05,
|
| 107 |
+
"loss": 1.2508,
|
| 108 |
+
"step": 70
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.142721217887726,
|
| 112 |
+
"grad_norm": 0.5110635757446289,
|
| 113 |
+
"learning_rate": 2.8101265822784812e-05,
|
| 114 |
+
"loss": 1.198,
|
| 115 |
+
"step": 75
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.1522359657469077,
|
| 119 |
+
"grad_norm": 0.4833797514438629,
|
| 120 |
+
"learning_rate": 3e-05,
|
| 121 |
+
"loss": 1.2047,
|
| 122 |
+
"step": 80
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.16175071360608945,
|
| 126 |
+
"grad_norm": 0.5108519792556763,
|
| 127 |
+
"learning_rate": 2.9999176442782448e-05,
|
| 128 |
+
"loss": 1.214,
|
| 129 |
+
"step": 85
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.17126546146527116,
|
| 133 |
+
"grad_norm": 0.5081245303153992,
|
| 134 |
+
"learning_rate": 2.999670586156266e-05,
|
| 135 |
+
"loss": 1.1913,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1807802093244529,
|
| 140 |
+
"grad_norm": 0.5639904737472534,
|
| 141 |
+
"learning_rate": 2.9992588527629303e-05,
|
| 142 |
+
"loss": 1.181,
|
| 143 |
+
"step": 95
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.19029495718363462,
|
| 147 |
+
"grad_norm": 0.5344358086585999,
|
| 148 |
+
"learning_rate": 2.998682489309705e-05,
|
| 149 |
+
"loss": 1.1447,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.19980970504281637,
|
| 154 |
+
"grad_norm": 0.5228133201599121,
|
| 155 |
+
"learning_rate": 2.9979415590856944e-05,
|
| 156 |
+
"loss": 1.1065,
|
| 157 |
+
"step": 105
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.2093244529019981,
|
| 161 |
+
"grad_norm": 0.5331887602806091,
|
| 162 |
+
"learning_rate": 2.9970361434506903e-05,
|
| 163 |
+
"loss": 1.1229,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.21883920076117983,
|
| 168 |
+
"grad_norm": 0.5858896374702454,
|
| 169 |
+
"learning_rate": 2.9959663418262358e-05,
|
| 170 |
+
"loss": 1.1146,
|
| 171 |
+
"step": 115
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.22835394862036157,
|
| 175 |
+
"grad_norm": 0.6496222615242004,
|
| 176 |
+
"learning_rate": 2.9947322716847114e-05,
|
| 177 |
+
"loss": 1.0549,
|
| 178 |
+
"step": 120
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.23786869647954328,
|
| 182 |
+
"grad_norm": 0.5937091708183289,
|
| 183 |
+
"learning_rate": 2.9933340685364335e-05,
|
| 184 |
+
"loss": 1.0713,
|
| 185 |
+
"step": 125
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.24738344433872503,
|
| 189 |
+
"grad_norm": 0.5986030101776123,
|
| 190 |
+
"learning_rate": 2.9917718859147743e-05,
|
| 191 |
+
"loss": 1.0802,
|
| 192 |
+
"step": 130
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.25689819219790677,
|
| 196 |
+
"grad_norm": 0.6456966400146484,
|
| 197 |
+
"learning_rate": 2.990045895359304e-05,
|
| 198 |
+
"loss": 0.9931,
|
| 199 |
+
"step": 135
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2664129400570885,
|
| 203 |
+
"grad_norm": 0.6817247867584229,
|
| 204 |
+
"learning_rate": 2.9881562863969535e-05,
|
| 205 |
+
"loss": 1.0154,
|
| 206 |
+
"step": 140
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2759276879162702,
|
| 210 |
+
"grad_norm": 0.6692793369293213,
|
| 211 |
+
"learning_rate": 2.986103266521201e-05,
|
| 212 |
+
"loss": 0.9671,
|
| 213 |
+
"step": 145
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.285442435775452,
|
| 217 |
+
"grad_norm": 0.7419658303260803,
|
| 218 |
+
"learning_rate": 2.983887061169293e-05,
|
| 219 |
+
"loss": 1.0442,
|
| 220 |
+
"step": 150
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2949571836346337,
|
| 224 |
+
"grad_norm": 0.7353575229644775,
|
| 225 |
+
"learning_rate": 2.981507913697484e-05,
|
| 226 |
+
"loss": 0.9401,
|
| 227 |
+
"step": 155
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.3044719314938154,
|
| 231 |
+
"grad_norm": 0.6603230237960815,
|
| 232 |
+
"learning_rate": 2.9789660853543176e-05,
|
| 233 |
+
"loss": 0.8802,
|
| 234 |
+
"step": 160
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.3139866793529971,
|
| 238 |
+
"grad_norm": 0.7152225375175476,
|
| 239 |
+
"learning_rate": 2.9762618552519366e-05,
|
| 240 |
+
"loss": 0.9031,
|
| 241 |
+
"step": 165
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3235014272121789,
|
| 245 |
+
"grad_norm": 0.850296139717102,
|
| 246 |
+
"learning_rate": 2.973395520335437e-05,
|
| 247 |
+
"loss": 0.901,
|
| 248 |
+
"step": 170
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3330161750713606,
|
| 252 |
+
"grad_norm": 0.9184004664421082,
|
| 253 |
+
"learning_rate": 2.9703673953502598e-05,
|
| 254 |
+
"loss": 0.8877,
|
| 255 |
+
"step": 175
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3425309229305423,
|
| 259 |
+
"grad_norm": 0.7570393681526184,
|
| 260 |
+
"learning_rate": 2.967177812807631e-05,
|
| 261 |
+
"loss": 0.9255,
|
| 262 |
+
"step": 180
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3520456707897241,
|
| 266 |
+
"grad_norm": 0.8172934055328369,
|
| 267 |
+
"learning_rate": 2.9638271229480458e-05,
|
| 268 |
+
"loss": 0.9145,
|
| 269 |
+
"step": 185
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.3615604186489058,
|
| 273 |
+
"grad_norm": 0.7811704874038696,
|
| 274 |
+
"learning_rate": 2.9603156937028142e-05,
|
| 275 |
+
"loss": 0.9125,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.37107516650808753,
|
| 280 |
+
"grad_norm": 0.9186906814575195,
|
| 281 |
+
"learning_rate": 2.9566439106536554e-05,
|
| 282 |
+
"loss": 0.8972,
|
| 283 |
+
"step": 195
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.38058991436726924,
|
| 287 |
+
"grad_norm": 0.9056068062782288,
|
| 288 |
+
"learning_rate": 2.9528121769903604e-05,
|
| 289 |
+
"loss": 0.8137,
|
| 290 |
+
"step": 200
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.390104662226451,
|
| 294 |
+
"grad_norm": 0.8188087940216064,
|
| 295 |
+
"learning_rate": 2.948820913466518e-05,
|
| 296 |
+
"loss": 0.842,
|
| 297 |
+
"step": 205
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.39961941008563273,
|
| 301 |
+
"grad_norm": 1.1099237203598022,
|
| 302 |
+
"learning_rate": 2.944670558353312e-05,
|
| 303 |
+
"loss": 0.8465,
|
| 304 |
+
"step": 210
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.40913415794481445,
|
| 308 |
+
"grad_norm": 0.921646237373352,
|
| 309 |
+
"learning_rate": 2.9403615673913977e-05,
|
| 310 |
+
"loss": 0.8026,
|
| 311 |
+
"step": 215
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.4186489058039962,
|
| 315 |
+
"grad_norm": 0.9761255383491516,
|
| 316 |
+
"learning_rate": 2.9358944137408553e-05,
|
| 317 |
+
"loss": 0.8248,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.42816365366317793,
|
| 322 |
+
"grad_norm": 0.9675289988517761,
|
| 323 |
+
"learning_rate": 2.931269587929236e-05,
|
| 324 |
+
"loss": 0.7885,
|
| 325 |
+
"step": 225
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.43767840152235965,
|
| 329 |
+
"grad_norm": 0.9839529991149902,
|
| 330 |
+
"learning_rate": 2.9264875977976966e-05,
|
| 331 |
+
"loss": 0.7722,
|
| 332 |
+
"step": 230
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.44719314938154137,
|
| 336 |
+
"grad_norm": 5.82216739654541,
|
| 337 |
+
"learning_rate": 2.9215489684452354e-05,
|
| 338 |
+
"loss": 0.8365,
|
| 339 |
+
"step": 235
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.45670789724072314,
|
| 343 |
+
"grad_norm": 0.9137938618659973,
|
| 344 |
+
"learning_rate": 2.9164542421710325e-05,
|
| 345 |
+
"loss": 0.8072,
|
| 346 |
+
"step": 240
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.46622264509990485,
|
| 350 |
+
"grad_norm": 0.982866108417511,
|
| 351 |
+
"learning_rate": 2.9112039784148993e-05,
|
| 352 |
+
"loss": 0.814,
|
| 353 |
+
"step": 245
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.47573739295908657,
|
| 357 |
+
"grad_norm": 0.996077299118042,
|
| 358 |
+
"learning_rate": 2.905798753695852e-05,
|
| 359 |
+
"loss": 0.7403,
|
| 360 |
+
"step": 250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.48525214081826834,
|
| 364 |
+
"grad_norm": 1.0766255855560303,
|
| 365 |
+
"learning_rate": 2.9002391615488008e-05,
|
| 366 |
+
"loss": 0.7461,
|
| 367 |
+
"step": 255
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.49476688867745006,
|
| 371 |
+
"grad_norm": 1.0079656839370728,
|
| 372 |
+
"learning_rate": 2.894525812459377e-05,
|
| 373 |
+
"loss": 0.7074,
|
| 374 |
+
"step": 260
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.5042816365366318,
|
| 378 |
+
"grad_norm": 0.9577733278274536,
|
| 379 |
+
"learning_rate": 2.888659333796898e-05,
|
| 380 |
+
"loss": 0.7233,
|
| 381 |
+
