Upload folder using huggingface_hub
Browse files- instruct/123_128_e3_3e-5/adapter_config.json +39 -0
- instruct/123_128_e3_3e-5/adapter_model.safetensors +3 -0
- instruct/123_128_e3_3e-5/added_tokens.json +9 -0
- instruct/123_128_e3_3e-5/all_results.json +9 -0
- instruct/123_128_e3_3e-5/chat_template.jinja +62 -0
- instruct/123_128_e3_3e-5/config.json +32 -0
- instruct/123_128_e3_3e-5/merges.txt +0 -0
- instruct/123_128_e3_3e-5/special_tokens_map.json +33 -0
- instruct/123_128_e3_3e-5/tokenizer.json +0 -0
- instruct/123_128_e3_3e-5/tokenizer_config.json +234 -0
- instruct/123_128_e3_3e-5/train_results.json +9 -0
- instruct/123_128_e3_3e-5/trainer_state.json +2549 -0
- instruct/123_128_e3_3e-5/training_args.bin +3 -0
- instruct/123_128_e3_3e-5/vocab.json +0 -0
instruct/123_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 |
+
"q_proj",
|
| 28 |
+
"down_proj",
|
| 29 |
+
"k_proj",
|
| 30 |
+
"o_proj",
|
| 31 |
+
"gate_proj",
|
| 32 |
+
"v_proj",
|
| 33 |
+
"up_proj"
|
| 34 |
+
],
|
| 35 |
+
"task_type": "CAUSAL_LM",
|
| 36 |
+
"trainable_token_indices": null,
|
| 37 |
+
"use_dora": false,
|
| 38 |
+
"use_rslora": false
|
| 39 |
+
}
|
instruct/123_128_e3_3e-5/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0a8d5639340cb3c1226239354b9a03efb14f51b1516671cf77ea9968af4b418
|
| 3 |
+
size 791751704
|
instruct/123_128_e3_3e-5/added_tokens.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|end_of_cite|>": 49156,
|
| 3 |
+
"<|end_of_plugin|>": 49158,
|
| 4 |
+
"<|end_of_role|>": 49153,
|
| 5 |
+
"<|start_of_cite|>": 49155,
|
| 6 |
+
"<|start_of_plugin|>": 49157,
|
| 7 |
+
"<|start_of_role|>": 49152,
|
| 8 |
+
"<|tool_call|>": 49154
|
| 9 |
+
}
|
instruct/123_128_e3_3e-5/all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.2654197702349292e+18,
|
| 4 |
+
"train_loss": 0.3680797904198252,
|
| 5 |
+
"train_runtime": 1122.1168,
|
| 6 |
+
"train_samples": 19127,
|
| 7 |
+
"train_samples_per_second": 51.136,
|
| 8 |
+
"train_steps_per_second": 1.599
|
| 9 |
+
}
|
instruct/123_128_e3_3e-5/chat_template.jinja
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{# Alias tools -> available_tools #}
|
| 2 |
+
{%- if tools and not available_tools -%}
|
| 3 |
+
{%- set available_tools = tools -%}
|
| 4 |
+
{%- endif -%}
|
| 5 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 6 |
+
{%- set system_message = messages[0]['content'] %}
|
| 7 |
+
{%- set loop_messages = messages[1:] %}
|
| 8 |
+
{%- else %}
|
| 9 |
+
{%- set system_message = "Knowledge Cutoff Date: April 2024.
|
| 10 |
+
Today's Date: " + strftime_now('%B %d, %Y') + ".
|
| 11 |
+
You are Granite, developed by IBM." %}
|
| 12 |
+
{%- if available_tools and documents %}
|
| 13 |
+
{%- set system_message = system_message + " You are a helpful assistant with access to the following tools. When a tool is required to answer the user's query, respond only with <|tool_call|> followed by a JSON list of tools used. If a tool does not exist in the provided list of tools, notify the user that you do not have the ability to fulfill the request.
|
| 14 |
+
Write the response to the user's input by strictly aligning with the facts in the provided documents. If the information needed to answer the question is not available in the documents, inform the user that the question cannot be answered based on the available data." %}
|
| 15 |
+
{%- elif available_tools %}
|
| 16 |
+
{%- set system_message = system_message + " You are a helpful assistant with access to the following tools. When a tool is required to answer the user's query, respond only with <|tool_call|> followed by a JSON list of tools used. If a tool does not exist in the provided list of tools, notify the user that you do not have the ability to fulfill the request." %}
|
| 17 |
+
{%- elif documents %}
|
| 18 |
+
{%- set system_message = system_message + " Write the response to the user's input by strictly aligning with the facts in the provided documents. If the information needed to answer the question is not available in the documents, inform the user that the question cannot be answered based on the available data." %}
|
| 19 |
+
{%- elif thinking %}
|
| 20 |
+
{%- set system_message = system_message + " You are a helpful AI assistant.
|
| 21 |
+
Respond to every user query in a comprehensive and detailed way. You can write down your thoughts and reasoning process before responding. In the thought process, engage in a comprehensive cycle of analysis, summarization, exploration, reassessment, reflection, backtracing, and iteration to develop well-considered thinking process. In the response section, based on various attempts, explorations, and reflections from the thoughts section, systematically present the final solution that you deem correct. The response should summarize the thought process. Write your thoughts between <think></think> and write your response between <response></response> for each user query." %}
|
| 22 |
+
{%- else %}
|
| 23 |
+
{%- set system_message = system_message + " You are a helpful AI assistant." %}
|
| 24 |
+
{%- endif %}
|
| 25 |
+
{%- if 'citations' in controls and documents %}
|
| 26 |
+
{%- set system_message = system_message + '
|
| 27 |
+
Use the symbols <|start_of_cite|> and <|end_of_cite|> to indicate when a fact comes from a document in the search result, e.g <|start_of_cite|> {document_id: 1}my fact <|end_of_cite|> for a fact from document 1. Afterwards, list all the citations with their corresponding documents in an ordered list.' %}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{%- if 'hallucinations' in controls and documents %}
|
| 30 |
+
{%- set system_message = system_message + '
|
| 31 |
+
Finally, after the response is written, include a numbered list of sentences from the response with a corresponding risk value that are hallucinated and not based in the documents.' %}
|
| 32 |
+
{%- endif %}
|
| 33 |
+
{%- set loop_messages = messages %}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{{- '<|start_of_role|>system<|end_of_role|>' + system_message + '<|end_of_text|>
|
| 36 |
+
' }}
|
| 37 |
+
{%- if available_tools %}
|
| 38 |
+
{{- '<|start_of_role|>available_tools<|end_of_role|>' }}
|
| 39 |
+
{{- available_tools | tojson(indent=4) }}
|
| 40 |
+
{{- '<|end_of_text|>
|
| 41 |
+
' }}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{%- if documents %}
|
| 44 |
+
{%- for document in documents %}
|
| 45 |
+
{{- '<|start_of_role|>document {"document_id": "' + document['doc_id'] | string + '"}<|end_of_role|>
|
| 46 |
+
' }}
|
| 47 |
+
{{- document['text'] }}
|
| 48 |
+
{{- '<|end_of_text|>
|
| 49 |
+
' }}
|
| 50 |
+
{%- endfor %}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- for message in loop_messages %}
|
| 53 |
+
{{- '<|start_of_role|>' + message['role'] + '<|end_of_role|>' + message['content'] + '<|end_of_text|>
|
| 54 |
+
' }}
|
| 55 |
+
{%- if loop.last and add_generation_prompt %}
|
| 56 |
+
{{- '<|start_of_role|>assistant' }}
|
| 57 |
+
{%- if controls %}
|
| 58 |
+
{{- ' ' + controls | tojson()}}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|end_of_role|>' }}
|
| 61 |
+
{%- endif %}
|
| 62 |
+
{%- endfor %}
|
instruct/123_128_e3_3e-5/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"GraniteForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"attention_multiplier": 0.0078125,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"embedding_multiplier": 12.0,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 4096,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 12800,
|
| 15 |
+
"logits_scaling": 16.0,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"model_type": "granite",
|
| 19 |
+
"num_attention_heads": 32,
|
| 20 |
+
"num_hidden_layers": 40,
|
| 21 |
+
"num_key_value_heads": 8,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"residual_multiplier": 0.22,
|
| 24 |
+
"rms_norm_eps": 1e-05,
|
| 25 |
+
"rope_scaling": null,
|
| 26 |
+
"rope_theta": 10000000.0,
|
| 27 |
+
"tie_word_embeddings": true,
|
| 28 |
+
"torch_dtype": "bfloat16",
|
| 29 |
+
"transformers_version": "4.52.4",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 49159
|
| 32 |
+
}
|
instruct/123_128_e3_3e-5/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
instruct/123_128_e3_3e-5/special_tokens_map.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|start_of_role|>",
|
| 4 |
+
"<|end_of_role|>",
|
| 5 |
+
"<|tool_call|>",
|
| 6 |
+
"<|start_of_cite|>",
|
| 7 |
+
"<|end_of_cite|>",
|
| 8 |
+
"<|start_of_plugin|>",
|
| 9 |
+
"<|end_of_plugin|>"
|
| 10 |
+
],
|
| 11 |
+
"bos_token": {
|
| 12 |
+
"content": "<|end_of_text|>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false
|
| 17 |
+
},
|
| 18 |
+
"eos_token": {
|
| 19 |
+
"content": "<|end_of_text|>",
|
| 20 |
+
"lstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"rstrip": false,
|
| 23 |
+
"single_word": false
|
| 24 |
+
},
|
| 25 |
+
"pad_token": "<|end_of_plugin|>",
|
| 26 |
+
"unk_token": {
|
| 27 |
+
"content": "<|end_of_text|>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": false,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
}
|
| 33 |
+
}
|
instruct/123_128_e3_3e-5/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
instruct/123_128_e3_3e-5/tokenizer_config.json
ADDED
|
@@ -0,0 +1,234 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<|end_of_text|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<fim_prefix>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "<fim_middle>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"3": {
|
| 30 |
+
"content": "<fim_suffix>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"4": {
|
| 38 |
+
"content": "<fim_pad>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"5": {
|
| 46 |
+
"content": "<filename>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"6": {
|
| 54 |
+
"content": "<gh_stars>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"7": {
|
| 62 |
+
"content": "<issue_start>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"8": {
|
| 70 |
+
"content": "<issue_comment>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"9": {
|
| 78 |
+
"content": "<issue_closed>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"10": {
|
| 86 |
+
"content": "<jupyter_start>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"11": {
|
| 94 |
+
"content": "<jupyter_text>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"12": {
|
| 102 |
+
"content": "<jupyter_code>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"13": {
|
| 110 |
+
"content": "<jupyter_output>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"14": {
|
| 118 |
+
"content": "<empty_output>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": true
|
| 124 |
+
},
|
| 125 |
+
"15": {
|
| 126 |
+
"content": "<commit_before>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": true
|
| 132 |
+
},
|
| 133 |
+
"16": {
|
| 134 |
+
"content": "<commit_msg>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": true
|
| 140 |
+
},
|
| 141 |
+
"17": {
|
| 142 |
+
"content": "<commit_after>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": true
|
| 148 |
+
},
|
| 149 |
+
"18": {
|
| 150 |
+
"content": "<reponame>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": true
|
| 156 |
+
},
|
| 157 |
+
"49152": {
|
| 158 |
+
"content": "<|start_of_role|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": true
|
| 164 |
+
},
|
| 165 |
+
"49153": {
|
| 166 |
+
"content": "<|end_of_role|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": true
|
| 172 |
+
},
|
| 173 |
+
"49154": {
|
| 174 |
+
"content": "<|tool_call|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": true
|
| 180 |
+
},
|
| 181 |
+
"49155": {
|
| 182 |
+
"content": "<|start_of_cite|>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": true
|
| 188 |
+
},
|
| 189 |
+
"49156": {