"step": 265
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.5137963843958135,
|
| 385 |
+
"grad_norm": 1.0857473611831665,
|
| 386 |
+
"learning_rate": 2.882640369745477e-05,
|
| 387 |
+
"loss": 0.6986,
|
| 388 |
+
"step": 270
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.5233111322549953,
|
| 392 |
+
"grad_norm": 1.030419945716858,
|
| 393 |
+
"learning_rate": 2.876469581233285e-05,
|
| 394 |
+
"loss": 0.6686,
|
| 395 |
+
"step": 275
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.532825880114177,
|
| 399 |
+
"grad_norm": 1.0613384246826172,
|
| 400 |
+
"learning_rate": 2.870147645859978e-05,
|
| 401 |
+
"loss": 0.6428,
|
| 402 |
+
"step": 280
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.5423406279733587,
|
| 406 |
+
"grad_norm": 1.1087590456008911,
|
| 407 |
+
"learning_rate": 2.863675257822289e-05,
|
| 408 |
+
"loss": 0.5937,
|
| 409 |
+
"step": 285
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5518553758325404,
|
| 413 |
+
"grad_norm": 1.0359351634979248,
|
| 414 |
+
"learning_rate": 2.8570531278378044e-05,
|
| 415 |
+
"loss": 0.6644,
|
| 416 |
+
"step": 290
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5613701236917221,
|
| 420 |
+
"grad_norm": 1.2866332530975342,
|
| 421 |
+
"learning_rate": 2.8502819830669148e-05,
|
| 422 |
+
"loss": 0.65,
|
| 423 |
+
"step": 295
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.570884871550904,
|
| 427 |
+
"grad_norm": 0.9716510772705078,
|
| 428 |
+
"learning_rate": 2.843362567032974e-05,
|
| 429 |
+
"loss": 0.637,
|
| 430 |
+
"step": 300
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5803996194100857,
|
| 434 |
+
"grad_norm": 1.1773654222488403,
|
| 435 |
+
"learning_rate": 2.8362956395406512e-05,
|
| 436 |
+
"loss": 0.6127,
|
| 437 |
+
"step": 305
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5899143672692674,
|
| 441 |
+
"grad_norm": 1.1006251573562622,
|
| 442 |
+
"learning_rate": 2.8290819765924984e-05,
|
| 443 |
+
"loss": 0.5662,
|
| 444 |
+
"step": 310
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5994291151284491,
|
| 448 |
+
"grad_norm": 1.0656607151031494,
|
| 449 |
+
"learning_rate": 2.8217223703037398e-05,
|
| 450 |
+
"loss": 0.5699,
|
| 451 |
+
"step": 315
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6089438629876308,
|
| 455 |
+
"grad_norm": 1.1515144109725952,
|
| 456 |
+
"learning_rate": 2.8142176288152935e-05,
|
| 457 |
+
"loss": 0.5958,
|
| 458 |
+
"step": 320
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.6184586108468125,
|
| 462 |
+
"grad_norm": 1.0917718410491943,
|
| 463 |
+
"learning_rate": 2.8065685762050274e-05,
|
| 464 |
+
"loss": 0.6488,
|
| 465 |
+
"step": 325
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.6279733587059942,
|
| 469 |
+
"grad_norm": 1.072685718536377,
|
| 470 |
+
"learning_rate": 2.7987760523972735e-05,
|
| 471 |
+
"loss": 0.6258,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.6374881065651761,
|
| 476 |
+
"grad_norm": 1.153988242149353,
|
| 477 |
+
"learning_rate": 2.7908409130705952e-05,
|
| 478 |
+
"loss": 0.5789,
|
| 479 |
+
"step": 335
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6470028544243578,
|
| 483 |
+
"grad_norm": 1.2856593132019043,
|
| 484 |
+
"learning_rate": 2.7827640295638274e-05,
|
| 485 |
+
"loss": 0.5803,
|
| 486 |
+
"step": 340
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.6565176022835395,
|
| 490 |
+
"grad_norm": 1.0754746198654175,
|
| 491 |
+
"learning_rate": 2.7745462887803984e-05,
|
| 492 |
+
"loss": 0.5613,
|
| 493 |
+
"step": 345
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6660323501427212,
|
| 497 |
+
"grad_norm": 1.0461519956588745,
|
| 498 |
+
"learning_rate": 2.7661885930909385e-05,
|
| 499 |
+
"loss": 0.5795,
|
| 500 |
+
"step": 350
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6755470980019029,
|
| 504 |
+
"grad_norm": 1.1279278993606567,
|
| 505 |
+
"learning_rate": 2.7576918602341957e-05,
|
| 506 |
+
"loss": 0.544,
|
| 507 |
+
"step": 355
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6850618458610847,
|
| 511 |
+
"grad_norm": 1.0936821699142456,
|
| 512 |
+
"learning_rate": 2.7490570232162595e-05,
|
| 513 |
+
"loss": 0.5318,
|
| 514 |
+
"step": 360
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6945765937202664,
|
| 518 |
+
"grad_norm": 1.1369750499725342,
|
| 519 |
+
"learning_rate": 2.74028503020811e-05,
|
| 520 |
+
"loss": 0.4475,
|
| 521 |
+
"step": 365
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.7040913415794482,
|
| 525 |
+
"grad_norm": 1.275067925453186,
|
| 526 |
+
"learning_rate": 2.7313768444415004e-05,
|
| 527 |
+
"loss": 0.491,
|
| 528 |
+
"step": 370
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.7136060894386299,
|
| 532 |
+
"grad_norm": 1.1067943572998047,
|
| 533 |
+
"learning_rate": 2.722333444103189e-05,
|
| 534 |
+
"loss": 0.5225,
|
| 535 |
+
"step": 375
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.7231208372978116,
|
| 539 |
+
"grad_norm": 1.056262493133545,
|
| 540 |
+
"learning_rate": 2.7131558222275243e-05,
|
| 541 |
+
"loss": 0.5236,
|
| 542 |
+
"step": 380
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.7326355851569933,
|
| 546 |
+
"grad_norm": 1.149281620979309,
|
| 547 |
+
"learning_rate": 2.7038449865874053e-05,
|
| 548 |
+
"loss": 0.5403,
|
| 549 |
+
"step": 385
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.7421503330161751,
|
| 553 |
+
"grad_norm": 1.227942705154419,
|
| 554 |
+
"learning_rate": 2.694401959583618e-05,
|
| 555 |
+
"loss": 0.543,
|
| 556 |
+
"step": 390
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.7516650808753568,
|
| 560 |
+
"grad_norm": 1.1342272758483887,
|
| 561 |
+
"learning_rate": 2.684827778132567e-05,
|
| 562 |
+
"loss": 0.5135,
|
| 563 |
+
"step": 395
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7611798287345385,
|
| 567 |
+
"grad_norm": 1.2289284467697144,
|
| 568 |
+
"learning_rate": 2.675123493552418e-05,
|
| 569 |
+
"loss": 0.502,
|
| 570 |
+
"step": 400
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.7706945765937203,
|
| 574 |
+
"grad_norm": 1.17427396774292,
|
| 575 |
+
"learning_rate": 2.6652901714476516e-05,
|
| 576 |
+
"loss": 0.4953,
|
| 577 |
+
"step": 405
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.780209324452902,
|
| 581 |
+
"grad_norm": 1.144631028175354,
|
| 582 |
+
"learning_rate": 2.6553288915920533e-05,
|
| 583 |
+
"loss": 0.5017,
|
| 584 |
+
"step": 410
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7897240723120837,
|
| 588 |
+
"grad_norm": 1.3134822845458984,
|
| 589 |
+
"learning_rate": 2.6452407478101472e-05,
|
| 590 |
+
"loss": 0.4931,
|
| 591 |
+
"step": 415
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7992388201712655,
|
| 595 |
+
"grad_norm": 1.1979602575302124,
|
| 596 |
+
"learning_rate": 2.6350268478570813e-05,
|
| 597 |
+
"loss": 0.4383,
|
| 598 |
+
"step": 420
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.8087535680304472,
|
| 602 |
+
"grad_norm": 1.1338562965393066,
|
| 603 |
+
"learning_rate": 2.6246883132969934e-05,
|
| 604 |
+
"loss": 0.5283,
|
| 605 |
+
"step": 425
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.8182683158896289,
|
| 609 |
+
"grad_norm": 1.2282143831253052,
|
| 610 |
+
"learning_rate": 2.6142262793798507e-05,
|
| 611 |
+
"loss": 0.4431,
|
| 612 |
+
"step": 430
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.8277830637488106,
|
| 616 |
+
"grad_norm": 1.1187149286270142,
|
| 617 |
+
"learning_rate": 2.6036418949167922e-05,
|
| 618 |
+
"loss": 0.4622,
|
| 619 |
+
"step": 435
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.8372978116079924,
|
| 623 |
+
"grad_norm": 1.214730143547058,
|
| 624 |
+
"learning_rate": 2.5929363221539802e-05,
|
| 625 |
+
"loss": 0.4658,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.8468125594671742,
|
| 630 |
+
"grad_norm": 1.1169562339782715,
|
| 631 |
+
"learning_rate": 2.582110736644977e-05,
|
| 632 |
+
"loss": 0.4353,
|
| 633 |
+
"step": 445
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8563273073263559,
|
| 637 |
+
"grad_norm": 1.2651045322418213,
|
| 638 |
+
"learning_rate": 2.5711663271216597e-05,
|
| 639 |
+
"loss": 0.4366,
|
| 640 |
+
"step": 450
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8658420551855376,
|
| 644 |
+
"grad_norm": 1.2140588760375977,
|
| 645 |
+
"learning_rate": 2.5601042953636895e-05,
|
| 646 |
+
"loss": 0.4641,
|
| 647 |
+
"step": 455
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.8753568030447193,
|
| 651 |
+
"grad_norm": 1.2653535604476929,
|
| 652 |
+
"learning_rate": 2.5489258560665455e-05,
|
| 653 |
+
"loss": 0.4349,
|
| 654 |
+
"step": 460
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.884871550903901,
|
| 658 |
+
"grad_norm": 1.1089608669281006,
|
| 659 |
+
"learning_rate": 2.5376322367081433e-05,
|
| 660 |
+
"loss": 0.3995,
|
| 661 |
+
"step": 465
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.8943862987630827,
|
| 665 |
+
"grad_norm": 1.2086623907089233,
|
| 666 |
+
"learning_rate": 2.5262246774140466e-05,
|
| 667 |
+
"loss": 0.4266,
|
| 668 |
+
"step": 470
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.9039010466222646,
|
| 672 |
+
"grad_norm": 1.2142243385314941,
|
| 673 |
+
"learning_rate": 2.5147044308212947e-05,
|
| 674 |
+
"loss": 0.4156,
|
| 675 |
+
"step": 475
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.9134157944814463,
|
| 679 |
+
"grad_norm": 1.1891040802001953,
|
| 680 |
+
"learning_rate": 2.503072761940852e-05,
|
| 681 |
+
"loss": 0.4163,
|
| 682 |
+
"step": 480
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.922930542340628,
|
| 686 |
+
"grad_norm": 1.1879245042800903,
|
| 687 |
+
"learning_rate": 2.4913309480186988e-05,
|
| 688 |
+
"loss": 0.3968,
|
| 689 |
+
"step": 485
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.9324452901998097,
|
| 693 |
+
"grad_norm": 1.2805386781692505,
|
| 694 |
+
"learning_rate": 2.4794802783955827e-05,
|
| 695 |
+
"loss": 0.3904,
|
| 696 |
+
"step": 490
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.9419600380589914,
|
| 700 |
+
"grad_norm": 1.1774660348892212,
|
| 701 |
+
"learning_rate": 2.4675220543654374e-05,
|
| 702 |
+
"loss": 0.3922,
|
| 703 |
+
"step": 495
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.9514747859181731,
|
| 707 |
+
"grad_norm": 1.3504858016967773,
|
| 708 |
+
"learning_rate": 2.45545758903249e-05,
|
| 709 |
+
"loss": 0.4221,
|
| 710 |
+
"step": 500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.9609895337773549,
|
| 714 |
+
"grad_norm": 1.2761121988296509,
|
| 715 |
+
"learning_rate": 2.4432882071670746e-05,
|
| 716 |
+
"loss": 0.3763,
|
| 717 |
+
"step": 505
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.9705042816365367,
|
| 721 |
+
"grad_norm": 1.1477121114730835,
|
| 722 |
+
"learning_rate": 2.43101524506016e-05,
|
| 723 |
+
"loss": 0.3743,
|
| 724 |
+
"step": 510
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.9800190294957184,
|
| 728 |
+
"grad_norm": 1.2160308361053467,
|
| 729 |
+
"learning_rate": 2.418640050376616e-05,
|
| 730 |
+
"loss": 0.36,
|
| 731 |
+
"step": 515
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.9895337773549001,
|
| 735 |
+
"grad_norm": 1.1914492845535278,
|
| 736 |
+
"learning_rate": 2.40616398200723e-05,
|
| 737 |
+
"loss": 0.3396,
|
| 738 |
+
"step": 520
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.9990485252140818,
|
| 742 |
+
"grad_norm": 1.1502474546432495,
|
| 743 |
+
"learning_rate": 2.3935884099194875e-05,
|
| 744 |
+
"loss": 0.3661,
|
| 745 |
+
"step": 525
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.0076117982873454,
|
| 749 |
+
"grad_norm": 1.0102150440216064,
|
| 750 |
+
"learning_rate": 2.3809147150071444e-05,
|
| 751 |
+
"loss": 0.2706,
|
| 752 |
+
"step": 530
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.017126546146527,
|
| 756 |
+
"grad_norm": 1.0909314155578613,
|
| 757 |
+
"learning_rate": 2.368144288938589e-05,
|
| 758 |
+
"loss": 0.311,
|
| 759 |
+
"step": 535
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.0266412940057088,
|
| 763 |
+
"grad_norm": 1.3318074941635132,
|
| 764 |
+
"learning_rate": 2.3552785340040294e-05,
|
| 765 |
+
"loss": 0.313,
|
| 766 |
+
"step": 540
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.0361560418648905,
|
| 770 |
+
"grad_norm": 1.1899315118789673,
|
| 771 |
+
"learning_rate": 2.3423188629615107e-05,
|
| 772 |
+
"loss": 0.2696,
|
| 773 |
+
"step": 545
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.0456707897240722,
|
| 777 |
+
"grad_norm": 1.2354599237442017,
|
| 778 |
+
"learning_rate": 2.3292666988817818e-05,
|
| 779 |
+
"loss": 0.287,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.055185537583254,
|
| 784 |
+
"grad_norm": 1.3331857919692993,
|
| 785 |
+
"learning_rate": 2.316123474992035e-05,
|
| 786 |
+
"loss": 0.2917,
|
| 787 |
+
"step": 555
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.0647002854424357,
|
| 791 |
+
"grad_norm": 1.224503993988037,
|
| 792 |
+
"learning_rate": 2.3028906345185224e-05,
|
| 793 |
+
"loss": 0.2739,
|
| 794 |
+
"step": 560
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.0742150333016176,
|
| 798 |
+
"grad_norm": 1.1998045444488525,
|
| 799 |
+
"learning_rate": 2.2895696305280812e-05,
|
| 800 |
+
"loss": 0.2941,
|
| 801 |
+
"step": 565
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.0837297811607993,
|
| 805 |
+
"grad_norm": 1.3093836307525635,
|
| 806 |
+
"learning_rate": 2.2761619257685757e-05,
|
| 807 |
+
"loss": 0.3088,
|
| 808 |
+
"step": 570
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.093244529019981,
|
| 812 |
+
"grad_norm": 1.3205437660217285,
|
| 813 |
+
"learning_rate": 2.2626689925082772e-05,
|
| 814 |
+
"loss": 0.2638,
|
| 815 |
+
"step": 575
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.1027592768791628,
|
| 819 |
+
"grad_norm": 1.1075496673583984,
|
| 820 |
+
"learning_rate": 2.2490923123741937e-05,
|
| 821 |
+
"loss": 0.2666,
|
| 822 |
+
"step": 580
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.1122740247383445,
|
| 826 |
+
"grad_norm": 1.2905559539794922,
|
| 827 |
+
"learning_rate": 2.235433376189381e-05,
|
| 828 |
+
"loss": 0.2395,
|
| 829 |
+
"step": 585
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.1217887725975262,
|
| 833 |
+
"grad_norm": 1.0629035234451294,
|
| 834 |
+
"learning_rate": 2.2216936838092368e-05,
|
| 835 |
+
"loss": 0.2641,
|
| 836 |
+
"step": 590
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.131303520456708,
|
| 840 |
+
"grad_norm": 1.3710782527923584,
|
| 841 |
+
"learning_rate": 2.207874743956804e-05,
|
| 842 |
+
"loss": 0.2833,
|
| 843 |
+
"step": 595
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.1408182683158896,
|
| 847 |
+
"grad_norm": 1.2300183773040771,
|
| 848 |
+
"learning_rate": 2.1939780740571034e-05,
|
| 849 |
+
"loss": 0.2804,
|
| 850 |
+
"step": 600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.1503330161750713,
|
| 854 |
+
"grad_norm": 1.31852126121521,
|
| 855 |
+
"learning_rate": 2.1800052000705084e-05,
|
| 856 |
+
"loss": 0.2603,
|
| 857 |
+
"step": 605
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.159847764034253,
|
| 861 |
+
"grad_norm": 1.1332749128341675,
|
| 862 |
+
"learning_rate": 2.1659576563251807e-05,
|
| 863 |
+
"loss": 0.28,
|
| 864 |
+
"step": 610
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.1693625118934348,
|
| 868 |
+
"grad_norm": 1.3116353750228882,
|
| 869 |
+
"learning_rate": 2.1518369853485935e-05,
|
| 870 |
+
"loss": 0.2505,
|
| 871 |
+
"step": 615
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.1788772597526165,
|
| 875 |
+
"grad_norm": 1.2103403806686401,
|
| 876 |
+
"learning_rate": 2.1376447376981462e-05,
|
| 877 |
+
"loss": 0.2262,
|
| 878 |
+
"step": 620
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.1883920076117982,
|
| 882 |
+
"grad_norm": 1.1109930276870728,
|
| 883 |
+
"learning_rate": 2.123382471790903e-05,
|
| 884 |
+
"loss": 0.2438,
|
| 885 |
+
"step": 625
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.1979067554709801,
|
| 889 |
+
"grad_norm": 1.3755519390106201,
|
| 890 |
+
"learning_rate": 2.109051753732469e-05,
|
| 891 |
+
"loss": 0.2505,
|
| 892 |
+
"step": 630
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.2074215033301616,
|
| 896 |
+
"grad_norm": 1.2475131750106812,
|
| 897 |
+
"learning_rate": 2.0946541571450143e-05,
|
| 898 |
+
"loss": 0.2557,
|
| 899 |
+
"step": 635
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.2169362511893436,
|
| 903 |
+
"grad_norm": 1.2654348611831665,
|
| 904 |
+
"learning_rate": 2.080191262994485e-05,
|
| 905 |
+
"loss": 0.2184,
|
| 906 |
+
"step": 640
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.2264509990485253,
|
| 910 |
+
"grad_norm": 1.23422372341156,
|
| 911 |
+
"learning_rate": 2.0656646594169957e-05,
|
| 912 |
+
"loss": 0.2422,
|
| 913 |
+
"step": 645
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.235965746907707,
|
| 917 |
+
"grad_norm": 1.2899601459503174,
|
| 918 |
+
"learning_rate": 2.0510759415444427e-05,
|
| 919 |
+
"loss": 0.251,
|
| 920 |
+
"step": 650
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.2454804947668887,
|
| 924 |
+
"grad_norm": 1.1003848314285278,
|
| 925 |
+
"learning_rate": 2.036426711329346e-05,