|
| 190 |
+
"content": "<|end_of_cite|>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": true
|
| 196 |
+
},
|
| 197 |
+
"49157": {
|
| 198 |
+
"content": "<|start_of_plugin|>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": true
|
| 204 |
+
},
|
| 205 |
+
"49158": {
|
| 206 |
+
"content": "<|end_of_plugin|>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": true
|
| 212 |
+
}
|
| 213 |
+
},
|
| 214 |
+
"additional_special_tokens": [
|
| 215 |
+
"<|start_of_role|>",
|
| 216 |
+
"<|end_of_role|>",
|
| 217 |
+
"<|tool_call|>",
|
| 218 |
+
"<|start_of_cite|>",
|
| 219 |
+
"<|end_of_cite|>",
|
| 220 |
+
"<|start_of_plugin|>",
|
| 221 |
+
"<|end_of_plugin|>"
|
| 222 |
+
],
|
| 223 |
+
"bos_token": "<|end_of_text|>",
|
| 224 |
+
"clean_up_tokenization_spaces": true,
|
| 225 |
+
"eos_token": "<|end_of_text|>",
|
| 226 |
+
"errors": "replace",
|
| 227 |
+
"extra_special_tokens": {},
|
| 228 |
+
"model_max_length": 8192,
|
| 229 |
+
"pad_token": "<|end_of_plugin|>",
|
| 230 |
+
"padding_side": "left",
|
| 231 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 232 |
+
"unk_token": "<|end_of_text|>",
|
| 233 |
+
"vocab_size": 49152
|
| 234 |
+
}
|
instruct/123_128_e3_3e-5/train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.2654197702349292e+18,
|
| 4 |
+
"train_loss": 0.3680797904198252,
|
| 5 |
+
"train_runtime": 1122.1168,
|
| 6 |
+
"train_samples": 19127,
|
| 7 |
+
"train_samples_per_second": 51.136,
|
| 8 |
+
"train_steps_per_second": 1.599
|
| 9 |
+
}
|
instruct/123_128_e3_3e-5/trainer_state.json
ADDED
|
@@ -0,0 +1,2549 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1794,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008361204013377926,
|
| 14 |
+
"grad_norm": 2.428767204284668,
|
| 15 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 16 |
+
"loss": 1.5744,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.016722408026755852,
|
| 21 |
+
"grad_norm": 2.3047614097595215,
|
| 22 |
+
"learning_rate": 3e-06,
|
| 23 |
+
"loss": 1.5362,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02508361204013378,
|
| 28 |
+
"grad_norm": 1.2143248319625854,
|
| 29 |
+
"learning_rate": 4.666666666666667e-06,
|
| 30 |
+
"loss": 1.5454,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.033444816053511704,
|
| 35 |
+
"grad_norm": 1.051906943321228,
|
| 36 |
+
"learning_rate": 6.333333333333333e-06,
|
| 37 |
+
"loss": 1.4653,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04180602006688963,
|
| 42 |
+
"grad_norm": 0.5391985177993774,
|
| 43 |
+
"learning_rate": 8e-06,
|
| 44 |
+
"loss": 1.4307,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05016722408026756,
|
| 49 |
+
"grad_norm": 0.48843422532081604,
|
| 50 |
+
"learning_rate": 9.666666666666667e-06,
|
| 51 |
+
"loss": 1.4223,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.05852842809364549,
|
| 56 |
+
"grad_norm": 0.39418190717697144,
|
| 57 |
+
"learning_rate": 1.1333333333333334e-05,
|
| 58 |
+
"loss": 1.3866,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06688963210702341,
|
| 63 |
+
"grad_norm": 0.4181228280067444,
|
| 64 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 65 |
+
"loss": 1.4421,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.07525083612040134,
|
| 70 |
+
"grad_norm": 0.44273337721824646,
|
| 71 |
+
"learning_rate": 1.4666666666666666e-05,
|
| 72 |
+
"loss": 1.3984,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08361204013377926,
|
| 77 |
+
"grad_norm": 0.4200684726238251,
|
| 78 |
+
"learning_rate": 1.633333333333333e-05,
|
| 79 |
+
"loss": 1.3884,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09197324414715718,
|
| 84 |
+
"grad_norm": 0.44915077090263367,
|
| 85 |
+
"learning_rate": 1.8e-05,
|
| 86 |
+
"loss": 1.3844,
|
| 87 |
+
"step": 55
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10033444816053512,
|
| 91 |
+
"grad_norm": 0.44476261734962463,
|
| 92 |
+
"learning_rate": 1.9666666666666666e-05,
|
| 93 |
+
"loss": 1.3493,
|
| 94 |
+
"step": 60
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.10869565217391304,
|
| 98 |
+
"grad_norm": 0.3984792232513428,
|
| 99 |
+
"learning_rate": 2.1333333333333335e-05,
|
| 100 |
+
"loss": 1.326,
|
| 101 |
+
"step": 65
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.11705685618729098,
|
| 105 |
+
"grad_norm": 0.43860700726509094,
|
| 106 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 107 |
+
"loss": 1.327,
|
| 108 |
+
"step": 70
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.1254180602006689,
|
| 112 |
+
"grad_norm": 0.41491463780403137,
|
| 113 |
+
"learning_rate": 2.4666666666666665e-05,
|
| 114 |
+
"loss": 1.3195,
|
| 115 |
+
"step": 75
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.13377926421404682,
|
| 119 |
+
"grad_norm": 0.4761362373828888,
|
| 120 |
+
"learning_rate": 2.6333333333333334e-05,
|
| 121 |
+
"loss": 1.2602,
|
| 122 |
+
"step": 80
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.14214046822742474,
|
| 126 |
+
"grad_norm": 0.43643829226493835,
|
| 127 |
+
"learning_rate": 2.8e-05,
|
| 128 |
+
"loss": 1.3016,
|
| 129 |
+
"step": 85
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1505016722408027,
|
| 133 |
+
"grad_norm": 0.4489060342311859,
|
| 134 |
+
"learning_rate": 2.966666666666667e-05,
|
| 135 |
+
"loss": 1.3011,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1588628762541806,
|
| 140 |
+
"grad_norm": 0.479536771774292,
|
| 141 |
+
"learning_rate": 2.9999592113020268e-05,
|
| 142 |
+
"loss": 1.2701,
|
| 143 |
+
"step": 95
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.16722408026755853,
|
| 147 |
+
"grad_norm": 0.5242592692375183,
|
| 148 |
+
"learning_rate": 2.9997935110183593e-05,
|
| 149 |
+
"loss": 1.2632,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.17558528428093645,
|
| 154 |
+
"grad_norm": 0.5387066006660461,
|
| 155 |
+
"learning_rate": 2.9995003639251288e-05,
|
| 156 |
+
"loss": 1.2208,
|
| 157 |
+
"step": 105
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.18394648829431437,
|
| 161 |
+
"grad_norm": 0.46687281131744385,
|
| 162 |
+
"learning_rate": 2.9990797949328723e-05,
|
| 163 |
+
"loss": 1.2081,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.19230769230769232,
|
| 168 |
+
"grad_norm": 0.5380125045776367,
|
| 169 |
+
"learning_rate": 2.9985318397799606e-05,
|
| 170 |
+
"loss": 1.202,
|
| 171 |
+
"step": 115
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.20066889632107024,
|
| 175 |
+
"grad_norm": 0.6144979000091553,
|
| 176 |
+
"learning_rate": 2.9978565450295605e-05,
|
| 177 |
+
"loss": 1.1171,
|
| 178 |
+
"step": 120
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.20903010033444816,
|
| 182 |
+
"grad_norm": 0.6049258708953857,
|
| 183 |
+
"learning_rate": 2.9970539680656798e-05,
|
| 184 |
+
"loss": 1.178,
|
| 185 |
+
"step": 125
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.21739130434782608,
|
| 189 |
+
"grad_norm": 0.7016860246658325,
|
| 190 |
+
"learning_rate": 2.9961241770882885e-05,
|
| 191 |
+
"loss": 1.1369,
|
| 192 |
+
"step": 130
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.225752508361204,
|
| 196 |
+
"grad_norm": 0.574518620967865,
|
| 197 |
+
"learning_rate": 2.995067251107526e-05,
|
| 198 |
+
"loss": 1.1286,
|
| 199 |
+
"step": 135
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.23411371237458195,
|
| 203 |
+
"grad_norm": 0.6339816451072693,
|
| 204 |
+
"learning_rate": 2.9938832799369854e-05,
|
| 205 |
+
"loss": 1.1623,
|
| 206 |
+
"step": 140
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.24247491638795987,
|
| 210 |
+
"grad_norm": 0.5623111128807068,
|
| 211 |
+
"learning_rate": 2.9925723641860815e-05,
|
| 212 |
+
"loss": 1.0667,
|
| 213 |
+
"step": 145
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.2508361204013378,
|
| 217 |
+
"grad_norm": 0.6678268313407898,
|
| 218 |
+
"learning_rate": 2.991134615251503e-05,
|
| 219 |
+
"loss": 1.0533,
|
| 220 |
+
"step": 150
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2591973244147157,
|
| 224 |
+
"grad_norm": 0.6887169480323792,
|
| 225 |
+
"learning_rate": 2.9895701553077442e-05,
|
| 226 |
+
"loss": 1.0845,
|
| 227 |
+
"step": 155
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.26755852842809363,
|
| 231 |
+
"grad_norm": 0.8076543211936951,
|
| 232 |
+
"learning_rate": 2.9878791172967256e-05,
|
| 233 |
+
"loss": 1.0421,
|
| 234 |
+
"step": 160
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.27591973244147155,
|
| 238 |
+
"grad_norm": 0.6907469630241394,
|
| 239 |
+
"learning_rate": 2.9860616449164947e-05,
|
| 240 |
+
"loss": 1.067,
|
| 241 |
+
"step": 165
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2842809364548495,
|
| 245 |
+
"grad_norm": 0.8173097968101501,
|
| 246 |
+
"learning_rate": 2.9841178926090155e-05,
|
| 247 |
+
"loss": 1.015,
|
| 248 |
+
"step": 170
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.29264214046822745,
|
| 252 |
+
"grad_norm": 0.8232016563415527,
|
| 253 |
+
"learning_rate": 2.982048025547046e-05,
|
| 254 |
+
"loss": 0.9976,
|
| 255 |
+
"step": 175
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3010033444816054,
|
| 259 |
+
"grad_norm": 0.7820318341255188,
|
| 260 |
+
"learning_rate": 2.9798522196201016e-05,
|
| 261 |
+
"loss": 1.0716,
|
| 262 |
+
"step": 180
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3093645484949833,
|
| 266 |
+
"grad_norm": 0.7637665867805481,
|
| 267 |
+
"learning_rate": 2.9775306614195075e-05,
|
| 268 |
+
"loss": 0.9719,
|
| 269 |
+
"step": 185
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.3177257525083612,
|
| 273 |
+
"grad_norm": 0.9077698588371277,
|
| 274 |
+
"learning_rate": 2.975083548222545e-05,
|
| 275 |
+
"loss": 0.9558,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.32608695652173914,
|
| 280 |
+
"grad_norm": 0.8507328629493713,
|
| 281 |
+
"learning_rate": 2.9725110879756868e-05,
|
| 282 |
+
"loss": 0.9066,
|
| 283 |
+
"step": 195
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.33444816053511706,
|
| 287 |
+
"grad_norm": 0.8806769847869873,
|
| 288 |
+
"learning_rate": 2.9698134992769248e-05,
|
| 289 |
+
"loss": 0.9665,
|
| 290 |
+
"step": 200
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.342809364548495,
|
| 294 |
+
"grad_norm": 0.8611456751823425,
|
| 295 |
+
"learning_rate": 2.9669910113571973e-05,
|
| 296 |
+
"loss": 1.0279,
|
| 297 |
+
"step": 205
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.3511705685618729,
|
| 301 |
+
"grad_norm": 0.8885751962661743,
|
| 302 |
+
"learning_rate": 2.9640438640609082e-05,
|
| 303 |
+
"loss": 0.8928,
|
| 304 |
+
"step": 210
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3595317725752508,
|
| 308 |
+
"grad_norm": 0.8427619934082031,
|
| 309 |
+
"learning_rate": 2.9609723078255458e-05,
|
| 310 |
+
"loss": 0.9126,
|
| 311 |
+
"step": 215
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.36789297658862874,
|
| 315 |
+
"grad_norm": 0.8372911810874939,
|
| 316 |
+
"learning_rate": 2.9577766036604024e-05,
|
| 317 |
+
"loss": 0.9419,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3762541806020067,
|
| 322 |
+
"grad_norm": 0.8554261326789856,
|
| 323 |
+
"learning_rate": 2.954457023124394e-05,
|
| 324 |
+
"loss": 0.9542,
|
| 325 |
+
"step": 225
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.38461538461538464,
|
| 329 |
+
"grad_norm": 0.9268397092819214,
|
| 330 |
+
"learning_rate": 2.9510138483029843e-05,
|
| 331 |
+
"loss": 0.8502,
|
| 332 |
+
"step": 230
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.39297658862876256,
|
| 336 |
+
"grad_norm": 0.9707223176956177,
|
| 337 |
+
"learning_rate": 2.947447371784215e-05,
|
| 338 |
+
"loss": 0.867,
|
| 339 |
+
"step": 235
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.4013377926421405,
|
| 343 |
+
"grad_norm": 0.9051746129989624,
|
| 344 |
+
"learning_rate": 2.9437578966338413e-05,
|
| 345 |