|
| 926 |
+
"loss": 0.2318,
|
| 927 |
+
"step": 655
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.2549952426260704,
|
| 931 |
+
"grad_norm": 1.2367238998413086,
|
| 932 |
+
"learning_rate": 2.0217185773689424e-05,
|
| 933 |
+
"loss": 0.2329,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.2645099904852521,
|
| 938 |
+
"grad_norm": 1.1256778240203857,
|
| 939 |
+
"learning_rate": 2.0069531547285493e-05,
|
| 940 |
+
"loss": 0.2431,
|
| 941 |
+
"step": 665
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.2740247383444339,
|
| 945 |
+
"grad_norm": 1.2910668849945068,
|
| 946 |
+
"learning_rate": 1.9921320647642177e-05,
|
| 947 |
+
"loss": 0.2371,
|
| 948 |
+
"step": 670
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.2835394862036156,
|
| 952 |
+
"grad_norm": 1.2745888233184814,
|
| 953 |
+
"learning_rate": 1.9772569349446962e-05,
|
| 954 |
+
"loss": 0.2635,
|
| 955 |
+
"step": 675
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.2930542340627973,
|
| 959 |
+
"grad_norm": 1.1055139303207397,
|
| 960 |
+
"learning_rate": 1.962329398672721e-05,
|
| 961 |
+
"loss": 0.2363,
|
| 962 |
+
"step": 680
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.302568981921979,
|
| 966 |
+
"grad_norm": 1.0831804275512695,
|
| 967 |
+
"learning_rate": 1.9473510951056575e-05,
|
| 968 |
+
"loss": 0.2278,
|
| 969 |
+
"step": 685
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.3120837297811607,
|
| 973 |
+
"grad_norm": 1.3675286769866943,
|
| 974 |
+
"learning_rate": 1.9323236689755073e-05,
|
| 975 |
+
"loss": 0.2341,
|
| 976 |
+
"step": 690
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.3215984776403427,
|
| 980 |
+
"grad_norm": 1.136629343032837,
|
| 981 |
+
"learning_rate": 1.9172487704083033e-05,
|
| 982 |
+
"loss": 0.2235,
|
| 983 |
+
"step": 695
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.3311132254995242,
|
| 987 |
+
"grad_norm": 1.2047828435897827,
|
| 988 |
+
"learning_rate": 1.9021280547429146e-05,
|
| 989 |
+
"loss": 0.2432,
|
| 990 |
+
"step": 700
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.340627973358706,
|
| 994 |
+
"grad_norm": 1.0443748235702515,
|
| 995 |
+
"learning_rate": 1.8869631823492768e-05,
|
| 996 |
+
"loss": 0.1921,
|
| 997 |
+
"step": 705
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.3501427212178878,
|
| 1001 |
+
"grad_norm": 1.1521350145339966,
|
| 1002 |
+
"learning_rate": 1.871755818446073e-05,
|
| 1003 |
+
"loss": 0.2177,
|
| 1004 |
+
"step": 710
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.3596574690770695,
|
| 1008 |
+
"grad_norm": 1.0329209566116333,
|
| 1009 |
+
"learning_rate": 1.8565076329178762e-05,
|
| 1010 |
+
"loss": 0.2069,
|
| 1011 |
+
"step": 715
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.3691722169362512,
|
| 1015 |
+
"grad_norm": 1.1129894256591797,
|
| 1016 |
+
"learning_rate": 1.8412203001317856e-05,
|
| 1017 |
+
"loss": 0.1888,
|
| 1018 |
+
"step": 720
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.378686964795433,
|
| 1022 |
+
"grad_norm": 1.0250636339187622,
|
| 1023 |
+
"learning_rate": 1.8258954987535686e-05,
|
| 1024 |
+
"loss": 0.1804,
|
| 1025 |
+
"step": 725
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.3882017126546147,
|
| 1029 |
+
"grad_norm": 1.140295386314392,
|
| 1030 |
+
"learning_rate": 1.8105349115633297e-05,
|
| 1031 |
+
"loss": 0.2153,
|
| 1032 |
+
"step": 730
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.3977164605137964,
|
| 1036 |
+
"grad_norm": 1.02838933467865,
|
| 1037 |
+
"learning_rate": 1.7951402252707277e-05,
|
| 1038 |
+
"loss": 0.1775,
|
| 1039 |
+
"step": 735
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.407231208372978,
|
| 1043 |
+
"grad_norm": 1.0924123525619507,
|
| 1044 |
+
"learning_rate": 1.7797131303297644e-05,
|
| 1045 |
+
"loss": 0.2179,
|
| 1046 |
+
"step": 740
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.4167459562321598,
|
| 1050 |
+
"grad_norm": 1.0600032806396484,
|
| 1051 |
+
"learning_rate": 1.7642553207531577e-05,
|
| 1052 |
+
"loss": 0.1951,
|
| 1053 |
+
"step": 745
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.4262607040913415,
|
| 1057 |
+
"grad_norm": 1.3156516551971436,
|
| 1058 |
+
"learning_rate": 1.7487684939263266e-05,
|
| 1059 |
+
"loss": 0.1858,
|
| 1060 |
+
"step": 750
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.4357754519505233,
|
| 1064 |
+
"grad_norm": 1.1005526781082153,
|
| 1065 |
+
"learning_rate": 1.7332543504210058e-05,
|
| 1066 |
+
"loss": 0.1877,
|
| 1067 |
+
"step": 755
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.445290199809705,
|
| 1071 |
+
"grad_norm": 1.0171141624450684,
|
| 1072 |
+
"learning_rate": 1.7177145938085094e-05,
|
| 1073 |
+
"loss": 0.1857,
|
| 1074 |
+
"step": 760
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.4548049476688867,
|
| 1078 |
+
"grad_norm": 1.158037781715393,
|
| 1079 |
+
"learning_rate": 1.702150930472667e-05,
|
| 1080 |
+
"loss": 0.2049,
|
| 1081 |
+
"step": 765
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.4643196955280686,
|
| 1085 |
+
"grad_norm": 1.1181069612503052,
|
| 1086 |
+
"learning_rate": 1.686565069422448e-05,
|
| 1087 |
+
"loss": 0.1755,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.4738344433872501,
|
| 1092 |
+
"grad_norm": 1.154498815536499,
|
| 1093 |
+
"learning_rate": 1.6709587221043e-05,
|
| 1094 |
+
"loss": 0.1654,
|
| 1095 |
+
"step": 775
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.483349191246432,
|
| 1099 |
+
"grad_norm": 1.1790727376937866,
|
| 1100 |
+
"learning_rate": 1.65533360221422e-05,
|
| 1101 |
+
"loss": 0.1787,
|
| 1102 |
+
"step": 780
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.4928639391056138,
|
| 1106 |
+
"grad_norm": 1.0942307710647583,
|
| 1107 |
+
"learning_rate": 1.6396914255095753e-05,
|
| 1108 |
+
"loss": 0.1995,
|
| 1109 |
+
"step": 785
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.5023786869647955,
|
| 1113 |
+
"grad_norm": 0.9170376062393188,
|
| 1114 |
+
"learning_rate": 1.624033909620703e-05,
|
| 1115 |
+
"loss": 0.1886,
|
| 1116 |
+
"step": 790
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.5118934348239772,
|
| 1120 |
+
"grad_norm": 1.0581458806991577,
|
| 1121 |
+
"learning_rate": 1.6083627738622986e-05,
|
| 1122 |
+
"loss": 0.1816,
|
| 1123 |
+
"step": 795
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.521408182683159,
|
| 1127 |
+
"grad_norm": 0.8717968463897705,
|
| 1128 |
+
"learning_rate": 1.5926797390446244e-05,
|
| 1129 |
+
"loss": 0.162,
|
| 1130 |
+
"step": 800
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.5309229305423406,
|
| 1134 |
+
"grad_norm": 1.1979910135269165,
|
| 1135 |
+
"learning_rate": 1.5769865272845484e-05,
|
| 1136 |
+
"loss": 0.1624,
|
| 1137 |
+
"step": 805
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.5404376784015223,
|
| 1141 |
+
"grad_norm": 1.0542930364608765,
|
| 1142 |
+
"learning_rate": 1.561284861816446e-05,
|
| 1143 |
+
"loss": 0.1718,
|
| 1144 |
+
"step": 810
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.549952426260704,
|
| 1148 |
+
"grad_norm": 1.068016529083252,
|
| 1149 |
+
"learning_rate": 1.5455764668029732e-05,
|
| 1150 |
+
"loss": 0.1829,
|
| 1151 |
+
"step": 815
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.5594671741198858,
|
| 1155 |
+
"grad_norm": 1.2204563617706299,
|
| 1156 |
+
"learning_rate": 1.5298630671457422e-05,
|
| 1157 |
+
"loss": 0.1741,
|
| 1158 |
+
"step": 820
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.5689819219790677,
|
| 1162 |
+
"grad_norm": 1.1641712188720703,
|
| 1163 |
+
"learning_rate": 1.5141463882959134e-05,
|
| 1164 |
+
"loss": 0.1568,
|
| 1165 |
+
"step": 825
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.5784966698382492,
|
| 1169 |
+
"grad_norm": 0.9883166551589966,
|
| 1170 |
+
"learning_rate": 1.4984281560647267e-05,
|
| 1171 |
+
"loss": 0.1593,
|
| 1172 |
+
"step": 830
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.5880114176974311,
|
| 1176 |
+
"grad_norm": 1.2379170656204224,
|
| 1177 |
+
"learning_rate": 1.4827100964339957e-05,
|
| 1178 |
+
"loss": 0.1735,
|
| 1179 |
+
"step": 835
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.5975261655566126,
|
| 1183 |
+
"grad_norm": 1.2986499071121216,
|
| 1184 |
+
"learning_rate": 1.4669939353665808e-05,
|
| 1185 |
+
"loss": 0.1663,
|
| 1186 |
+
"step": 840
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.6070409134157946,
|
| 1190 |
+
"grad_norm": 0.9190987944602966,
|
| 1191 |
+
"learning_rate": 1.4512813986168666e-05,
|
| 1192 |
+
"loss": 0.1624,
|
| 1193 |
+
"step": 845
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.616555661274976,
|
| 1197 |
+
"grad_norm": 1.029982089996338,
|
| 1198 |
+
"learning_rate": 1.4355742115412583e-05,
|
| 1199 |
+
"loss": 0.1497,
|
| 1200 |
+
"step": 850
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.626070409134158,
|
| 1204 |
+
"grad_norm": 0.9184136986732483,
|
| 1205 |
+
"learning_rate": 1.4198740989087277e-05,
|
| 1206 |
+
"loss": 0.1546,
|
| 1207 |
+
"step": 855
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.6355851569933397,
|
| 1211 |
+
"grad_norm": 0.991702139377594,
|
| 1212 |
+
"learning_rate": 1.4041827847114183e-05,
|
| 1213 |
+
"loss": 0.1532,
|
| 1214 |
+
"step": 860
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.6450999048525214,
|
| 1218 |
+
"grad_norm": 1.0909984111785889,
|
| 1219 |
+
"learning_rate": 1.3885019919753377e-05,
|
| 1220 |
+
"loss": 0.1406,
|
| 1221 |
+
"step": 865
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.6546146527117032,
|
| 1225 |
+
"grad_norm": 1.0758049488067627,
|
| 1226 |
+
"learning_rate": 1.3728334425711571e-05,
|
| 1227 |
+
"loss": 0.1467,
|
| 1228 |
+
"step": 870
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.6641294005708849,
|
| 1232 |
+
"grad_norm": 0.9771386981010437,
|
| 1233 |
+
"learning_rate": 1.3571788570251364e-05,
|
| 1234 |
+
"loss": 0.1696,
|
| 1235 |
+
"step": 875
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.6736441484300666,
|
| 1239 |
+
"grad_norm": 1.1889567375183105,
|
| 1240 |
+
"learning_rate": 1.3415399543301983e-05,
|
| 1241 |
+
"loss": 0.1351,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.6831588962892483,
|
| 1246 |
+
"grad_norm": 0.9384522438049316,
|
| 1247 |
+
"learning_rate": 1.3259184517571674e-05,
|
| 1248 |
+
"loss": 0.1431,
|
| 1249 |
+
"step": 885
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.69267364414843,
|
| 1253 |
+
"grad_norm": 0.9385328888893127,
|
| 1254 |
+
"learning_rate": 1.3103160646662029e-05,
|
| 1255 |
+
"loss": 0.1362,
|
| 1256 |
+
"step": 890
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.7021883920076117,
|
| 1260 |
+
"grad_norm": 1.0427623987197876,
|
| 1261 |
+
"learning_rate": 1.294734506318438e-05,
|
| 1262 |
+
"loss": 0.1361,
|
| 1263 |
+
"step": 895
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.7117031398667937,
|
| 1267 |
+
"grad_norm": 0.8941156268119812,
|
| 1268 |
+
"learning_rate": 1.2791754876878513e-05,
|
| 1269 |
+
"loss": 0.1303,
|
| 1270 |
+
"step": 900
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.7212178877259752,
|
| 1274 |
+
"grad_norm": 1.1560466289520264,
|
| 1275 |
+
"learning_rate": 1.263640717273388e-05,
|
| 1276 |
+
"loss": 0.1205,
|
| 1277 |
+
"step": 905
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.730732635585157,
|
| 1281 |
+
"grad_norm": 1.1242207288742065,
|
| 1282 |
+
"learning_rate": 1.2481319009113548e-05,
|
| 1283 |
+
"loss": 0.1376,
|
| 1284 |
+
"step": 910
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.7402473834443386,
|
| 1288 |
+
"grad_norm": 0.8830758333206177,
|
| 1289 |
+
"learning_rate": 1.2326507415881051e-05,
|
| 1290 |
+
"loss": 0.1334,
|
| 1291 |
+
"step": 915
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.7497621313035205,
|
| 1295 |
+
"grad_norm": 1.1340227127075195,
|
| 1296 |
+
"learning_rate": 1.2171989392530383e-05,
|
| 1297 |
+
"loss": 0.1571,
|
| 1298 |
+
"step": 920
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.759276879162702,
|
| 1302 |
+
"grad_norm": 1.0231640338897705,
|
| 1303 |
+
"learning_rate": 1.2017781906319326e-05,
|
| 1304 |
+
"loss": 0.1399,
|
| 1305 |
+
"step": 925
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.768791627021884,
|
| 1309 |
+
"grad_norm": 0.9151033163070679,
|
| 1310 |
+
"learning_rate": 1.1863901890406316e-05,
|
| 1311 |
+
"loss": 0.1396,
|
| 1312 |
+
"step": 930
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.7783063748810657,
|
| 1316 |
+
"grad_norm": 1.1734161376953125,
|
| 1317 |
+
"learning_rate": 1.1710366241991054e-05,
|
| 1318 |
+
"loss": 0.1216,
|
| 1319 |
+
"step": 935
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.7878211227402474,
|
| 1323 |
+
"grad_norm": 1.0312820672988892,
|
| 1324 |
+
"learning_rate": 1.1557191820459061e-05,
|
| 1325 |
+
"loss": 0.1367,
|
| 1326 |
+
"step": 940
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.7973358705994291,
|
| 1330 |
+
"grad_norm": 0.9692572355270386,
|
| 1331 |
+
"learning_rate": 1.140439544553039e-05,
|
| 1332 |
+
"loss": 0.1307,
|
| 1333 |
+
"step": 945
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.8068506184586108,
|
| 1337 |
+
"grad_norm": 0.9553561806678772,
|
| 1338 |
+
"learning_rate": 1.1251993895412692e-05,
|
| 1339 |
+
"loss": 0.12,
|
| 1340 |
+
"step": 950
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.8163653663177926,
|
| 1344 |
+
"grad_norm": 1.0238513946533203,
|
| 1345 |
+
"learning_rate": 1.1100003904958839e-05,
|
| 1346 |
+
"loss": 0.1241,
|
| 1347 |
+
"step": 955
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.8258801141769743,
|
| 1351 |
+
"grad_norm": 0.9677141308784485,
|
| 1352 |
+
"learning_rate": 1.0948442163829318e-05,
|
| 1353 |
+
"loss": 0.1254,
|
| 1354 |
+
"step": 960
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.8353948620361562,
|
| 1358 |
+
"grad_norm": 0.9944155812263489,
|
| 1359 |
+
"learning_rate": 1.0797325314659571e-05,
|
| 1360 |
+
"loss": 0.1315,
|
| 1361 |
+
"step": 965
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.8449096098953377,
|
| 1365 |
+
"grad_norm": 1.1988720893859863,
|
| 1366 |
+
"learning_rate": 1.064666995123251e-05,
|
| 1367 |
+
"loss": 0.1245,
|
| 1368 |
+
"step": 970
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.8544243577545196,
|
| 1372 |
+
"grad_norm": 0.944276750087738,
|
| 1373 |
+
"learning_rate": 1.049649261665639e-05,
|
| 1374 |
+
"loss": 0.1108,
|
| 1375 |
+
"step": 975
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.8639391056137011,
|
| 1379 |
+
"grad_norm": 0.9537860155105591,
|
| 1380 |
+
"learning_rate": 1.0346809801548247e-05,
|
| 1381 |
+
"loss": 0.1182,
|
| 1382 |
+
"step": 980
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.873453853472883,
|
| 1386 |
+
"grad_norm": 0.9839835166931152,
|
| 1387 |
+
"learning_rate": 1.019763794222311e-05,
|
| 1388 |
+
"loss": 0.1205,
|
| 1389 |
+
"step": 985
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.8829686013320646,
|
| 1393 |
+
"grad_norm": 0.898876965045929,
|
| 1394 |
+
"learning_rate": 1.0048993418889176e-05,
|
| 1395 |
+
"loss": 0.1096,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.8924833491912465,
|
| 1400 |
+
"grad_norm": 1.068875789642334,
|
| 1401 |
+
"learning_rate": 9.90089255384911e-06,
|
| 1402 |
+
"loss": 0.1067,
|
| 1403 |
+
"step": 995
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.9019980970504282,
|
| 1407 |
+
"grad_norm": 1.0773367881774902,
|
| 1408 |
+
"learning_rate": 9.753351609707754e-06,
|
| 1409 |
+
"loss": 0.1028,
|
| 1410 |
+
"step": 1000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.91151284490961,
|
| 1414 |
+
"grad_norm": 0.9173557162284851,
|
| 1415 |
+
"learning_rate": 9.606386787586371e-06,
|
| 1416 |
+
"loss": 0.0979,
|
| 1417 |
+
"step": 1005
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.9210275927687916,
|
| 1421 |
+
"grad_norm": 0.9160549640655518,
|
| 1422 |
+
"learning_rate": 9.460014225343625e-06,
|
| 1423 |
+
"loss": 0.1114,
|
| 1424 |
+
"step": 1010
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.9305423406279734,
|
| 1428 |
+
"grad_norm": 1.0966682434082031,
|
| 1429 |
+
"learning_rate": 9.314249995803519e-06,
|
| 1430 |
+
"loss": 0.1125,
|
| 1431 |
+
"step": 1015
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.940057088487155,
|
| 1435 |
+
"grad_norm": 1.09756338596344,
|
| 1436 |
+
"learning_rate": 9.169110104990505e-06,
|
| 1437 |
+
"loss": 0.1188,
|
| 1438 |
+
"step": 1020
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.9495718363463368,
|
| 1442 |
+
"grad_norm": 0.9152058959007263,
|
| 1443 |
+
"learning_rate": 9.024610490371862e-06,
|
| 1444 |
+
"loss": 0.1115,
|
| 1445 |
+
"step": 1025
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.9590865842055185,
|
| 1449 |
+
"grad_norm": 1.0392881631851196,
|
| 1450 |
+
"learning_rate": 8.88076701910766e-06,
|
| 1451 |
+
"loss": 0.1181,
|
| 1452 |
+
"step": 1030
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.9686013320647002,
|
| 1456 |
+
"grad_norm": 0.9156733751296997,
|
| 1457 |
+
"learning_rate": 8.737595486308433e-06,
|
| 1458 |
+
"loss": 0.1027,
|
| 1459 |
+
"step": 1035
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.9781160799238822,