+
"loss": 0.817,
|
| 346 |
+
"step": 240
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.4096989966555184,
|
| 350 |
+
"grad_norm": 1.0233160257339478,
|
| 351 |
+
"learning_rate": 2.9399457363695803e-05,
|
| 352 |
+
"loss": 0.8224,
|
| 353 |
+
"step": 245
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.4180602006688963,
|
| 357 |
+
"grad_norm": 0.9360896348953247,
|
| 358 |
+
"learning_rate": 2.9360112149344675e-05,
|
| 359 |
+
"loss": 0.8383,
|
| 360 |
+
"step": 250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.42642140468227424,
|
| 364 |
+
"grad_norm": 1.0247498750686646,
|
| 365 |
+
"learning_rate": 2.931954666669331e-05,
|
| 366 |
+
"loss": 0.8753,
|
| 367 |
+
"step": 255
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.43478260869565216,
|
| 371 |
+
"grad_norm": 1.0177501440048218,
|
| 372 |
+
"learning_rate": 2.9277764362843793e-05,
|
| 373 |
+
"loss": 0.8282,
|
| 374 |
+
"step": 260
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.4431438127090301,
|
| 378 |
+
"grad_norm": 0.9408722519874573,
|
| 379 |
+
"learning_rate": 2.92347687882991e-05,
|
| 380 |
+
"loss": 0.8643,
|
| 381 |
+
"step": 265
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.451505016722408,
|
| 385 |
+
"grad_norm": 0.9376658201217651,
|
| 386 |
+
"learning_rate": 2.919056359666138e-05,
|
| 387 |
+
"loss": 0.7742,
|
| 388 |
+
"step": 270
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.459866220735786,
|
| 392 |
+
"grad_norm": 1.0052692890167236,
|
| 393 |
+
"learning_rate": 2.9145152544321504e-05,
|
| 394 |
+
"loss": 0.7731,
|
| 395 |
+
"step": 275
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4682274247491639,
|
| 399 |
+
"grad_norm": 1.1361019611358643,
|
| 400 |
+
"learning_rate": 2.909853949013983e-05,
|
| 401 |
+
"loss": 0.7795,
|
| 402 |
+
"step": 280
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4765886287625418,
|
| 406 |
+
"grad_norm": 0.9572595357894897,
|
| 407 |
+
"learning_rate": 2.9050728395118328e-05,
|
| 408 |
+
"loss": 0.7359,
|
| 409 |
+
"step": 285
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.48494983277591974,
|
| 413 |
+
"grad_norm": 1.0119596719741821,
|
| 414 |
+
"learning_rate": 2.9001723322063956e-05,
|
| 415 |
+
"loss": 0.7846,
|
| 416 |
+
"step": 290
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.49331103678929766,
|
| 420 |
+
"grad_norm": 1.054924488067627,
|
| 421 |
+
"learning_rate": 2.8951528435243447e-05,
|
| 422 |
+
"loss": 0.7177,
|
| 423 |
+
"step": 295
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5016722408026756,
|
| 427 |
+
"grad_norm": 1.4812935590744019,
|
| 428 |
+
"learning_rate": 2.8900148000029415e-05,
|
| 429 |
+
"loss": 0.727,
|
| 430 |
+
"step": 300
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5100334448160535,
|
| 434 |
+
"grad_norm": 1.1209754943847656,
|
| 435 |
+
"learning_rate": 2.8847586382537927e-05,
|
| 436 |
+
"loss": 0.7224,
|
| 437 |
+
"step": 305
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5183946488294314,
|
| 441 |
+
"grad_norm": 1.1178785562515259,
|
| 442 |
+
"learning_rate": 2.8793848049257483e-05,
|
| 443 |
+
"loss": 0.7154,
|
| 444 |
+
"step": 310
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5267558528428093,
|
| 448 |
+
"grad_norm": 1.0837448835372925,
|
| 449 |
+
"learning_rate": 2.873893756666944e-05,
|
| 450 |
+
"loss": 0.6842,
|
| 451 |
+
"step": 315
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5351170568561873,
|
| 455 |
+
"grad_norm": 1.1102854013442993,
|
| 456 |
+
"learning_rate": 2.8682859600860013e-05,
|
| 457 |
+
"loss": 0.7478,
|
| 458 |
+
"step": 320
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5434782608695652,
|
| 462 |
+
"grad_norm": 1.112066626548767,
|
| 463 |
+
"learning_rate": 2.8625618917123743e-05,
|
| 464 |
+
"loss": 0.7406,
|
| 465 |
+
"step": 325
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5518394648829431,
|
| 469 |
+
"grad_norm": 1.0628342628479004,
|
| 470 |
+
"learning_rate": 2.8567220379558565e-05,
|
| 471 |
+
"loss": 0.6514,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.560200668896321,
|
| 476 |
+
"grad_norm": 1.3103107213974,
|
| 477 |
+
"learning_rate": 2.850766895065247e-05,
|
| 478 |
+
"loss": 0.6924,
|
| 479 |
+
"step": 335
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.568561872909699,
|
| 483 |
+
"grad_norm": 1.2194041013717651,
|
| 484 |
+
"learning_rate": 2.844696969086183e-05,
|
| 485 |
+
"loss": 0.6807,
|
| 486 |
+
"step": 340
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5769230769230769,
|
| 490 |
+
"grad_norm": 1.1114630699157715,
|
| 491 |
+
"learning_rate": 2.8385127758181354e-05,
|
| 492 |
+
"loss": 0.6986,
|
| 493 |
+
"step": 345
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5852842809364549,
|
| 497 |
+
"grad_norm": 1.1207576990127563,
|
| 498 |
+
"learning_rate": 2.8322148407705803e-05,
|
| 499 |
+
"loss": 0.7282,
|
| 500 |
+
"step": 350
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5936454849498328,
|
| 504 |
+
"grad_norm": 1.0895614624023438,
|
| 505 |
+
"learning_rate": 2.8258036991183414e-05,
|
| 506 |
+
"loss": 0.7151,
|
| 507 |
+
"step": 355
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6020066889632107,
|
| 511 |
+
"grad_norm": 1.1493154764175415,
|
| 512 |
+
"learning_rate": 2.8192798956561158e-05,
|
| 513 |
+
"loss": 0.6586,
|
| 514 |
+
"step": 360
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6103678929765887,
|
| 518 |
+
"grad_norm": 1.3103563785552979,
|
| 519 |
+
"learning_rate": 2.812643984752174e-05,
|
| 520 |
+
"loss": 0.6383,
|
| 521 |
+
"step": 365
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.6187290969899666,
|
| 525 |
+
"grad_norm": 1.2521815299987793,
|
| 526 |
+
"learning_rate": 2.805896530301257e-05,
|
| 527 |
+
"loss": 0.6361,
|
| 528 |
+
"step": 370
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6270903010033445,
|
| 532 |
+
"grad_norm": 1.1308311223983765,
|
| 533 |
+
"learning_rate": 2.7990381056766583e-05,
|
| 534 |
+
"loss": 0.6185,
|
| 535 |
+
"step": 375
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6354515050167224,
|
| 539 |
+
"grad_norm": 1.133890151977539,
|
| 540 |
+
"learning_rate": 2.7920692936814957e-05,
|
| 541 |
+
"loss": 0.6087,
|
| 542 |
+
"step": 380
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6438127090301003,
|
| 546 |
+
"grad_norm": 1.2794893980026245,
|
| 547 |
+
"learning_rate": 2.7849906864991932e-05,
|
| 548 |
+
"loss": 0.632,
|
| 549 |
+
"step": 385
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6521739130434783,
|
| 553 |
+
"grad_norm": 1.2479077577590942,
|
| 554 |
+
"learning_rate": 2.777802885643155e-05,
|
| 555 |
+
"loss": 0.5847,
|
| 556 |
+
"step": 390
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6605351170568562,
|
| 560 |
+
"grad_norm": 1.3422638177871704,
|
| 561 |
+
"learning_rate": 2.7705065019056545e-05,
|
| 562 |
+
"loss": 0.597,
|
| 563 |
+
"step": 395
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6688963210702341,
|
| 567 |
+
"grad_norm": 1.1692026853561401,
|
| 568 |
+
"learning_rate": 2.7631021553059285e-05,
|
| 569 |
+
"loss": 0.5709,
|
| 570 |
+
"step": 400
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.677257525083612,
|
| 574 |
+
"grad_norm": 4.228718280792236,
|
| 575 |
+
"learning_rate": 2.7555904750374923e-05,
|
| 576 |
+
"loss": 0.5535,
|
| 577 |
+
"step": 405
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.68561872909699,
|
| 581 |
+
"grad_norm": 2.559650182723999,
|
| 582 |
+
"learning_rate": 2.747972099414672e-05,
|
| 583 |
+
"loss": 0.5904,
|
| 584 |
+
"step": 410
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6939799331103679,
|
| 588 |
+
"grad_norm": 1.1375120878219604,
|
| 589 |
+
"learning_rate": 2.740247675818363e-05,
|
| 590 |
+
"loss": 0.5772,
|
| 591 |
+
"step": 415
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7023411371237458,
|
| 595 |
+
"grad_norm": 1.0933048725128174,
|
| 596 |
+
"learning_rate": 2.7324178606410204e-05,
|
| 597 |
+
"loss": 0.5342,
|
| 598 |
+
"step": 420
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7107023411371237,
|
| 602 |
+
"grad_norm": 1.12921142578125,
|
| 603 |
+
"learning_rate": 2.724483319230878e-05,
|
| 604 |
+
"loss": 0.5811,
|
| 605 |
+
"step": 425
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.7190635451505016,
|
| 609 |
+
"grad_norm": 1.2569283246994019,
|
| 610 |
+
"learning_rate": 2.7164447258354104e-05,
|
| 611 |
+
"loss": 0.5728,
|
| 612 |
+
"step": 430
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7274247491638796,
|
| 616 |
+
"grad_norm": 1.141815423965454,
|
| 617 |
+
"learning_rate": 2.7083027635440392e-05,
|
| 618 |
+
"loss": 0.5648,
|
| 619 |
+
"step": 435
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7357859531772575,
|
| 623 |
+
"grad_norm": 1.2865076065063477,
|
| 624 |
+
"learning_rate": 2.700058124230086e-05,
|
| 625 |
+
"loss": 0.5854,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7441471571906354,
|
| 630 |
+
"grad_norm": 1.3393298387527466,
|
| 631 |
+
"learning_rate": 2.6917115084919784e-05,
|
| 632 |
+
"loss": 0.5413,
|
| 633 |
+
"step": 445
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7525083612040134,
|
| 637 |
+
"grad_norm": 1.2076478004455566,
|
| 638 |
+
"learning_rate": 2.683263625593718e-05,
|
| 639 |
+
"loss": 0.528,
|
| 640 |
+
"step": 450
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7608695652173914,
|
| 644 |
+
"grad_norm": 1.2185428142547607,
|
| 645 |
+
"learning_rate": 2.6747151934046077e-05,
|
| 646 |
+
"loss": 0.4777,
|
| 647 |
+
"step": 455
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.7692307692307693,
|
| 651 |
+
"grad_norm": 1.346038579940796,
|
| 652 |
+
"learning_rate": 2.666066938338252e-05,
|
| 653 |
+
"loss": 0.5019,
|
| 654 |
+
"step": 460
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7775919732441472,
|
| 658 |
+
"grad_norm": 1.254745602607727,
|
| 659 |
+
"learning_rate": 2.6573195952908265e-05,
|
| 660 |
+
"loss": 0.4879,
|
| 661 |
+
"step": 465
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7859531772575251,
|
| 665 |
+
"grad_norm": 1.4664686918258667,
|
| 666 |
+
"learning_rate": 2.648473907578632e-05,
|
| 667 |
+
"loss": 0.4837,
|
| 668 |
+
"step": 470
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.794314381270903,
|
| 672 |
+
"grad_norm": 1.2159003019332886,
|
| 673 |
+
"learning_rate": 2.6395306268749274e-05,
|
| 674 |
+
"loss": 0.5527,
|
| 675 |
+
"step": 475
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.802675585284281,
|
| 679 |
+
"grad_norm": 1.196530818939209,
|
| 680 |
+
"learning_rate": 2.6304905131460584e-05,
|
| 681 |
+
"loss": 0.4946,
|
| 682 |
+
"step": 480
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8110367892976589,
|
| 686 |
+
"grad_norm": 1.215921401977539,
|
| 687 |
+
"learning_rate": 2.6213543345868757e-05,
|
| 688 |
+
"loss": 0.4575,
|
| 689 |
+
"step": 485
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8193979933110368,
|
| 693 |
+
"grad_norm": 1.309975504875183,
|
| 694 |
+
"learning_rate": 2.612122867555458e-05,
|
| 695 |
+
"loss": 0.5229,
|
| 696 |
+
"step": 490
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.8277591973244147,
|
| 700 |
+
"grad_norm": 1.1279640197753906,
|
| 701 |
+
"learning_rate": 2.60279689650714e-05,
|
| 702 |
+
"loss": 0.5184,
|
| 703 |
+
"step": 495
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.8361204013377926,
|
| 707 |
+
"grad_norm": 1.3578612804412842,
|
| 708 |
+
"learning_rate": 2.593377213927853e-05,
|
| 709 |
+
"loss": 0.4958,
|
| 710 |
+
"step": 500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8444816053511706,
|
| 714 |
+
"grad_norm": 1.3831884860992432,
|
| 715 |
+
"learning_rate": 2.58386462026678e-05,
|
| 716 |
+
"loss": 0.4543,
|
| 717 |
+
"step": 505
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8528428093645485,
|
| 721 |
+