|
| 1463 |
+
"grad_norm": 0.927276074886322,
|
| 1464 |
+
"learning_rate": 8.595111613300735e-06,
|
| 1465 |
+
"loss": 0.1071,
|
| 1466 |
+
"step": 1040
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.9876308277830637,
|
| 1470 |
+
"grad_norm": 0.7791060209274292,
|
| 1471 |
+
"learning_rate": 8.453331045900835e-06,
|
| 1472 |
+
"loss": 0.1005,
|
| 1473 |
+
"step": 1045
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.9971455756422456,
|
| 1477 |
+
"grad_norm": 0.8834468722343445,
|
| 1478 |
+
"learning_rate": 8.312269352696674e-06,
|
| 1479 |
+
"loss": 0.1028,
|
| 1480 |
+
"step": 1050
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 2.0057088487155093,
|
| 1484 |
+
"grad_norm": 0.7079776525497437,
|
| 1485 |
+
"learning_rate": 8.171942023338328e-06,
|
| 1486 |
+
"loss": 0.0754,
|
| 1487 |
+
"step": 1055
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 2.0152235965746907,
|
| 1491 |
+
"grad_norm": 0.7594757080078125,
|
| 1492 |
+
"learning_rate": 8.032364466837124e-06,
|
| 1493 |
+
"loss": 0.0897,
|
| 1494 |
+
"step": 1060
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 2.0247383444338727,
|
| 1498 |
+
"grad_norm": 0.7908365726470947,
|
| 1499 |
+
"learning_rate": 7.893552009873601e-06,
|
| 1500 |
+
"loss": 0.0908,
|
| 1501 |
+
"step": 1065
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 2.034253092293054,
|
| 1505 |
+
"grad_norm": 0.757188618183136,
|
| 1506 |
+
"learning_rate": 7.755519895114533e-06,
|
| 1507 |
+
"loss": 0.0753,
|
| 1508 |
+
"step": 1070
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 2.043767840152236,
|
| 1512 |
+
"grad_norm": 0.641106903553009,
|
| 1513 |
+
"learning_rate": 7.61828327953917e-06,
|
| 1514 |
+
"loss": 0.0705,
|
| 1515 |
+
"step": 1075
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 2.0532825880114176,
|
| 1519 |
+
"grad_norm": 0.7395414710044861,
|
| 1520 |
+
"learning_rate": 7.481857232774882e-06,
|
| 1521 |
+
"loss": 0.0857,
|
| 1522 |
+
"step": 1080
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 2.0627973358705995,
|
| 1526 |
+
"grad_norm": 0.9435325264930725,
|
| 1527 |
+
"learning_rate": 7.346256735442396e-06,
|
| 1528 |
+
"loss": 0.0822,
|
| 1529 |
+
"step": 1085
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 2.072312083729781,
|
| 1533 |
+
"grad_norm": 0.8866448998451233,
|
| 1534 |
+
"learning_rate": 7.21149667751082e-06,
|
| 1535 |
+
"loss": 0.0779,
|
| 1536 |
+
"step": 1090
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 2.081826831588963,
|
| 1540 |
+
"grad_norm": 0.6670547127723694,
|
| 1541 |
+
"learning_rate": 7.077591856662598e-06,
|
| 1542 |
+
"loss": 0.0808,
|
| 1543 |
+
"step": 1095
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 2.0913415794481445,
|
| 1547 |
+
"grad_norm": 0.9339617490768433,
|
| 1548 |
+
"learning_rate": 6.944556976668619e-06,
|
| 1549 |
+
"loss": 0.0766,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 2.1008563273073264,
|
| 1554 |
+
"grad_norm": 0.7617207169532776,
|
| 1555 |
+
"learning_rate": 6.812406645773632e-06,
|
| 1556 |
+
"loss": 0.0847,
|
| 1557 |
+
"step": 1105
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 2.110371075166508,
|
| 1561 |
+
"grad_norm": 0.727130115032196,
|
| 1562 |
+
"learning_rate": 6.6811553750921395e-06,
|
| 1563 |
+
"loss": 0.0715,
|
| 1564 |
+
"step": 1110
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 2.11988582302569,
|
| 1568 |
+
"grad_norm": 0.9042198061943054,
|
| 1569 |
+
"learning_rate": 6.5508175770149905e-06,
|
| 1570 |
+
"loss": 0.0782,
|
| 1571 |
+
"step": 1115
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 2.1294005708848713,
|
| 1575 |
+
"grad_norm": 0.7367148995399475,
|
| 1576 |
+
"learning_rate": 6.421407563626753e-06,
|
| 1577 |
+
"loss": 0.078,
|
| 1578 |
+
"step": 1120
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 2.1389153187440533,
|
| 1582 |
+
"grad_norm": 0.7439459562301636,
|
| 1583 |
+
"learning_rate": 6.292939545134172e-06,
|
| 1584 |
+
"loss": 0.0788,
|
| 1585 |
+
"step": 1125
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 2.148430066603235,
|
| 1589 |
+
"grad_norm": 0.6174789071083069,
|
| 1590 |
+
"learning_rate": 6.165427628305758e-06,
|
| 1591 |
+
"loss": 0.0701,
|
| 1592 |
+
"step": 1130
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.1579448144624167,
|
| 1596 |
+
"grad_norm": 0.926468014717102,
|
| 1597 |
+
"learning_rate": 6.038885814922769e-06,
|
| 1598 |
+
"loss": 0.0923,
|
| 1599 |
+
"step": 1135
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 2.1674595623215986,
|
| 1603 |
+
"grad_norm": 0.8429945111274719,
|
| 1604 |
+
"learning_rate": 5.913328000241704e-06,
|
| 1605 |
+
"loss": 0.0818,
|
| 1606 |
+
"step": 1140
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 2.17697431018078,
|
| 1610 |
+
"grad_norm": 0.6744865775108337,
|
| 1611 |
+
"learning_rate": 5.788767971468493e-06,
|
| 1612 |
+
"loss": 0.0796,
|
| 1613 |
+
"step": 1145
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 2.186489058039962,
|
| 1617 |
+
"grad_norm": 0.5986104011535645,
|
| 1618 |
+
"learning_rate": 5.665219406244569e-06,
|
| 1619 |
+
"loss": 0.0623,
|
| 1620 |
+
"step": 1150
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 2.1960038058991436,
|
| 1624 |
+
"grad_norm": 0.7097954154014587,
|
| 1625 |
+
"learning_rate": 5.542695871144943e-06,
|
| 1626 |
+
"loss": 0.073,
|
| 1627 |
+
"step": 1155
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 2.2055185537583255,
|
| 1631 |
+
"grad_norm": 0.7579834461212158,
|
| 1632 |
+
"learning_rate": 5.421210820188503e-06,
|
| 1633 |
+
"loss": 0.0743,
|
| 1634 |
+
"step": 1160
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.215033301617507,
|
| 1638 |
+
"grad_norm": 0.6872014999389648,
|
| 1639 |
+
"learning_rate": 5.300777593360649e-06,
|
| 1640 |
+
"loss": 0.0766,
|
| 1641 |
+
"step": 1165
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 2.224548049476689,
|
| 1645 |
+
"grad_norm": 0.8187681436538696,
|
| 1646 |
+
"learning_rate": 5.181409415148482e-06,
|
| 1647 |
+
"loss": 0.0727,
|
| 1648 |
+
"step": 1170
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 2.2340627973358704,
|
| 1652 |
+
"grad_norm": 0.8018311262130737,
|
| 1653 |
+
"learning_rate": 5.0631193930886295e-06,
|
| 1654 |
+
"loss": 0.078,
|
| 1655 |
+
"step": 1175
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 2.2435775451950524,
|
| 1659 |
+
"grad_norm": 0.6779195666313171,
|
| 1660 |
+
"learning_rate": 4.945920516327942e-06,
|
| 1661 |
+
"loss": 0.0719,
|
| 1662 |
+
"step": 1180
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 2.253092293054234,
|
| 1666 |
+
"grad_norm": 0.6903189420700073,
|
| 1667 |
+
"learning_rate": 4.829825654197206e-06,
|
| 1668 |
+
"loss": 0.075,
|
| 1669 |
+
"step": 1185
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 2.262607040913416,
|
| 1673 |
+
"grad_norm": 0.7453174591064453,
|
| 1674 |
+
"learning_rate": 4.7148475547979665e-06,
|
| 1675 |
+
"loss": 0.0766,
|
| 1676 |
+
"step": 1190
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.2721217887725977,
|
| 1680 |
+
"grad_norm": 0.8147085309028625,
|
| 1681 |
+
"learning_rate": 4.60099884360271e-06,
|
| 1682 |
+
"loss": 0.0732,
|
| 1683 |
+
"step": 1195
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.2816365366317792,
|
| 1687 |
+
"grad_norm": 0.7342116236686707,
|
| 1688 |
+
"learning_rate": 4.488292022068475e-06,
|
| 1689 |
+
"loss": 0.0798,
|
| 1690 |
+
"step": 1200
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 2.291151284490961,
|
| 1694 |
+
"grad_norm": 0.712863564491272,
|
| 1695 |
+
"learning_rate": 4.3767394662641106e-06,
|
| 1696 |
+
"loss": 0.0712,
|
| 1697 |
+
"step": 1205
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.3006660323501427,
|
| 1701 |
+
"grad_norm": 0.7741625905036926,
|
| 1702 |
+
"learning_rate": 4.266353425511277e-06,
|
| 1703 |
+
"loss": 0.0702,
|
| 1704 |
+
"step": 1210
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.3101807802093246,
|
| 1708 |
+
"grad_norm": 0.6669841408729553,
|
| 1709 |
+
"learning_rate": 4.157146021039383e-06,
|
| 1710 |
+
"loss": 0.0623,
|
| 1711 |
+
"step": 1215
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.319695528068506,
|
| 1715 |
+
"grad_norm": 0.7403586506843567,
|
| 1716 |
+
"learning_rate": 4.049129244654584e-06,
|
| 1717 |
+
"loss": 0.0696,
|
| 1718 |
+
"step": 1220
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.329210275927688,
|
| 1722 |
+
"grad_norm": 0.7681209444999695,
|
| 1723 |
+
"learning_rate": 3.942314957422994e-06,
|
| 1724 |
+
"loss": 0.0685,
|
| 1725 |
+
"step": 1225