"grad_norm": 1.1932613849639893,
|
| 722 |
+
"learning_rate": 2.57425992386834e-05,
|
| 723 |
+
"loss": 0.4094,
|
| 724 |
+
"step": 510
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.8612040133779264,
|
| 728 |
+
"grad_norm": 1.2629352807998657,
|
| 729 |
+
"learning_rate": 2.5645639409034935e-05,
|
| 730 |
+
"loss": 0.45,
|
| 731 |
+
"step": 515
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8695652173913043,
|
| 735 |
+
"grad_norm": 1.313223123550415,
|
| 736 |
+
"learning_rate": 2.5547774953003927e-05,
|
| 737 |
+
"loss": 0.4909,
|
| 738 |
+
"step": 520
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8779264214046822,
|
| 742 |
+
"grad_norm": 0.983102560043335,
|
| 743 |
+
"learning_rate": 2.5449014186743627e-05,
|
| 744 |
+
"loss": 0.4274,
|
| 745 |
+
"step": 525
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8862876254180602,
|
| 749 |
+
"grad_norm": 1.4979335069656372,
|
| 750 |
+
"learning_rate": 2.5349365502572372e-05,
|
| 751 |
+
"loss": 0.4315,
|
| 752 |
+
"step": 530
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8946488294314381,
|
| 756 |
+
"grad_norm": 1.2491788864135742,
|
| 757 |
+
"learning_rate": 2.5248837368260404e-05,
|
| 758 |
+
"loss": 0.4518,
|
| 759 |
+
"step": 535
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.903010033444816,
|
| 763 |
+
"grad_norm": 1.160321593284607,
|
| 764 |
+
"learning_rate": 2.5147438326310335e-05,
|
| 765 |
+
"loss": 0.5185,
|
| 766 |
+
"step": 540
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.9113712374581939,
|
| 770 |
+
"grad_norm": 1.224698543548584,
|
| 771 |
+
"learning_rate": 2.5045176993231244e-05,
|
| 772 |
+
"loss": 0.4338,
|
| 773 |
+
"step": 545
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.919732441471572,
|
| 777 |
+
"grad_norm": 1.219382882118225,
|
| 778 |
+
"learning_rate": 2.4942062058806447e-05,
|
| 779 |
+
"loss": 0.3894,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9280936454849499,
|
| 784 |
+
"grad_norm": 1.3528062105178833,
|
| 785 |
+
"learning_rate": 2.4838102285355104e-05,
|
| 786 |
+
"loss": 0.4524,
|
| 787 |
+
"step": 555
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.9364548494983278,
|
| 791 |
+
"grad_norm": 1.4987643957138062,
|
| 792 |
+
"learning_rate": 2.4733306506987612e-05,
|
| 793 |
+
"loss": 0.4504,
|
| 794 |
+
"step": 560
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.9448160535117057,
|
| 798 |
+
"grad_norm": 1.2112421989440918,
|
| 799 |
+
"learning_rate": 2.4627683628854914e-05,
|
| 800 |
+
"loss": 0.4175,
|
| 801 |
+
"step": 565
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.9531772575250836,
|
| 805 |
+
"grad_norm": 1.3907016515731812,
|
| 806 |
+
"learning_rate": 2.4521242626391794e-05,
|
| 807 |
+
"loss": 0.4228,
|
| 808 |
+
"step": 570
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.9615384615384616,
|
| 812 |
+
"grad_norm": 1.2605050802230835,
|
| 813 |
+
"learning_rate": 2.4413992544554156e-05,
|
| 814 |
+
"loss": 0.3908,
|
| 815 |
+
"step": 575
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9698996655518395,
|
| 819 |
+
"grad_norm": 1.4385762214660645,
|
| 820 |
+
"learning_rate": 2.4305942497050415e-05,
|
| 821 |
+
"loss": 0.3991,
|
| 822 |
+
"step": 580
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.9782608695652174,
|
| 826 |
+
"grad_norm": 1.1734962463378906,
|
| 827 |
+
"learning_rate": 2.4197101665567064e-05,
|
| 828 |
+
"loss": 0.4127,
|
| 829 |
+
"step": 585
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9866220735785953,
|
| 833 |
+
"grad_norm": 1.3539752960205078,
|
| 834 |
+
"learning_rate": 2.4087479298988467e-05,
|
| 835 |
+
"loss": 0.4084,
|
| 836 |
+
"step": 590
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.9949832775919732,
|
| 840 |
+
"grad_norm": 1.3667895793914795,
|
| 841 |
+
"learning_rate": 2.3977084712610862e-05,
|
| 842 |
+
"loss": 0.3944,
|
| 843 |
+
"step": 595
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.0033444816053512,
|
| 847 |
+
"grad_norm": 1.360552430152893,
|
| 848 |
+
"learning_rate": 2.386592728735084e-05,
|
| 849 |
+
"loss": 0.3582,
|
| 850 |
+
"step": 600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.0117056856187292,
|
| 854 |
+
"grad_norm": 1.3769139051437378,
|
| 855 |
+
"learning_rate": 2.3754016468948174e-05,
|
| 856 |
+
"loss": 0.3777,
|
| 857 |
+
"step": 605
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.020066889632107,
|
| 861 |
+
"grad_norm": 1.1203612089157104,
|
| 862 |
+
"learning_rate": 2.364136176716314e-05,
|
| 863 |
+
"loss": 0.2938,
|
| 864 |
+
"step": 610
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.028428093645485,
|
| 868 |
+
"grad_norm": 1.4935859441757202,
|
| 869 |
+
"learning_rate": 2.352797275496843e-05,
|
| 870 |
+
"loss": 0.3177,
|
| 871 |
+
"step": 615
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.0367892976588629,
|
| 875 |
+
"grad_norm": 1.1156256198883057,
|
| 876 |
+
"learning_rate": 2.3413859067735674e-05,
|
| 877 |
+
"loss": 0.3286,
|
| 878 |
+
"step": 620
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.0451505016722409,
|
| 882 |
+
"grad_norm": 1.2689229249954224,
|
| 883 |
+
"learning_rate": 2.3299030402416656e-05,
|
| 884 |
+
"loss": 0.3305,
|
| 885 |
+
"step": 625
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.0535117056856187,
|
| 889 |
+
"grad_norm": 1.1696500778198242,
|
| 890 |
+
"learning_rate": 2.318349651671932e-05,
|
| 891 |
+
"loss": 0.3318,
|
| 892 |
+
"step": 630
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.0618729096989967,
|
| 896 |
+
"grad_norm": 1.2685688734054565,
|
| 897 |
+
"learning_rate": 2.306726722827855e-05,
|
| 898 |
+
"loss": 0.3113,
|
| 899 |
+
"step": 635
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.0702341137123745,
|
| 903 |
+
"grad_norm": 1.5556966066360474,
|
| 904 |
+
"learning_rate": 2.2950352413821987e-05,
|
| 905 |
+
"loss": 0.3039,
|
| 906 |
+
"step": 640
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.0785953177257526,
|
| 910 |
+
"grad_norm": 1.1608028411865234,
|
| 911 |
+
"learning_rate": 2.2832762008330656e-05,
|
| 912 |
+
"loss": 0.3268,
|
| 913 |
+
"step": 645
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.0869565217391304,
|
| 917 |
+
"grad_norm": 1.2109490633010864,
|
| 918 |
+
"learning_rate": 2.2714506004194782e-05,
|
| 919 |
+
"loss": 0.3281,
|
| 920 |
+
"step": 650
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.0953177257525084,
|
| 924 |
+
"grad_norm": 1.103773832321167,
|
| 925 |
+
"learning_rate": 2.2595594450364658e-05,
|
| 926 |
+
"loss": 0.2896,
|
| 927 |
+
"step": 655
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.1036789297658862,
|
| 931 |
+
"grad_norm": 1.3912843465805054,
|
| 932 |
+
"learning_rate": 2.247603745149672e-05,
|
| 933 |
+
"loss": 0.3141,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.1120401337792643,
|
| 938 |
+
"grad_norm": 1.2037835121154785,
|
| 939 |
+
"learning_rate": 2.2355845167094898e-05,
|
| 940 |
+
"loss": 0.3075,
|
| 941 |
+
"step": 665
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.120401337792642,
|
| 945 |
+
"grad_norm": 1.1449512243270874,
|
| 946 |
+
"learning_rate": 2.2235027810647292e-05,
|
| 947 |
+
"loss": 0.2866,
|
| 948 |
+
"step": 670
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.12876254180602,
|
| 952 |
+
"grad_norm": 1.450993537902832,
|
| 953 |
+
"learning_rate": 2.2113595648758273e-05,
|
| 954 |
+
"loss": 0.2998,
|
| 955 |
+
"step": 675
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.137123745819398,
|
| 959 |
+
"grad_norm": 1.2919437885284424,
|
| 960 |
+
"learning_rate": 2.199155900027607e-05,
|
| 961 |
+
"loss": 0.2879,
|
| 962 |
+
"step": 680
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.145484949832776,
|
| 966 |
+
"grad_norm": 1.4151182174682617,
|
| 967 |
+
"learning_rate": 2.186892823541591e-05,
|
| 968 |
+
"loss": 0.2648,
|
| 969 |
+
"step": 685
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.1538461538461537,
|
| 973 |
+
"grad_norm": 1.2452718019485474,
|
| 974 |
+
"learning_rate": 2.1745713774878788e-05,
|
| 975 |
+
"loss": 0.2732,
|
| 976 |
+
"step": 690
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.1622073578595318,
|
| 980 |
+
"grad_norm": 1.206667184829712,
|
| 981 |
+
"learning_rate": 2.1621926088965956e-05,
|
| 982 |
+
"loss": 0.2953,
|
| 983 |
+
"step": 695
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.1705685618729098,
|
| 987 |
+
"grad_norm": 1.2790087461471558,
|
| 988 |
+
"learning_rate": 2.149757569668921e-05,
|
| 989 |
+
"loss": 0.2728,
|
| 990 |
+
"step": 700
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.1789297658862876,
|
| 994 |
+
"grad_norm": 1.1826021671295166,
|
| 995 |
+
"learning_rate": 2.1372673164877032e-05,
|
| 996 |
+
"loss": 0.256,
|
| 997 |
+
"step": 705
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.1872909698996654,
|
| 1001 |
+
"grad_norm": 1.2781542539596558,
|
| 1002 |
+
"learning_rate": 2.124722910727661e-05,
|
| 1003 |
+
"loss": 0.2492,
|
| 1004 |
+
"step": 710
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.1956521739130435,
|
| 1008 |
+
"grad_norm": 1.316070556640625,
|
| 1009 |
+
"learning_rate": 2.1121254183651974e-05,
|
| 1010 |
+
"loss": 0.2822,
|
| 1011 |
+
"step": 715
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.2040133779264215,
|
| 1015 |
+
"grad_norm": 1.221724510192871,
|
| 1016 |
+
"learning_rate": 2.0994759098878156e-05,
|
| 1017 |
+
"loss": 0.2848,
|
| 1018 |
+
"step": 720
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.2123745819397993,
|
| 1022 |
+
"grad_norm": 1.4165841341018677,
|
| 1023 |
+
"learning_rate": 2.086775460203152e-05,
|
| 1024 |
+
"loss": 0.25,
|
| 1025 |
+
"step": 725
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.2207357859531773,
|
| 1029 |
+
"grad_norm": 1.5173105001449585,
|
| 1030 |
+
"learning_rate": 2.074025148547635e-05,
|
| 1031 |
+
"loss": 0.2927,
|
| 1032 |
+
"step": 730
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.2290969899665551,
|
| 1036 |
+
"grad_norm": 1.1870125532150269,
|
| 1037 |
+
"learning_rate": 2.0612260583947753e-05,
|
| 1038 |
+
"loss": 0.2764,
|
| 1039 |
+
"step": 735
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.2374581939799332,
|
| 1043 |
+
"grad_norm": 1.174973487854004,
|
| 1044 |
+
"learning_rate": 2.048379277363098e-05,
|
| 1045 |
+
"loss": 0.2477,
|
| 1046 |
+
"step": 740
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.245819397993311,
|
| 1050 |
+
"grad_norm": 1.2467384338378906,
|
| 1051 |
+
"learning_rate": 2.0354858971237182e-05,
|
| 1052 |
+
"loss": 0.2305,
|
| 1053 |
+
"step": 745
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.254180602006689,
|
| 1057 |
+
"grad_norm": 1.23879873752594,
|
| 1058 |
+
"learning_rate": 2.022547013307577e-05,
|
| 1059 |
+
"loss": 0.2557,
|
| 1060 |
+
"step": 750
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.2625418060200668,
|
| 1064 |
+
"grad_norm": 1.2731516361236572,
|
| 1065 |
+
"learning_rate": 2.0095637254123392e-05,
|
| 1066 |
+
"loss": 0.2333,
|
| 1067 |
+
"step": 755
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.2709030100334449,
|
| 1071 |
+
"grad_norm": 1.246065616607666,
|
| 1072 |
+
"learning_rate": 1.99653713670896e-05,
|
| 1073 |
+
"loss": 0.2349,
|
| 1074 |
+
"step": 760
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.2792642140468227,
|
| 1078 |
+
"grad_norm": 1.249688744544983,
|
| 1079 |
+
"learning_rate": 1.9834683541479335e-05,
|
| 1080 |
+
"loss": 0.2294,
|
| 1081 |
+
"step": 765
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.2876254180602007,
|
| 1085 |
+
"grad_norm": 1.3604812622070312,
|
| 1086 |
+
"learning_rate": 1.9703584882652314e-05,
|
| 1087 |
+
"loss": 0.2642,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.2959866220735785,
|
| 1092 |
+
"grad_norm": 1.1947777271270752,
|
| 1093 |
+
"learning_rate": 1.95720865308793e-05,
|
| 1094 |
+
"loss": 0.236,
|
| 1095 |
+
"step": 775
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.3043478260869565,
|
| 1099 |
+
"grad_norm": 1.3032244443893433,
|
| 1100 |
+
"learning_rate": 1.9440199660395463e-05,
|
| 1101 |
+
"loss": 0.2432,
|
| 1102 |
+
"step": 780
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.3127090301003346,
|
| 1106 |
+
"grad_norm": 1.3009603023529053,
|
| 1107 |
+
"learning_rate": 1.930793547845082e-05,
|
| 1108 |
+
"loss": 0.2528,
|
| 1109 |
+
"step": 785
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.3210702341137124,
|
| 1113 |
+
"grad_norm": 1.1324580907821655,
|
| 1114 |
+
"learning_rate": 1.9175305224357912e-05,
|
| 1115 |
+
"loss": 0.2318,
|
| 1116 |
+
"step": 790
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.3294314381270902,
|
| 1120 |
+
"grad_norm": 1.499282956123352,
|
| 1121 |
+
"learning_rate": 1.904232016853671e-05,
|
| 1122 |
+
"loss": 0.2182,
|
| 1123 |
+
"step": 795
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.3377926421404682,
|
| 1127 |
+
"grad_norm": 1.2004643678665161,
|
| 1128 |
+
"learning_rate": 1.8908991611556907e-05,
|
| 1129 |
+
"loss": 0.2227,
|
| 1130 |
+
"step": 800
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.3461538461538463,
|
| 1134 |
+
"grad_norm": 1.279281735420227,
|
| 1135 |
+
"learning_rate": 1.8775330883177603e-05,
|
| 1136 |
+
"loss": 0.252,
|
| 1137 |
+
"step": 805
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.354515050167224,
|
| 1141 |
+
"grad_norm": 1.1809158325195312,
|
| 1142 |
+
"learning_rate": 1.864134934138461e-05,
|
| 1143 |
+
"loss": 0.211,
|
| 1144 |
+
"step": 810
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.3628762541806019,
|
| 1148 |
+
"grad_norm": 1.1708887815475464,
|
| 1149 |
+
"learning_rate": 1.8507058371425234e-05,
|
| 1150 |
+
"loss": 0.2203,
|
| 1151 |
+
"step": 815
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.37123745819398,
|
| 1155 |
+
"grad_norm": 1.0978436470031738,
|
| 1156 |
+
"learning_rate": 1.8372469384840833e-05,
|
| 1157 |
+
"loss": 0.2422,
|
| 1158 |
+
"step": 820
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.379598662207358,
|
| 1162 |
+
"grad_norm": 1.2381864786148071,
|
| 1163 |
+
"learning_rate": 1.8237593818497096e-05,
|
| 1164 |
+
"loss": 0.2286,
|
| 1165 |
+
"step": 825
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.3879598662207357,
|
| 1169 |
+
"grad_norm": 1.1258418560028076,
|
| 1170 |
+
"learning_rate": 1.810244313361217e-05,
|
| 1171 |
+
"loss": 0.2184,
|
| 1172 |
+
"step": 830
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.3963210702341138,
|
| 1176 |
+
"grad_norm": 1.1974269151687622,
|
| 1177 |
+
"learning_rate": 1.796702881478276e-05,
|
| 1178 |
+
"loss": 0.2148,
|
| 1179 |
+
"step": 835
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.4046822742474916,
|
| 1183 |
+
"grad_norm": 1.1859115362167358,
|
| 1184 |
+
"learning_rate": 1.7831362369008175e-05,
|
| 1185 |
+
"loss": 0.2313,
|
| 1186 |
+
"step": 840
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.4130434782608696,
|
| 1190 |
+
"grad_norm": 1.237671136856079,
|
| 1191 |
+
"learning_rate": 1.7695455324712527e-05,
|
| 1192 |
+
"loss": 0.2142,
|
| 1193 |
+
"step": 845
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.4214046822742474,
|
| 1197 |
+
"grad_norm": 1.1161662340164185,
|
| 1198 |
+
"learning_rate": 1.7559319230765092e-05,
|
| 1199 |
+
"loss": 0.2201,
|
| 1200 |
+
"step": 850
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.4297658862876255,
|
| 1204 |
+
"grad_norm": 1.1516402959823608,
|
| 1205 |
+
"learning_rate": 1.7422965655498916e-05,
|
| 1206 |
+
"loss": 0.1909,
|
| 1207 |
+
"step": 855
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.4381270903010033,
|
| 1211 |
+
"grad_norm": 1.3841661214828491,
|
| 1212 |
+
"learning_rate": 1.7286406185727794e-05,
|
| 1213 |
+
"loss": 0.2245,
|
| 1214 |
+
"step": 860
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.4464882943143813,
|
| 1218 |
+
"grad_norm": 1.1365573406219482,
|
| 1219 |
+
"learning_rate": 1.7149652425761658e-05,
|
| 1220 |
+
"loss": 0.1824,
|
| 1221 |
+
"step": 865
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.4548494983277591,
|
| 1225 |
+
"grad_norm": 1.5044974088668823,
|
| 1226 |
+
"learning_rate": 1.701271599642049e-05,
|
| 1227 |
+
"loss": 0.212,
|
| 1228 |
+
"step": 870
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.4632107023411371,
|
| 1232 |
+
"grad_norm": 1.3054333925247192,
|
| 1233 |
+
"learning_rate": 1.687560853404684e-05,
|
| 1234 |
+
"loss": 0.2081,
|
| 1235 |
+
"step": 875
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.471571906354515,
|
| 1239 |
+
"grad_norm": 1.1648956537246704,
|
| 1240 |
+
"learning_rate": 1.6738341689516973e-05,
|
| 1241 |
+
"loss": 0.204,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.479933110367893,
|
| 1246 |
+
"grad_norm": 1.2482563257217407,
|
| 1247 |
+
"learning_rate": 1.6600927127250877e-05,
|
| 1248 |
+
"loss": 0.1912,
|
| 1249 |
+
"step": 885
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.488294314381271,
|
| 1253 |
+
"grad_norm": 1.380182147026062,
|
| 1254 |
+
"learning_rate": 1.646337652422103e-05,
|
| 1255 |
+
"loss": 0.1959,
|
| 1256 |
+
"step": 890
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.4966555183946488,
|
| 1260 |
+
"grad_norm": 1.4260421991348267,
|
| 1261 |
+
"learning_rate": 1.6325701568960136e-05,
|
| 1262 |
+
"loss": 0.2279,
|
| 1263 |
+
"step": 895
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.5050167224080266,
|
| 1267 |
+
"grad_norm": 1.284216284751892,
|
| 1268 |
+
"learning_rate": 1.6187913960567895e-05,
|
| 1269 |
+
"loss": 0.1915,
|
| 1270 |
+
"step": 900
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.5133779264214047,
|
| 1274 |
+
"grad_norm": 1.2649959325790405,
|
| 1275 |
+
"learning_rate": 1.6050025407716835e-05,
|
| 1276 |
+
"loss": 0.1899,
|
| 1277 |
+
"step": 905
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.5217391304347827,
|
| 1281 |
+
"grad_norm": 1.3045867681503296,
|
| 1282 |
+
"learning_rate": 1.5912047627657365e-05,
|
| 1283 |
+
"loss": 0.1985,
|
| 1284 |
+
"step": 910
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.5301003344481605,
|
| 1288 |
+
"grad_norm": 1.148569941520691,
|
| 1289 |
+
"learning_rate": 1.5773992345222098e-05,
|
| 1290 |
+
"loss": 0.1858,
|
| 1291 |
+
"step": 915
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.5384615384615383,
|
| 1295 |
+
"grad_norm": 1.1118544340133667,
|
| 1296 |
+
"learning_rate": 1.5635871291829496e-05,
|
| 1297 |
+
"loss": 0.1826,
|
| 1298 |
+
"step": 920
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.5468227424749164,
|
| 1302 |
+
"grad_norm": 1.17817223072052,
|
| 1303 |
+
"learning_rate": 1.5497696204486986e-05,
|
| 1304 |
+
"loss": 0.1771,
|
| 1305 |
+
"step": 925
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.5551839464882944,
|
| 1309 |
+
"grad_norm": 1.4601613283157349,
|
| 1310 |
+
"learning_rate": 1.5359478824793624e-05,
|
| 1311 |
+
"loss": 0.1857,
|
| 1312 |
+
"step": 930
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.5635451505016722,
|
| 1316 |
+
"grad_norm": 1.1704844236373901,
|
| 1317 |
+
"learning_rate": 1.5221230897942265e-05,
|
| 1318 |
+
"loss": 0.1963,
|
| 1319 |
+
"step": 935
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.57190635451505,
|
| 1323 |
+
"grad_norm": 1.1007918119430542,
|
| 1324 |
+
"learning_rate": 1.508296417172159e-05,
|
| 1325 |
+
"loss": 0.1754,
|
| 1326 |
+
"step": 940
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.580267558528428,
|
| 1330 |
+
"grad_norm": 1.1216483116149902,
|
| 1331 |
+
"learning_rate": 1.4944690395517754e-05,
|
| 1332 |
+
"loss": 0.1743,
|
| 1333 |
+
"step": 945
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.588628762541806,
|
| 1337 |
+
"grad_norm": 1.1580818891525269,
|
| 1338 |
+
"learning_rate": 1.4806421319316007e-05,
|
| 1339 |
+
"loss": 0.1894,
|
| 1340 |
+
"step": 950
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.596989966555184,
|
| 1344 |
+
"grad_norm": 1.1143430471420288,
|
| 1345 |
+
"learning_rate": 1.46681686927022e-05,
|
| 1346 |
+
"loss": 0.1683,
|
| 1347 |
+
"step": 955
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.605351170568562,
|
| 1351 |
+
"grad_norm": 1.3078250885009766,
|
| 1352 |
+
"learning_rate": 1.4529944263864363e-05,
|
| 1353 |
+
"loss": 0.1791,
|
| 1354 |
+
"step": 960
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.6137123745819397,
|
| 1358 |
+
"grad_norm": 1.232184886932373,
|
| 1359 |
+
"learning_rate": 1.4391759778594387e-05,
|
| 1360 |
+
"loss": 0.1953,
|
| 1361 |
+
"step": 965
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.6220735785953178,
|
| 1365 |
+
"grad_norm": 1.1319248676300049,
|
| 1366 |
+
"learning_rate": 1.4253626979289911e-05,
|
| 1367 |
+
"loss": 0.1915,
|
| 1368 |
+
"step": 970
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.6304347826086958,
|
| 1372 |
+
"grad_norm": 0.9206845760345459,
|
| 1373 |
+
"learning_rate": 1.4115557603956494e-05,
|
| 1374 |
+
"loss": 0.1738,
|
| 1375 |
+
"step": 975
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.6387959866220736,
|
| 1379 |
+
"grad_norm": 1.2068586349487305,
|
| 1380 |
+
"learning_rate": 1.3977563385210164e-05,
|
| 1381 |
+
"loss": 0.1799,
|
| 1382 |
+
"step": 980
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.6471571906354514,
|
| 1386 |
+
"grad_norm": 1.2171850204467773,
|
| 1387 |
+
"learning_rate": 1.3839656049280413e-05,
|
| 1388 |
+
"loss": 0.1927,
|
| 1389 |
+
"step": 985
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.6555183946488294,
|
| 1393 |
+
"grad_norm": 1.346513032913208,
|
| 1394 |
+
"learning_rate": 1.370184731501378e-05,
|
| 1395 |
+
"loss": 0.1767,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.6638795986622075,
|
| 1400 |
+
"grad_norm": 1.0660912990570068,
|
| 1401 |
+
"learning_rate": 1.356414889287799e-05,
|
| 1402 |
+
"loss": 0.1733,
|
| 1403 |
+
"step": 995
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.6722408026755853,
|
| 1407 |
+
"grad_norm": 1.1387193202972412,
|
| 1408 |
+
"learning_rate": 1.3426572483966863e-05,
|
| 1409 |
+
"loss": 0.1554,
|
| 1410 |
+
"step": 1000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.680602006688963,
|
| 1414 |
+
"grad_norm": 1.1112877130508423,
|
| 1415 |
+
"learning_rate": 1.3289129779005995e-05,
|
| 1416 |
+
"loss": 0.1548,
|
| 1417 |
+
"step": 1005
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.6889632107023411,
|
| 1421 |
+
"grad_norm": 1.0891822576522827,
|
| 1422 |
+
"learning_rate": 1.3151832457359327e-05,
|
| 1423 |
+
"loss": 0.1732,
|
| 1424 |
+
"step": 1010
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.6973244147157192,
|
| 1428 |
+
"grad_norm": 1.085573434829712,
|
| 1429 |
+
"learning_rate": 1.3014692186036664e-05,
|
| 1430 |
+
"loss": 0.1564,
|
| 1431 |
+
"step": 1015
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.705685618729097,
|
| 1435 |
+
"grad_norm": 1.064923882484436,
|
| 1436 |
+
"learning_rate": 1.2877720618702274e-05,
|
| 1437 |
+
"loss": 0.1413,
|
| 1438 |
+
"step": 1020
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.7140468227424748,
|
| 1442 |
+
"grad_norm": 1.0230779647827148,
|
| 1443 |
+
"learning_rate": 1.274092939468458e-05,
|
| 1444 |
+
"loss": 0.168,
|
| 1445 |
+
"step": 1025
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.7224080267558528,
|
| 1449 |
+
"grad_norm": 1.0332120656967163,
|
| 1450 |
+
"learning_rate": 1.2604330137987131e-05,
|
| 1451 |
+
"loss": 0.1392,
|
| 1452 |
+
"step": 1030
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.7307692307692308,
|
| 1456 |
+
"grad_norm": 0.9238594174385071,
|
| 1457 |
+
"learning_rate": 1.2467934456300798e-05,
|
| 1458 |
+
"loss": 0.148,
|
| 1459 |
+
"step": 1035
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.7391304347826086,