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.3387250237868695,
|
| 1729 |
+
"grad_norm": 0.711617648601532,
|
| 1730 |
+
"learning_rate": 3.836714888368234e-06,
|
| 1731 |
+
"loss": 0.0776,
|
| 1732 |
+
"step": 1230
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.3482397716460515,
|
| 1736 |
+
"grad_norm": 0.6631252765655518,
|
| 1737 |
+
"learning_rate": 3.732340633183507e-06,
|
| 1738 |
+
"loss": 0.0669,
|
| 1739 |
+
"step": 1235
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.357754519505233,
|
| 1743 |
+
"grad_norm": 0.519625723361969,
|
| 1744 |
+
"learning_rate": 3.629203652958309e-06,
|
| 1745 |
+
"loss": 0.0751,
|
| 1746 |
+
"step": 1240
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 2.367269267364415,
|
| 1750 |
+
"grad_norm": 0.6091018915176392,
|
| 1751 |
+
"learning_rate": 3.5273152729198973e-06,
|
| 1752 |
+
"loss": 0.0692,
|
| 1753 |
+
"step": 1245
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 2.3767840152235964,
|
| 1757 |
+
"grad_norm": 0.5963409543037415,
|
| 1758 |
+
"learning_rate": 3.426686681189711e-06,
|
| 1759 |
+
"loss": 0.0659,
|
| 1760 |
+
"step": 1250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.3862987630827783,
|
| 1764 |
+
"grad_norm": 0.7319835424423218,
|
| 1765 |
+
"learning_rate": 3.3273289275548154e-06,
|
| 1766 |
+
"loss": 0.0685,
|
| 1767 |
+
"step": 1255
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 2.3958135109419603,
|
| 1771 |
+
"grad_norm": 0.5643691420555115,
|
| 1772 |
+
"learning_rate": 3.2292529222545608e-06,
|
| 1773 |
+
"loss": 0.0588,
|
| 1774 |
+
"step": 1260
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.4053282588011418,
|
| 1778 |
+
"grad_norm": 0.7442695498466492,
|
| 1779 |
+
"learning_rate": 3.132469434782553e-06,
|
| 1780 |
+
"loss": 0.0725,
|
| 1781 |
+
"step": 1265
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.4148430066603233,
|
| 1785 |
+
"grad_norm": 0.6718599796295166,
|
| 1786 |
+
"learning_rate": 3.036989092704079e-06,
|
| 1787 |
+
"loss": 0.0704,
|
| 1788 |
+
"step": 1270
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.424357754519505,
|
| 1792 |
+
"grad_norm": 0.7288857698440552,
|
| 1793 |
+
"learning_rate": 2.9428223804891114e-06,
|
| 1794 |
+
"loss": 0.0691,
|
| 1795 |
+
"step": 1275
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.433872502378687,
|
| 1799 |
+
"grad_norm": 0.6483878493309021,
|
| 1800 |
+
"learning_rate": 2.8499796383610625e-06,
|
| 1801 |
+
"loss": 0.0782,
|
| 1802 |
+
"step": 1280
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.4433872502378686,
|
| 1806 |
+
"grad_norm": 0.5524934530258179,
|
| 1807 |
+
"learning_rate": 2.758471061161309e-06,
|
| 1808 |
+
"loss": 0.0614,
|
| 1809 |
+
"step": 1285
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.4529019980970506,
|
| 1813 |
+
"grad_norm": 0.6097465753555298,
|
| 1814 |
+
"learning_rate": 2.6683066972297488e-06,
|
| 1815 |
+
"loss": 0.0719,
|
| 1816 |
+
"step": 1290
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.462416745956232,
|
| 1820 |
+
"grad_norm": 0.8196659088134766,
|
| 1821 |
+
"learning_rate": 2.579496447301404e-06,
|
| 1822 |
+
"loss": 0.0587,
|
| 1823 |
+
"step": 1295
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.471931493815414,
|
| 1827 |
+
"grad_norm": 0.6070199012756348,
|
| 1828 |
+
"learning_rate": 2.492050063419252e-06,
|
| 1829 |
+
"loss": 0.066,
|
| 1830 |
+
"step": 1300
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.4814462416745955,
|
| 1834 |
+
"grad_norm": 0.6622481942176819,
|
| 1835 |
+
"learning_rate": 2.405977147863372e-06,
|
| 1836 |
+
"loss": 0.0623,
|
| 1837 |
+
"step": 1305
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.4909609895337774,
|
| 1841 |
+
"grad_norm": 0.6231311559677124,
|
| 1842 |
+
"learning_rate": 2.3212871520965432e-06,
|
| 1843 |
+
"loss": 0.07,
|
| 1844 |
+
"step": 1310
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.500475737392959,
|
| 1848 |
+
"grad_norm": 0.7100708484649658,
|
| 1849 |
+
"learning_rate": 2.2379893757264007e-06,
|
| 1850 |
+
"loss": 0.0662,
|
| 1851 |
+
"step": 1315
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.509990485252141,
|
| 1855 |
+
"grad_norm": 0.631592333316803,
|
| 1856 |
+
"learning_rate": 2.1560929654842697e-06,
|
| 1857 |
+
"loss": 0.0606,
|
| 1858 |
+
"step": 1320
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.519505233111323,
|
| 1862 |
+
"grad_norm": 0.6099866032600403,
|
| 1863 |
+
"learning_rate": 2.0756069142207846e-06,
|
| 1864 |
+
"loss": 0.0649,
|
| 1865 |
+
"step": 1325
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.5290199809705043,
|
| 1869 |
+
"grad_norm": 0.5789472460746765,
|
| 1870 |
+
"learning_rate": 1.996540059918396e-06,
|
| 1871 |
+
"loss": 0.0774,
|
| 1872 |
+
"step": 1330
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.538534728829686,
|
| 1876 |
+
"grad_norm": 0.5259339809417725,
|
| 1877 |
+
"learning_rate": 1.918901084720916e-06,
|
| 1878 |
+
"loss": 0.0641,
|
| 1879 |
+
"step": 1335
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.5480494766888677,
|
| 1883 |
+
"grad_norm": 0.7160536646842957,
|
| 1884 |
+
"learning_rate": 1.8426985139801278e-06,
|
| 1885 |
+
"loss": 0.0618,
|
| 1886 |
+
"step": 1340
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.5575642245480497,
|
| 1890 |
+
"grad_norm": 0.5439906120300293,
|
| 1891 |
+
"learning_rate": 1.767940715319647e-06,
|
| 1892 |
+
"loss": 0.0508,
|
| 1893 |
+
"step": 1345
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.567078972407231,
|
| 1897 |
+
"grad_norm": 0.6428115963935852,
|
| 1898 |
+
"learning_rate": 1.6946358977160936e-06,
|
| 1899 |
+
"loss": 0.067,
|
| 1900 |
+
"step": 1350
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.576593720266413,
|
| 1904 |
+
"grad_norm": 0.49097028374671936,
|
| 1905 |
+
"learning_rate": 1.622792110597685e-06,
|
| 1906 |
+
"loss": 0.0561,
|
| 1907 |
+
"step": 1355
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.5861084681255946,
|
| 1911 |
+
"grad_norm": 0.6506499648094177,
|
| 1912 |
+
"learning_rate": 1.5524172429603444e-06,
|
| 1913 |
+
"loss": 0.0702,
|
| 1914 |
+
"step": 1360
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.5956232159847765,
|
| 1918 |
+
"grad_norm": 0.5794188380241394,
|
| 1919 |
+
"learning_rate": 1.4835190225014255e-06,
|
| 1920 |
+
"loss": 0.0584,
|
| 1921 |
+
"step": 1365
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.605137963843958,
|
| 1925 |
+
"grad_norm": 0.639162003993988,
|
| 1926 |
+
"learning_rate": 1.416105014771163e-06,
|
| 1927 |
+
"loss": 0.0619,
|
| 1928 |
+
"step": 1370
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.61465271170314,
|
| 1932 |
+
"grad_norm": 0.6618750095367432,
|
| 1933 |
+
"learning_rate": 1.3501826223419066e-06,
|
| 1934 |
+
"loss": 0.0646,
|
| 1935 |
+
"step": 1375
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.6241674595623214,
|
| 1939 |
+
"grad_norm": 0.5845932960510254,
|
| 1940 |
+
"learning_rate": 1.285759083995266e-06,
|
| 1941 |
+
"loss": 0.0629,
|
| 1942 |
+
"step": 1380
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.6336822074215034,
|
| 1946 |
+
"grad_norm": 0.6518992185592651,
|
| 1947 |
+
"learning_rate": 1.222841473927238e-06,
|
| 1948 |
+
"loss": 0.0709,
|
| 1949 |
+
"step": 1385
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.6431969552806853,
|
| 1953 |
+
"grad_norm": 0.5598560571670532,
|
| 1954 |
+
"learning_rate": 1.1614367009714116e-06,
|
| 1955 |
+
"loss": 0.0617,
|
| 1956 |
+
"step": 1390
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.652711703139867,
|
| 1960 |
+
"grad_norm": 0.5705521702766418,
|
| 1961 |
+
"learning_rate": 1.1015515078403115e-06,
|
| 1962 |
+
"loss": 0.0667,
|
| 1963 |
+
"step": 1395
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.6622264509990483,
|
| 1967 |
+
"grad_norm": 0.702721893787384,
|
| 1968 |
+
"learning_rate": 1.04319247038501e-06,
|
| 1969 |
+
"loss": 0.0579,
|
| 1970 |
+
"step": 1400
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.6717411988582302,
|
| 1974 |
+
"grad_norm": 0.7045192718505859,
|
| 1975 |
+
"learning_rate": 9.8636599687304e-07,
|
| 1976 |
+
"loss": 0.0654,
|
| 1977 |
+
"step": 1405
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.681255946717412,
|
| 1981 |
+
"grad_norm": 0.5023800134658813,
|
| 1982 |
+
"learning_rate": 9.310783272847223e-07,
|
| 1983 |
+
"loss": 0.0605,
|
| 1984 |
+
"step": 1410
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.6907706945765937,
|
| 1988 |
+
"grad_norm": 0.7537912726402283,
|
| 1989 |
+
"learning_rate": 8.773355326279686e-07,
|
| 1990 |
+
"loss": 0.065,
|
| 1991 |
+
"step": 1415
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.7002854424357756,
|
| 1995 |
+
"grad_norm": 0.48378339409828186,
|
| 1996 |
+
"learning_rate": 8.251435142716357e-07,
|
| 1997 |
+
"loss": 0.0636,
|
| 1998 |
+
"step": 1420
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.709800190294957,
|
| 2002 |
+
"grad_norm": 0.6163333654403687,
|
| 2003 |
+
"learning_rate": 7.745080032975144e-07,
|
| 2004 |
+
"loss": 0.0636,
|
| 2005 |
+
"step": 1425
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.719314938154139,
|
| 2009 |
+
"grad_norm": 0.698273777961731,
|
| 2010 |
+
"learning_rate": 7.254345598710044e-07,
|
| 2011 |
+
"loss": 0.0658,
|
| 2012 |
+
"step": 1430
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.7288296860133205,
|
| 2016 |
+
"grad_norm": 0.5646078586578369,
|
| 2017 |
+
"learning_rate": 6.779285726305762e-07,
|
| 2018 |
+
"loss": 0.0662,
|
| 2019 |
+
"step": 1435
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.7383444338725025,
|
| 2023 |
+
"grad_norm": 0.6498432159423828,
|
| 2024 |
+
"learning_rate": 6.319952580960559e-07,
|
| 2025 |
+
"loss": 0.0701,
|
| 2026 |
+
"step": 1440
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.747859181731684,
|
| 2030 |
+
"grad_norm": 0.5409531593322754,
|
| 2031 |
+
"learning_rate": 5.876396600958034e-07,
|
| 2032 |
+
"loss": 0.0591,
|
| 2033 |
+
"step": 1445
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.757373929590866,
|
| 2037 |
+
"grad_norm": 0.519465982913971,
|
| 2038 |
+
"learning_rate": 5.448666492128673e-07,
|
| 2039 |
+
"loss": 0.0577,
|
| 2040 |
+
"step": 1450
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.766888677450048,
|
| 2044 |
+
"grad_norm": 0.5574569702148438,
|
| 2045 |
+
"learning_rate": 5.036809222501593e-07,
|
| 2046 |
+
"loss": 0.0637,
|
| 2047 |
+
"step": 1455
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.7764034253092293,
|
| 2051 |
+
"grad_norm": 0.6084758043289185,
|
| 2052 |
+
"learning_rate": 4.6408700171470474e-07,
|
| 2053 |
+
"loss": 0.0644,
|
| 2054 |
+
"step": 1460
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.785918173168411,
|
| 2058 |
+
"grad_norm": 0.6547659635543823,
|
| 2059 |
+
"learning_rate": 4.2608923532104047e-07,
|
| 2060 |
+
"loss": 0.058,
|
| 2061 |
+
"step": 1465
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.7954329210275928,
|
| 2065 |
+
"grad_norm": 0.6344557404518127,
|
| 2066 |
+
"learning_rate": 3.896917955138046e-07,
|
| 2067 |
+
"loss": 0.0737,
|
| 2068 |
+
"step": 1470
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.8049476688867747,
|
| 2072 |
+
"grad_norm": 0.5339006781578064,
|
| 2073 |
+
"learning_rate": 3.5489867900956153e-07,
|
| 2074 |
+
"loss": 0.0648,
|
| 2075 |
+
"step": 1475
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.814462416745956,
|
| 2079 |
+
"grad_norm": 0.5979055166244507,
|
| 2080 |
+
"learning_rate": 3.217137063579395e-07,
|
| 2081 |
+
"loss": 0.066,
|
| 2082 |
+
"step": 1480
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.8239771646051377,
|
| 2086 |
+
"grad_norm": 0.5584600567817688,
|
| 2087 |
+
"learning_rate": 2.901405215221081e-07,
|
| 2088 |
+
"loss": 0.066,
|
| 2089 |
+
"step": 1485
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.8334919124643196,
|
| 2093 |
+
"grad_norm": 0.6321412920951843,
|
| 2094 |
+
"learning_rate": 2.6018259147862986e-07,
|
| 2095 |
+
"loss": 0.0655,
|
| 2096 |
+
"step": 1490
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.8430066603235016,
|
| 2100 |
+
"grad_norm": 0.5815407633781433,
|
| 2101 |
+
"learning_rate": 2.3184320583677643e-07,
|
| 2102 |
+
"loss": 0.0638,
|
| 2103 |
+
"step": 1495
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.852521408182683,
|
| 2107 |
+
"grad_norm": 0.6009070873260498,
|
| 2108 |
+
"learning_rate": 2.0512547647728918e-07,
|
| 2109 |
+
"loss": 0.0592,
|
| 2110 |
+
"step": 1500
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.862036156041865,
|
| 2114 |
+
"grad_norm": 0.6436778903007507,
|
| 2115 |
+
"learning_rate": 1.8003233721068434e-07,
|
| 2116 |
+
"loss": 0.0552,
|
| 2117 |
+
"step": 1505
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.8715509039010465,
|
| 2121 |
+
"grad_norm": 0.5648630261421204,
|
| 2122 |
+
"learning_rate": 1.5656654345508892e-07,
|
| 2123 |
+
"loss": 0.0578,
|
| 2124 |
+
"step": 1510
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.8810656517602284,
|
| 2128 |
+
"grad_norm": 0.5980017781257629,
|
| 2129 |
+
"learning_rate": 1.347306719336777e-07,
|
| 2130 |
+
"loss": 0.0581,
|
| 2131 |
+
"step": 1515
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.89058039961941,
|
| 2135 |
+
"grad_norm": 0.5065551996231079,
|
| 2136 |
+
"learning_rate": 1.1452712039172808e-07,
|
| 2137 |
+
"loss": 0.0638,
|
| 2138 |
+
"step": 1520
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.900095147478592,
|
| 2142 |
+
"grad_norm": 0.5917147994041443,
|
| 2143 |
+
"learning_rate": 9.595810733333543e-08,
|
| 2144 |
+
"loss": 0.0601,
|
| 2145 |
+
"step": 1525
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.9096098953377734,
|
| 2149 |
+
"grad_norm": 0.5745869874954224,
|
| 2150 |
+
"learning_rate": 7.902567177779319e-08,
|
| 2151 |
+
"loss": 0.0642,
|
| 2152 |
+
"step": 1530
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.9191246431969553,
|
| 2156 |
+
"grad_norm": 0.7027199864387512,
|
| 2157 |
+
"learning_rate": 6.373167303570515e-08,
|
| 2158 |
+
"loss": 0.0681,
|
| 2159 |
+
"step": 1535
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.9286393910561372,
|
| 2163 |
+
"grad_norm": 0.5298718810081482,
|
| 2164 |
+
"learning_rate": 5.007779050481221e-08,
|
| 2165 |
+
"loss": 0.0555,
|
| 2166 |
+
"step": 1540
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.9381541389153187,
|
| 2170 |
+
"grad_norm": 0.5928686857223511,
|
| 2171 |
+
"learning_rate": 3.806552348557812e-08,
|
| 2172 |
+
"loss": 0.0656,
|
| 2173 |
+
"step": 1545
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.9476688867745002,
|
| 2177 |
+
"grad_norm": 0.6328768134117126,
|
| 2178 |
+
"learning_rate": 2.7696191016566287e-08,
|
| 2179 |
+
"loss": 0.0579,
|
| 2180 |
+
"step": 1550
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.957183634633682,
|
| 2184 |
+
"grad_norm": 0.559458315372467,
|
| 2185 |
+
"learning_rate": 1.8970931729590525e-08,
|
| 2186 |
+
"loss": 0.0614,
|
| 2187 |
+
"step": 1555
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.966698382492864,
|
| 2191 |
+
"grad_norm": 0.6970070004463196,
|
| 2192 |
+
"learning_rate": 1.1890703724681817e-08,
|
| 2193 |
+
"loss": 0.0658,
|
| 2194 |
+
"step": 1560
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.9762131303520456,
|
| 2198 |
+
"grad_norm": 0.5203301310539246,
|
| 2199 |
+
"learning_rate": 6.456284464892437e-09,
|
| 2200 |
+
"loss": 0.0574,
|
| 2201 |
+
"step": 1565
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.9857278782112275,
|
| 2205 |
+
"grad_norm": 0.5908868312835693,
|
| 2206 |
+
"learning_rate": 2.6682706909159085e-09,
|
| 2207 |
+
"loss": 0.062,
|
| 2208 |
+
"step": 1570
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.995242626070409,
|
| 2212 |
+
"grad_norm": 0.47268006205558777,
|
| 2213 |
+
"learning_rate": 5.270783555644254e-10,
|
| 2214 |
+
"loss": 0.0617,
|
| 2215 |
+
"step": 1575
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 3.0,
|
| 2219 |
+
"step": 1578,
|
| 2220 |
+
"total_flos": 2.066537064018477e+18,
|
| 2221 |
+
"train_loss": 0.3455555534464778,
|
| 2222 |
+
"train_runtime": 975.374,
|
| 2223 |
+
"train_samples_per_second": 51.7,
|
| 2224 |
+
"train_steps_per_second": 1.618
|
| 2225 |
+
}
|
| 2226 |
+
],
|
| 2227 |
+
"logging_steps": 5,
|
| 2228 |
+
"max_steps": 1578,
|
| 2229 |
+
"num_input_tokens_seen": 0,
|
| 2230 |
+
"num_train_epochs": 3,
|
| 2231 |
+
"save_steps": 20000,
|
| 2232 |
+
"stateful_callbacks": {
|
| 2233 |
+
"TrainerControl": {
|
| 2234 |
+
"args": {
|
| 2235 |
+
"should_epoch_stop": false,
|
| 2236 |
+
"should_evaluate": false,
|
| 2237 |
+
"should_log": false,
|
| 2238 |
+
"should_save": false,
|
| 2239 |
+
"should_training_stop": false
|
| 2240 |
+
},
|
| 2241 |
+
"attributes": {}
|
| 2242 |
+
}
|
| 2243 |
+
},
|
| 2244 |
+
"total_flos": 2.066537064018477e+18,
|
| 2245 |
+
"train_batch_size": 2,
|
| 2246 |
+
"trial_name": null,
|
| 2247 |
+
"trial_params": null
|
| 2248 |
+
}
|
139_128_e3_3e-5/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a574412b3405dd861b1c499a10007feb3aa28b0d7a41ac24a0f7dcceec1330a
|
| 3 |
+
size 8273
|
139_128_e3_3e-5/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|