|
| 1463 |
+
"grad_norm": 1.226245641708374,
|
| 1464 |
+
"learning_rate": 1.2331753940017418e-05,
|
| 1465 |
+
"loss": 0.1332,
|
| 1466 |
+
"step": 1040
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.7474916387959865,
|
| 1470 |
+
"grad_norm": 0.947795033454895,
|
| 1471 |
+
"learning_rate": 1.2195800161244891e-05,
|
| 1472 |
+
"loss": 0.1687,
|
| 1473 |
+
"step": 1045
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.7558528428093645,
|
| 1477 |
+
"grad_norm": 1.1598882675170898,
|
| 1478 |
+
"learning_rate": 1.20600846728238e-05,
|
| 1479 |
+
"loss": 0.1437,
|
| 1480 |
+
"step": 1050
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.7642140468227425,
|
| 1484 |
+
"grad_norm": 1.3257359266281128,
|
| 1485 |
+
"learning_rate": 1.1924619007345725e-05,
|
| 1486 |
+
"loss": 0.1392,
|
| 1487 |
+
"step": 1055
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.7725752508361206,
|
| 1491 |
+
"grad_norm": 1.0451935529708862,
|
| 1492 |
+
"learning_rate": 1.178941467617322e-05,
|
| 1493 |
+
"loss": 0.1384,
|
| 1494 |
+
"step": 1060
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.7809364548494984,
|
| 1498 |
+
"grad_norm": 1.077939748764038,
|
| 1499 |
+
"learning_rate": 1.165448316846164e-05,
|
| 1500 |
+
"loss": 0.1608,
|
| 1501 |
+
"step": 1065
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.7892976588628762,
|
| 1505 |
+
"grad_norm": 1.2124754190444946,
|
| 1506 |
+
"learning_rate": 1.1519835950182829e-05,
|
| 1507 |
+
"loss": 0.1439,
|
| 1508 |
+
"step": 1070
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.7976588628762542,
|
| 1512 |
+
"grad_norm": 0.9916610717773438,
|
| 1513 |
+
"learning_rate": 1.1385484463150784e-05,
|
| 1514 |
+
"loss": 0.133,
|
| 1515 |
+
"step": 1075
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.8060200668896322,
|
| 1519 |
+
"grad_norm": 1.0639203786849976,
|
| 1520 |
+
"learning_rate": 1.1251440124049373e-05,
|
| 1521 |
+
"loss": 0.1431,
|
| 1522 |
+
"step": 1080
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.81438127090301,
|
| 1526 |
+
"grad_norm": 1.0501058101654053,
|
| 1527 |
+
"learning_rate": 1.1117714323462188e-05,
|
| 1528 |
+
"loss": 0.1452,
|
| 1529 |
+
"step": 1085
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.8227424749163879,
|
| 1533 |
+
"grad_norm": 0.9028212428092957,
|
| 1534 |
+
"learning_rate": 1.0984318424904617e-05,
|
| 1535 |
+
"loss": 0.132,
|
| 1536 |
+
"step": 1090
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.8311036789297659,
|
| 1540 |
+
"grad_norm": 0.9625601172447205,
|
| 1541 |
+
"learning_rate": 1.0851263763858218e-05,
|
| 1542 |
+
"loss": 0.1408,
|
| 1543 |
+
"step": 1095
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.839464882943144,
|
| 1547 |
+
"grad_norm": 1.0606342554092407,
|
| 1548 |
+
"learning_rate": 1.0718561646807458e-05,
|
| 1549 |
+
"loss": 0.136,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.8478260869565217,
|
| 1554 |
+
"grad_norm": 1.1000688076019287,
|
| 1555 |
+
"learning_rate": 1.0586223350278952e-05,
|
| 1556 |
+
"loss": 0.1541,
|
| 1557 |
+
"step": 1105
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.8561872909698995,
|
| 1561 |
+
"grad_norm": 0.9197486639022827,
|
| 1562 |
+
"learning_rate": 1.0454260119883217e-05,
|
| 1563 |
+
"loss": 0.1156,
|
| 1564 |
+
"step": 1110
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.8645484949832776,
|
| 1568 |
+
"grad_norm": 1.1074347496032715,
|
| 1569 |
+
"learning_rate": 1.0322683169359056e-05,
|
| 1570 |
+
"loss": 0.1241,
|
| 1571 |
+
"step": 1115
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.8729096989966556,
|
| 1575 |
+
"grad_norm": 0.9901503324508667,
|
| 1576 |
+
"learning_rate": 1.0191503679620662e-05,
|
| 1577 |
+
"loss": 0.1289,
|
| 1578 |
+
"step": 1120
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.8812709030100334,
|
| 1582 |
+
"grad_norm": 0.9402830600738525,
|
| 1583 |
+
"learning_rate": 1.006073279780751e-05,
|
| 1584 |
+
"loss": 0.1263,
|
| 1585 |
+
"step": 1125
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.8896321070234112,
|
| 1589 |
+
"grad_norm": 1.1589277982711792,
|
| 1590 |
+
"learning_rate": 9.930381636337112e-06,
|
| 1591 |
+
"loss": 0.1387,
|
| 1592 |
+
"step": 1130
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.8979933110367893,
|
| 1596 |
+
"grad_norm": 1.022166132926941,
|
| 1597 |
+
"learning_rate": 9.800461271960713e-06,
|
| 1598 |
+
"loss": 0.1213,
|
| 1599 |
+
"step": 1135
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.9063545150501673,
|
| 1603 |
+
"grad_norm": 0.9341086745262146,
|
| 1604 |
+
"learning_rate": 9.670982744822051e-06,
|
| 1605 |
+
"loss": 0.1086,
|
| 1606 |
+
"step": 1140
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.914715719063545,
|
| 1610 |
+
"grad_norm": 0.9373572468757629,
|
| 1611 |
+
"learning_rate": 9.541957057519197e-06,
|
| 1612 |
+
"loss": 0.1033,
|
| 1613 |
+
"step": 1145
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.9230769230769231,
|
| 1617 |
+
"grad_norm": 1.281521201133728,
|
| 1618 |
+
"learning_rate": 9.413395174169601e-06,
|
| 1619 |
+
"loss": 0.1339,
|
| 1620 |
+
"step": 1150
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.931438127090301,
|
| 1624 |
+
"grad_norm": 0.9795116782188416,
|
| 1625 |
+
"learning_rate": 9.285308019478391e-06,
|
| 1626 |
+
"loss": 0.1223,
|
| 1627 |
+
"step": 1155
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.939799331103679,
|
| 1631 |
+
"grad_norm": 1.102449655532837,
|
| 1632 |
+
"learning_rate": 9.157706477810037e-06,
|
| 1633 |
+
"loss": 0.1164,
|
| 1634 |
+
"step": 1160
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.948160535117057,
|
| 1638 |
+
"grad_norm": 0.9929519295692444,
|
| 1639 |
+
"learning_rate": 9.030601392263456e-06,
|
| 1640 |
+
"loss": 0.1412,
|
| 1641 |
+
"step": 1165
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.9565217391304348,
|
| 1645 |
+
"grad_norm": 1.066251516342163,
|
| 1646 |
+
"learning_rate": 8.904003563750579e-06,
|
| 1647 |
+
"loss": 0.1211,
|
| 1648 |
+
"step": 1170
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.9648829431438126,
|
| 1652 |
+
"grad_norm": 0.8123632669448853,
|
| 1653 |
+
"learning_rate": 8.77792375007854e-06,
|
| 1654 |
+
"loss": 0.1273,
|
| 1655 |
+
"step": 1175
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.9732441471571907,
|
| 1659 |
+
"grad_norm": 0.9759752750396729,
|
| 1660 |
+
"learning_rate": 8.652372665035534e-06,
|
| 1661 |
+
"loss": 0.1099,
|
| 1662 |
+
"step": 1180
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.9816053511705687,
|
| 1666 |
+
"grad_norm": 1.0519074201583862,
|
| 1667 |
+
"learning_rate": 8.527360977480371e-06,
|
| 1668 |
+
"loss": 0.1248,
|
| 1669 |
+
"step": 1185
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.9899665551839465,
|
| 1673 |
+
"grad_norm": 1.0605065822601318,
|
| 1674 |
+
"learning_rate": 8.402899310435914e-06,
|
| 1675 |
+
"loss": 0.1293,
|
| 1676 |
+
"step": 1190
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.9983277591973243,
|
| 1680 |
+
"grad_norm": 1.2480859756469727,
|
| 1681 |
+
"learning_rate": 8.278998240186322e-06,
|
| 1682 |
+
"loss": 0.1153,
|
| 1683 |
+
"step": 1195
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.0066889632107023,
|
| 1687 |
+
"grad_norm": 0.8604503273963928,
|
| 1688 |
+
"learning_rate": 8.155668295378378e-06,
|
| 1689 |
+
"loss": 0.0907,
|
| 1690 |
+
"step": 1200
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 2.0150501672240804,
|
| 1694 |
+
"grad_norm": 1.0056833028793335,
|
| 1695 |
+
"learning_rate": 8.03291995612676e-06,
|
| 1696 |
+
"loss": 0.0975,
|
| 1697 |
+
"step": 1205
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.0234113712374584,
|
| 1701 |
+
"grad_norm": 1.1255475282669067,
|
| 1702 |
+
"learning_rate": 7.910763653123491e-06,
|
| 1703 |
+
"loss": 0.0896,
|
| 1704 |
+
"step": 1210
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.031772575250836,
|
| 1708 |
+
"grad_norm": 0.9034900665283203,
|
| 1709 |
+
"learning_rate": 7.789209766751577e-06,
|
| 1710 |
+
"loss": 0.0991,
|
| 1711 |
+
"step": 1215
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.040133779264214,
|
| 1715 |
+
"grad_norm": 0.8074113726615906,
|
| 1716 |
+
"learning_rate": 7.668268626202943e-06,
|
| 1717 |
+
"loss": 0.0901,
|
| 1718 |
+
"step": 1220
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.048494983277592,
|
| 1722 |
+
"grad_norm": 0.8036563396453857,
|
| 1723 |
+
"learning_rate": 7.547950508600669e-06,
|
| 1724 |
+
"loss": 0.0833,
|
| 1725 |
+
"step": 1225
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.05685618729097,
|
| 1729 |
+
"grad_norm": 0.9691311120986938,
|
| 1730 |
+
"learning_rate": 7.4282656381256825e-06,
|
| 1731 |
+
"loss": 0.0971,
|
| 1732 |
+
"step": 1230
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.0652173913043477,
|
| 1736 |
+
"grad_norm": 0.8114463686943054,
|
| 1737 |
+
"learning_rate": 7.309224185147965e-06,
|
| 1738 |
+
"loss": 0.0841,
|
| 1739 |
+
"step": 1235
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.0735785953177257,
|
| 1743 |
+
"grad_norm": 0.6964769959449768,
|
| 1744 |
+
"learning_rate": 7.1908362653622976e-06,
|
| 1745 |
+
"loss": 0.0892,
|
| 1746 |
+
"step": 1240
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 2.0819397993311037,
|
| 1750 |
+
"grad_norm": 0.8318316340446472,
|
| 1751 |
+
"learning_rate": 7.073111938928663e-06,
|
| 1752 |
+
"loss": 0.0919,
|
| 1753 |
+
"step": 1245
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 2.0903010033444818,
|
| 1757 |
+
"grad_norm": 0.758114218711853,
|
| 1758 |
+
"learning_rate": 6.956061209617378e-06,
|
| 1759 |
+
"loss": 0.0936,
|
| 1760 |
+
"step": 1250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.0986622073578594,
|
| 1764 |
+
"grad_norm": 0.6917577385902405,
|
| 1765 |
+
"learning_rate": 6.839694023959016e-06,
|
| 1766 |
+
"loss": 0.0877,
|
| 1767 |
+
"step": 1255
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 2.1070234113712374,
|
| 1771 |
+
"grad_norm": 1.0525777339935303,
|
| 1772 |
+
"learning_rate": 6.724020270399187e-06,
|
| 1773 |
+
"loss": 0.0917,
|
| 1774 |
+
"step": 1260
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.1153846153846154,
|
| 1778 |
+
"grad_norm": 0.8548959493637085,
|
| 1779 |
+
"learning_rate": 6.60904977845823e-06,
|
| 1780 |
+
"loss": 0.0874,
|
| 1781 |
+
"step": 1265
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.1237458193979935,
|
| 1785 |
+
"grad_norm": 1.1010277271270752,
|
| 1786 |
+
"learning_rate": 6.494792317895986e-06,
|
| 1787 |
+
"loss": 0.088,
|
| 1788 |
+
"step": 1270
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.132107023411371,
|
| 1792 |
+
"grad_norm": 0.8416399955749512,
|
| 1793 |
+
"learning_rate": 6.3812575978815474e-06,
|
| 1794 |
+
"loss": 0.0915,
|
| 1795 |
+
"step": 1275
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.140468227424749,
|
| 1799 |
+
"grad_norm": 0.8504528403282166,
|
| 1800 |
+
"learning_rate": 6.268455266168247e-06,
|
| 1801 |
+
"loss": 0.0784,
|
| 1802 |
+
"step": 1280
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.148829431438127,
|
| 1806 |
+
"grad_norm": 0.9198423027992249,
|
| 1807 |
+
"learning_rate": 6.1563949082738184e-06,
|
| 1808 |
+
"loss": 0.0929,
|
| 1809 |
+
"step": 1285
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.157190635451505,
|
| 1813 |
+
"grad_norm": 0.9377270936965942,
|
| 1814 |
+
"learning_rate": 6.045086046665843e-06,
|
| 1815 |
+
"loss": 0.0813,
|
| 1816 |
+
"step": 1290
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.165551839464883,
|
| 1820 |
+
"grad_norm": 0.8508732914924622,
|
| 1821 |
+
"learning_rate": 5.934538139952569e-06,
|
| 1822 |
+
"loss": 0.0852,
|
| 1823 |
+
"step": 1295
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.1739130434782608,
|
| 1827 |
+
"grad_norm": 0.7873074412345886,
|
| 1828 |
+
"learning_rate": 5.824760582079179e-06,
|
| 1829 |
+
"loss": 0.0879,
|
| 1830 |
+
"step": 1300
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.182274247491639,
|
| 1834 |
+
"grad_norm": 0.8733201026916504,
|
| 1835 |
+
"learning_rate": 5.715762701529505e-06,
|
| 1836 |
+
"loss": 0.0795,
|
| 1837 |
+
"step": 1305
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.190635451505017,
|
| 1841 |
+
"grad_norm": 0.9689776301383972,
|
| 1842 |
+
"learning_rate": 5.6075537605333185e-06,
|
| 1843 |
+
"loss": 0.0939,
|
| 1844 |
+
"step": 1310
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.198996655518395,
|
| 1848 |
+
"grad_norm": 0.8010507822036743,
|
| 1849 |
+
"learning_rate": 5.500142954279293e-06,
|
| 1850 |
+
"loss": 0.08,
|
| 1851 |
+
"step": 1315
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.2073578595317724,
|
| 1855 |
+
"grad_norm": 0.8791949152946472,
|
| 1856 |
+
"learning_rate": 5.393539410133614e-06,
|
| 1857 |
+
"loss": 0.091,
|
| 1858 |
+
"step": 1320
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.2157190635451505,
|
| 1862 |
+
"grad_norm": 0.8463504314422607,
|
| 1863 |
+
"learning_rate": 5.2877521868643586e-06,
|
| 1864 |
+
"loss": 0.0701,
|
| 1865 |
+
"step": 1325
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.2240802675585285,
|
| 1869 |
+
"grad_norm": 0.7704371213912964,
|
| 1870 |
+
"learning_rate": 5.182790273871728e-06,
|
| 1871 |
+
"loss": 0.078,
|
| 1872 |
+
"step": 1330
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.2324414715719065,
|
| 1876 |
+
"grad_norm": 0.8468684554100037,
|
| 1877 |
+
"learning_rate": 5.078662590424175e-06,
|
| 1878 |
+
"loss": 0.093,
|
| 1879 |
+
"step": 1335
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.240802675585284,
|
| 1883 |
+
"grad_norm": 0.8309755325317383,
|
| 1884 |
+
"learning_rate": 4.975377984900459e-06,
|
| 1885 |
+
"loss": 0.0852,
|
| 1886 |
+
"step": 1340
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.249163879598662,
|
| 1890 |
+
"grad_norm": 0.6142987012863159,
|
| 1891 |
+
"learning_rate": 4.872945234037738e-06,
|
| 1892 |
+
"loss": 0.0795,
|
| 1893 |
+
"step": 1345
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.25752508361204,
|
| 1897 |
+
"grad_norm": 0.6354855895042419,
|
| 1898 |
+
"learning_rate": 4.771373042185779e-06,
|
| 1899 |
+
"loss": 0.0749,
|
| 1900 |
+
"step": 1350
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.265886287625418,
|
| 1904 |
+
"grad_norm": 0.7661406397819519,
|
| 1905 |
+
"learning_rate": 4.670670040567284e-06,
|
| 1906 |
+
"loss": 0.0871,
|
| 1907 |
+
"step": 1355
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.274247491638796,
|
| 1911 |
+
"grad_norm": 0.7454207539558411,
|
| 1912 |
+
"learning_rate": 4.570844786544428e-06,
|
| 1913 |
+
"loss": 0.0776,
|
| 1914 |
+
"step": 1360
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.282608695652174,
|
| 1918 |
+
"grad_norm": 0.9089949727058411,
|
| 1919 |
+
"learning_rate": 4.4719057628917e-06,
|
| 1920 |
+
"loss": 0.0779,
|
| 1921 |
+
"step": 1365
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.290969899665552,
|
| 1925 |
+
"grad_norm": 0.8225701451301575,
|
| 1926 |
+
"learning_rate": 4.373861377075077e-06,
|
| 1927 |
+
"loss": 0.0793,
|
| 1928 |
+
"step": 1370
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.29933110367893,
|
| 1932 |
+
"grad_norm": 0.6854401230812073,
|
| 1933 |
+
"learning_rate": 4.276719960537565e-06,
|
| 1934 |
+
"loss": 0.0694,
|
| 1935 |
+
"step": 1375
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.3076923076923075,
|
| 1939 |
+
"grad_norm": 0.6923999786376953,
|
| 1940 |
+
"learning_rate": 4.180489767991246e-06,
|
| 1941 |
+
"loss": 0.0821,
|
| 1942 |
+
"step": 1380
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.3160535117056855,
|
| 1946 |
+
"grad_norm": 0.6919062733650208,
|
| 1947 |
+
"learning_rate": 4.085178976715823e-06,
|
| 1948 |
+
"loss": 0.0778,
|
| 1949 |
+
"step": 1385
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.3244147157190636,
|
| 1953 |
+
"grad_norm": 0.7127764225006104,
|
| 1954 |
+
"learning_rate": 3.990795685863719e-06,
|
| 1955 |
+
"loss": 0.0762,
|
| 1956 |
+
"step": 1390
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.3327759197324416,
|
| 1960 |
+
"grad_norm": 0.6489531993865967,
|
| 1961 |
+
"learning_rate": 3.897347915771874e-06,
|
| 1962 |
+
"loss": 0.0713,
|
| 1963 |
+
"step": 1395
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.3411371237458196,
|
| 1967 |
+
"grad_norm": 0.7847180962562561,
|
| 1968 |
+
"learning_rate": 3.8048436072801996e-06,
|
| 1969 |
+
"loss": 0.0725,
|
| 1970 |
+
"step": 1400
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.349498327759197,
|
| 1974 |
+
"grad_norm": 0.8207539916038513,
|
| 1975 |
+
"learning_rate": 3.7132906210567756e-06,
|
| 1976 |
+
"loss": 0.0834,
|
| 1977 |
+
"step": 1405
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.3578595317725752,
|
| 1981 |
+
"grad_norm": 0.6497763991355896,
|
| 1982 |
+
"learning_rate": 3.622696736929896e-06,
|
| 1983 |
+
"loss": 0.0766,
|
| 1984 |
+
"step": 1410
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.3662207357859533,
|
| 1988 |
+
"grad_norm": 0.6899621486663818,
|
| 1989 |
+
"learning_rate": 3.5330696532269786e-06,
|
| 1990 |
+
"loss": 0.0822,
|
| 1991 |
+
"step": 1415
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.374581939799331,
|
| 1995 |
+
"grad_norm": 0.6841697096824646,
|
| 1996 |
+
"learning_rate": 3.4444169861203765e-06,
|
| 1997 |
+
"loss": 0.0781,
|
| 1998 |
+
"step": 1420
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.382943143812709,
|
| 2002 |
+
"grad_norm": 0.8676159977912903,
|
| 2003 |
+
"learning_rate": 3.3567462689801735e-06,
|
| 2004 |
+
"loss": 0.0746,
|
| 2005 |
+
"step": 1425
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.391304347826087,
|
| 2009 |
+
"grad_norm": 0.6744688749313354,
|
| 2010 |
+
"learning_rate": 3.270064951734056e-06,
|
| 2011 |
+
"loss": 0.0747,
|
| 2012 |
+
"step": 1430
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.399665551839465,
|
| 2016 |
+
"grad_norm": 0.5460439920425415,
|
| 2017 |
+
"learning_rate": 3.1843804002342296e-06,
|
| 2018 |
+
"loss": 0.0807,
|
| 2019 |
+
"step": 1435
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.408026755852843,
|
| 2023 |
+
"grad_norm": 0.8848717212677002,
|
| 2024 |
+
"learning_rate": 3.0996998956314744e-06,
|
| 2025 |
+
"loss": 0.0787,
|
| 2026 |
+
"step": 1440
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.4163879598662206,
|
| 2030 |
+
"grad_norm": 0.5185514688491821,
|
| 2031 |
+
"learning_rate": 3.016030633756469e-06,
|
| 2032 |
+
"loss": 0.0733,
|
| 2033 |
+
"step": 1445
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.4247491638795986,
|
| 2037 |
+
"grad_norm": 0.6057285070419312,
|
| 2038 |
+
"learning_rate": 2.9333797245082823e-06,
|
| 2039 |
+
"loss": 0.0819,
|
| 2040 |
+
"step": 1450
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.4331103678929766,
|
| 2044 |
+
"grad_norm": 0.7948001623153687,
|
| 2045 |
+
"learning_rate": 2.851754191250201e-06,
|
| 2046 |
+
"loss": 0.0767,
|
| 2047 |
+
"step": 1455
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.4414715719063547,
|
| 2051 |
+
"grad_norm": 0.6949142217636108,
|
| 2052 |
+
"learning_rate": 2.7711609702129254e-06,
|
| 2053 |
+
"loss": 0.0744,
|
| 2054 |
+
"step": 1460
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.4498327759197323,
|
| 2058 |
+
"grad_norm": 0.6895713210105896,
|
| 2059 |
+
"learning_rate": 2.6916069099051506e-06,
|
| 2060 |
+
"loss": 0.075,
|
| 2061 |
+
"step": 1465
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.4581939799331103,
|
| 2065 |
+
"grad_norm": 0.6978183388710022,
|
| 2066 |
+
"learning_rate": 2.613098770531589e-06,
|
| 2067 |
+
"loss": 0.0759,
|
| 2068 |
+
"step": 1470
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.4665551839464883,
|
| 2072 |
+
"grad_norm": 0.8206489682197571,
|
| 2073 |
+
"learning_rate": 2.5356432234185458e-06,
|
| 2074 |
+
"loss": 0.076,
|
| 2075 |
+
"step": 1475
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.4749163879598663,
|
| 2079 |
+
"grad_norm": 0.5426347255706787,
|
| 2080 |
+
"learning_rate": 2.459246850446977e-06,
|
| 2081 |
+
"loss": 0.0759,
|
| 2082 |
+
"step": 1480
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.483277591973244,
|
| 2086 |
+
"grad_norm": 0.7795494198799133,
|
| 2087 |
+
"learning_rate": 2.383916143493222e-06,
|
| 2088 |
+
"loss": 0.0708,
|
| 2089 |
+
"step": 1485
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.491638795986622,
|
| 2093 |
+
"grad_norm": 0.7876139879226685,
|
| 2094 |
+
"learning_rate": 2.309657503877321e-06,
|
| 2095 |
+
"loss": 0.0697,
|
| 2096 |
+
"step": 1490
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.5,
|
| 2100 |
+
"grad_norm": 0.6578056812286377,
|
| 2101 |
+
"learning_rate": 2.236477241819067e-06,
|
| 2102 |
+
"loss": 0.0733,
|
| 2103 |
+
"step": 1495
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.508361204013378,
|
| 2107 |
+
"grad_norm": 0.7559707164764404,
|
| 2108 |
+
"learning_rate": 2.1643815759017955e-06,
|
| 2109 |
+
"loss": 0.0663,
|
| 2110 |
+
"step": 1500
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.516722408026756,
|
| 2114 |
+
"grad_norm": 0.6470872163772583,
|
| 2115 |
+
"learning_rate": 2.0933766325439253e-06,
|
| 2116 |
+
"loss": 0.0715,
|
| 2117 |
+
"step": 1505
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.5250836120401337,
|
| 2121 |
+
"grad_norm": 0.7392722368240356,
|
| 2122 |
+
"learning_rate": 2.023468445478393e-06,
|
| 2123 |
+
"loss": 0.0841,
|
| 2124 |
+
"step": 1510
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.5334448160535117,
|
| 2128 |
+
"grad_norm": 0.7570933699607849,
|
| 2129 |
+
"learning_rate": 1.9546629552398997e-06,
|
| 2130 |
+
"loss": 0.0729,
|
| 2131 |
+
"step": 1515
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.5418060200668897,
|
| 2135 |
+
"grad_norm": 0.793716311454773,
|
| 2136 |
+
"learning_rate": 1.8869660086601242e-06,
|
| 2137 |
+
"loss": 0.0687,
|
| 2138 |
+
"step": 1520
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.5501672240802673,
|
| 2142 |
+
"grad_norm": 0.5936105251312256,
|
| 2143 |
+
"learning_rate": 1.8203833583708757e-06,
|
| 2144 |
+
"loss": 0.0665,
|
| 2145 |
+
"step": 1525
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.5585284280936453,
|
| 2149 |
+
"grad_norm": 0.6305376887321472,
|
| 2150 |
+
"learning_rate": 1.7549206623152596e-06,
|
| 2151 |
+
"loss": 0.0745,
|
| 2152 |
+
"step": 1530
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.5668896321070234,
|
| 2156 |
+
"grad_norm": 0.6653011441230774,
|
| 2157 |
+
"learning_rate": 1.6905834832668798e-06,
|
| 2158 |
+
"loss": 0.0699,
|
| 2159 |
+
"step": 1535
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.5752508361204014,
|
| 2163 |
+
"grad_norm": 0.6668432950973511,
|
| 2164 |
+
"learning_rate": 1.627377288357142e-06,
|
| 2165 |
+
"loss": 0.0791,
|
| 2166 |
+
"step": 1540
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.5836120401337794,
|
| 2170 |
+
"grad_norm": 0.5528368353843689,
|
| 2171 |
+
"learning_rate": 1.5653074486106777e-06,
|
| 2172 |
+
"loss": 0.0762,
|
| 2173 |
+
"step": 1545
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.591973244147157,
|
| 2177 |
+
"grad_norm": 0.6813206076622009,
|
| 2178 |
+
"learning_rate": 1.5043792384889282e-06,
|
| 2179 |
+
"loss": 0.0728,
|
| 2180 |
+
"step": 1550
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.600334448160535,
|
| 2184 |
+
"grad_norm": 0.6437457799911499,
|
| 2185 |
+
"learning_rate": 1.4445978354419437e-06,
|
| 2186 |
+
"loss": 0.0667,
|
| 2187 |
+
"step": 1555
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.608695652173913,
|
| 2191 |
+
"grad_norm": 0.7440688014030457,
|
| 2192 |
+
"learning_rate": 1.3859683194684258e-06,
|
| 2193 |
+
"loss": 0.0807,
|
| 2194 |
+
"step": 1560
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.617056856187291,
|
| 2198 |
+
"grad_norm": 0.6096461415290833,
|
| 2199 |
+
"learning_rate": 1.328495672684053e-06,
|
| 2200 |
+
"loss": 0.0633,
|
| 2201 |
+
"step": 1565
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.625418060200669,
|
| 2205 |
+
"grad_norm": 0.6187028288841248,
|
| 2206 |
+
"learning_rate": 1.2721847788981022e-06,
|
| 2207 |
+
"loss": 0.0716,
|
| 2208 |
+
"step": 1570
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.6337792642140467,
|
| 2212 |
+
"grad_norm": 0.5957960486412048,
|
| 2213 |
+
"learning_rate": 1.2170404231984562e-06,
|
| 2214 |
+
"loss": 0.0653,
|
| 2215 |
+
"step": 1575
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.6421404682274248,
|
| 2219 |
+
"grad_norm": 0.5660960674285889,
|
| 2220 |
+
"learning_rate": 1.1630672915449848e-06,
|
| 2221 |
+
"loss": 0.0737,
|
| 2222 |
+
"step": 1580
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.650501672240803,
|
| 2226 |
+
"grad_norm": 0.6390895247459412,
|
| 2227 |
+
"learning_rate": 1.110269970371336e-06,
|
| 2228 |
+
"loss": 0.072,
|
| 2229 |
+
"step": 1585
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.6588628762541804,
|
| 2233 |
+
"grad_norm": 0.7064967155456543,
|
| 2234 |
+
"learning_rate": 1.058652946195217e-06,
|
| 2235 |
+
"loss": 0.0742,
|
| 2236 |
+
"step": 1590
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.6672240802675584,
|
| 2240 |
+
"grad_norm": 0.66982102394104,
|
| 2241 |
+
"learning_rate": 1.0082206052371284e-06,
|
| 2242 |
+
"loss": 0.0685,
|
| 2243 |
+
"step": 1595
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.6755852842809364,
|
| 2247 |
+
"grad_norm": 0.5818952322006226,
|
| 2248 |
+
"learning_rate": 9.58977233047651e-07,
|
| 2249 |
+
"loss": 0.0656,
|
| 2250 |
+
"step": 1600
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.6839464882943145,
|
| 2254 |
+
"grad_norm": 0.6678537130355835,
|
| 2255 |
+
"learning_rate": 9.109270141432779e-07,
|
| 2256 |
+
"loss": 0.0702,
|
| 2257 |
+
"step": 1605
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.6923076923076925,
|
| 2261 |
+
"grad_norm": 0.7119616866111755,
|
| 2262 |
+
"learning_rate": 8.640740316508211e-07,
|
| 2263 |
+
"loss": 0.0819,
|
| 2264 |
+
"step": 1610
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.70066889632107,
|
| 2268 |
+
"grad_norm": 0.7172545194625854,
|
| 2269 |
+
"learning_rate": 8.184222669604463e-07,
|
| 2270 |
+
"loss": 0.0704,
|
| 2271 |
+
"step": 1615
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.709030100334448,
|
| 2275 |
+
"grad_norm": 0.7231026887893677,
|
| 2276 |
+
"learning_rate": 7.739755993873492e-07,
|
| 2277 |
+
"loss": 0.0678,
|
| 2278 |
+
"step": 1620
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 2.717391304347826,
|
| 2282 |
+
"grad_norm": 0.606120765209198,
|
| 2283 |
+
"learning_rate": 7.30737805842116e-07,
|
| 2284 |
+
"loss": 0.0772,
|
| 2285 |
+
"step": 1625
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.7257525083612038,
|
| 2289 |
+
"grad_norm": 0.7016251087188721,
|
| 2290 |
+
"learning_rate": 6.887125605097527e-07,
|
| 2291 |
+
"loss": 0.0819,
|
| 2292 |
+
"step": 1630
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.734113712374582,
|
| 2296 |
+
"grad_norm": 0.5494547486305237,
|
| 2297 |
+
"learning_rate": 6.479034345374856e-07,
|
| 2298 |
+
"loss": 0.0646,
|
| 2299 |
+
"step": 1635
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.74247491638796,
|
| 2303 |
+
"grad_norm": 0.7460204362869263,
|
| 2304 |
+
"learning_rate": 6.083138957312973e-07,
|
| 2305 |
+
"loss": 0.0699,
|
| 2306 |
+
"step": 1640
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 2.750836120401338,
|
| 2310 |
+
"grad_norm": 0.5424668192863464,
|
| 2311 |
+
"learning_rate": 5.699473082612433e-07,
|
| 2312 |
+
"loss": 0.0696,
|
| 2313 |
+
"step": 1645
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.759197324414716,
|
| 2317 |
+
"grad_norm": 0.634181559085846,
|
| 2318 |
+
"learning_rate": 5.328069323755636e-07,
|
| 2319 |
+
"loss": 0.059,
|
| 2320 |
+
"step": 1650
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.7675585284280935,
|
| 2324 |
+
"grad_norm": 0.5853201150894165,
|
| 2325 |
+
"learning_rate": 4.968959241236648e-07,
|
| 2326 |
+
"loss": 0.0726,
|
| 2327 |
+
"step": 1655
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.7759197324414715,
|
| 2331 |
+
"grad_norm": 0.7359809875488281,
|
| 2332 |
+
"learning_rate": 4.622173350879105e-07,
|
| 2333 |
+
"loss": 0.0851,
|
| 2334 |
+
"step": 1660
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.7842809364548495,
|
| 2338 |
+
"grad_norm": 0.6230887770652771,
|
| 2339 |
+
"learning_rate": 4.287741121243216e-07,
|
| 2340 |
+
"loss": 0.0681,
|
| 2341 |
+
"step": 1665
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.7926421404682276,
|
| 2345 |
+
"grad_norm": 0.673726499080658,
|
| 2346 |
+
"learning_rate": 3.965690971121483e-07,
|
| 2347 |
+
"loss": 0.0653,
|
| 2348 |
+
"step": 1670
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.8010033444816056,
|
| 2352 |
+
"grad_norm": 0.637654185295105,
|
| 2353 |
+
"learning_rate": 3.656050267123984e-07,
|
| 2354 |
+
"loss": 0.0716,
|
| 2355 |
+
"step": 1675
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.809364548494983,
|
| 2359 |
+
"grad_norm": 0.5485715270042419,
|
| 2360 |
+
"learning_rate": 3.358845321352716e-07,
|
| 2361 |
+
"loss": 0.0679,
|
| 2362 |
+
"step": 1680
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.817725752508361,
|
| 2366 |
+
"grad_norm": 0.5812855958938599,
|
| 2367 |
+
"learning_rate": 3.0741013891657175e-07,
|
| 2368 |
+
"loss": 0.0732,
|
| 2369 |
+
"step": 1685
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.8260869565217392,
|
| 2373 |
+
"grad_norm": 0.6797609329223633,
|
| 2374 |
+
"learning_rate": 2.8018426670310016e-07,
|
| 2375 |
+
"loss": 0.078,
|
| 2376 |
+
"step": 1690
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.834448160535117,
|
| 2380 |
+
"grad_norm": 0.5995035767555237,
|
| 2381 |
+
"learning_rate": 2.5420922904703823e-07,
|
| 2382 |
+
"loss": 0.0682,
|
| 2383 |
+
"step": 1695
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.842809364548495,
|
| 2387 |
+
"grad_norm": 0.6953979134559631,
|
| 2388 |
+
"learning_rate": 2.294872332093567e-07,
|
| 2389 |
+
"loss": 0.0714,
|
| 2390 |
+
"step": 1700
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.851170568561873,
|
| 2394 |
+
"grad_norm": 0.6152311563491821,
|
| 2395 |
+
"learning_rate": 2.060203799722421e-07,
|
| 2396 |
+
"loss": 0.0618,
|
| 2397 |
+
"step": 1705
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.859531772575251,
|
| 2401 |
+
"grad_norm": 0.7466418743133545,
|
| 2402 |
+
"learning_rate": 1.8381066346058962e-07,
|
| 2403 |
+
"loss": 0.0624,
|
| 2404 |
+
"step": 1710
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 2.867892976588629,
|
| 2408 |
+
"grad_norm": 0.6382126808166504,
|
| 2409 |
+
"learning_rate": 1.6285997097254368e-07,
|
| 2410 |
+
"loss": 0.0765,
|
| 2411 |
+
"step": 1715
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 2.8762541806020065,
|
| 2415 |
+
"grad_norm": 0.6276755332946777,
|
| 2416 |
+
"learning_rate": 1.4317008281912457e-07,
|
| 2417 |
+
"loss": 0.0722,
|
| 2418 |
+
"step": 1720
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 2.8846153846153846,
|
| 2422 |
+
"grad_norm": 0.5620287656784058,
|
| 2423 |
+
"learning_rate": 1.2474267217294443e-07,
|
| 2424 |
+
"loss": 0.0748,
|
| 2425 |
+
"step": 1725
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.8929765886287626,
|
| 2429 |
+
"grad_norm": 0.7062171101570129,
|
| 2430 |
+
"learning_rate": 1.0757930492602597e-07,
|
| 2431 |
+
"loss": 0.0755,
|
| 2432 |
+
"step": 1730
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 2.90133779264214,
|
| 2436 |
+
"grad_norm": 0.6054103970527649,
|
| 2437 |
+
"learning_rate": 9.168143955673892e-08,
|
| 2438 |
+
"loss": 0.0646,
|
| 2439 |
+
"step": 1735
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 2.9096989966555182,
|
| 2443 |
+
"grad_norm": 0.655422568321228,
|
| 2444 |
+
"learning_rate": 7.705042700586918e-08,
|
| 2445 |
+
"loss": 0.0711,
|
| 2446 |
+
"step": 1740
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 2.9180602006688963,
|
| 2450 |
+
"grad_norm": 0.7701873779296875,
|
| 2451 |
+
"learning_rate": 6.368751056181233e-08,
|
| 2452 |
+
"loss": 0.0729,
|
| 2453 |
+
"step": 1745
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 2.9264214046822743,
|
| 2457 |
+
"grad_norm": 0.49546903371810913,
|
| 2458 |
+
"learning_rate": 5.1593825754928106e-08,
|
| 2459 |
+
"loss": 0.0661,
|
| 2460 |
+
"step": 1750
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 2.9347826086956523,
|
| 2464 |
+
"grad_norm": 0.5851170420646667,
|
| 2465 |
+
"learning_rate": 4.0770400261050944e-08,
|
| 2466 |
+
"loss": 0.0735,
|
| 2467 |
+
"step": 1755
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 2.94314381270903,
|
| 2471 |
+
"grad_norm": 0.447939395904541,
|
| 2472 |
+
"learning_rate": 3.121815381415483e-08,
|
| 2473 |
+
"loss": 0.0655,
|
| 2474 |
+
"step": 1760
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 2.951505016722408,
|
| 2478 |
+
"grad_norm": 0.5624101758003235,
|
| 2479 |
+
"learning_rate": 2.293789812820246e-08,
|
| 2480 |
+
"loss": 0.0626,
|
| 2481 |
+
"step": 1765
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 2.959866220735786,
|
| 2485 |
+
"grad_norm": 0.6843060851097107,
|
| 2486 |
+
"learning_rate": 1.59303368281688e-08,
|
| 2487 |
+
"loss": 0.0714,
|
| 2488 |
+
"step": 1770
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 2.968227424749164,
|
| 2492 |
+
"grad_norm": 0.5496982336044312,
|
| 2493 |
+
"learning_rate": 1.019606539024387e-08,
|
| 2494 |
+
"loss": 0.0602,
|
| 2495 |
+
"step": 1775
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 2.976588628762542,
|
| 2499 |
+
"grad_norm": 0.5980583429336548,
|
| 2500 |
+
"learning_rate": 5.735571091241565e-09,
|
| 2501 |
+
"loss": 0.0762,
|
| 2502 |
+
"step": 1780
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 2.9849498327759196,
|
| 2506 |
+
"grad_norm": 0.6778466105461121,
|
| 2507 |
+
"learning_rate": 2.549232967184456e-09,
|
| 2508 |
+
"loss": 0.0742,
|
| 2509 |
+
"step": 1785
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 2.9933110367892977,
|
| 2513 |
+
"grad_norm": 0.5680410265922546,
|
| 2514 |
+
"learning_rate": 6.373217810978771e-10,
|
| 2515 |
+
"loss": 0.0646,
|
| 2516 |
+
"step": 1790
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 3.0,
|
| 2520 |
+
"step": 1794,
|
| 2521 |
+
"total_flos": 2.2654197702349292e+18,
|
| 2522 |
+
"train_loss": 0.3680797904198252,
|
| 2523 |
+
"train_runtime": 1122.1168,
|
| 2524 |
+
"train_samples_per_second": 51.136,
|
| 2525 |
+
"train_steps_per_second": 1.599
|
| 2526 |
+
}
|
| 2527 |
+
],
|
| 2528 |
+
"logging_steps": 5,
|
| 2529 |
+
"max_steps": 1794,
|
| 2530 |
+
"num_input_tokens_seen": 0,
|
| 2531 |
+
"num_train_epochs": 3,
|
| 2532 |
+
"save_steps": 20000,
|
| 2533 |
+
"stateful_callbacks": {
|
| 2534 |
+
"TrainerControl": {
|
| 2535 |
+
"args": {
|
| 2536 |
+
"should_epoch_stop": false,
|
| 2537 |
+
"should_evaluate": false,
|
| 2538 |
+
"should_log": false,
|
| 2539 |
+
"should_save": false,
|
| 2540 |
+
"should_training_stop": false
|
| 2541 |
+
},
|
| 2542 |
+
"attributes": {}
|
| 2543 |
+
}
|
| 2544 |
+
},
|
| 2545 |
+
"total_flos": 2.2654197702349292e+18,
|
| 2546 |
+
"train_batch_size": 2,
|
| 2547 |
+
"trial_name": null,
|
| 2548 |
+
"trial_params": null
|
| 2549 |
+
}
|
instruct/123_128_e3_3e-5/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c5857908a1ba5a4e43abda89343bc06ccce52e9625fbb638c9e3316f7075a75
|
| 3 |
+
size 8273
|
instruct/123_128_e3_3e-5/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|