Upload folder using huggingface_hub
Browse files- 136_128_e3_3e-5/adapter_config.json +39 -0
- 136_128_e3_3e-5/adapter_model.safetensors +3 -0
- 136_128_e3_3e-5/added_tokens.json +9 -0
- 136_128_e3_3e-5/all_results.json +9 -0
- 136_128_e3_3e-5/chat_template.jinja +62 -0
- 136_128_e3_3e-5/config.json +32 -0
- 136_128_e3_3e-5/merges.txt +0 -0
- 136_128_e3_3e-5/special_tokens_map.json +33 -0
- 136_128_e3_3e-5/tokenizer.json +0 -0
- 136_128_e3_3e-5/tokenizer_config.json +234 -0
- 136_128_e3_3e-5/train_results.json +9 -0
- 136_128_e3_3e-5/trainer_state.json +2185 -0
- 136_128_e3_3e-5/training_args.bin +3 -0
- 136_128_e3_3e-5/vocab.json +0 -0
136_128_e3_3e-5/adapter_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "ibm-granite/granite-3.3-8b-instruct",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"corda_config": null,
|
| 7 |
+
"eva_config": null,
|
| 8 |
+
"exclude_modules": null,
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 256,
|
| 17 |
+
"lora_bias": false,
|
| 18 |
+
"lora_dropout": 0.05,
|
| 19 |
+
"megatron_config": null,
|
| 20 |
+
"megatron_core": "megatron.core",
|
| 21 |
+
"modules_to_save": null,
|
| 22 |
+
"peft_type": "LORA",
|
| 23 |
+
"r": 128,
|
| 24 |
+
"rank_pattern": {},
|
| 25 |
+
"revision": null,
|
| 26 |
+
"target_modules": [
|
| 27 |
+
"up_proj",
|
| 28 |
+
"v_proj",
|
| 29 |
+
"o_proj",
|
| 30 |
+
"down_proj",
|
| 31 |
+
"k_proj",
|
| 32 |
+
"gate_proj",
|
| 33 |
+
"q_proj"
|
| 34 |
+
],
|
| 35 |
+
"task_type": "CAUSAL_LM",
|
| 36 |
+
"trainable_token_indices": null,
|
| 37 |
+
"use_dora": false,
|
| 38 |
+
"use_rslora": false
|
| 39 |
+
}
|
136_128_e3_3e-5/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:35ffda86604865506ff855ccf7fc11236300aef2f16c4cfa02817271e983aebf
|
| 3 |
+
size 791751704
|
136_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 |
+
}
|
136_128_e3_3e-5/all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.0550851304557117e+18,
|
| 4 |
+
"train_loss": 0.3431719825186919,
|
| 5 |
+
"train_runtime": 983.7331,
|
| 6 |
+
"train_samples": 16347,
|
| 7 |
+
"train_samples_per_second": 49.852,
|
| 8 |
+
"train_steps_per_second": 1.558
|
| 9 |
+
}
|
136_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 %}
|
136_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 |
+
}
|
136_128_e3_3e-5/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
136_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 |
+
}
|
136_128_e3_3e-5/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
136_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 |
+
}
|
136_128_e3_3e-5/train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.0550851304557117e+18,
|
| 4 |
+
"train_loss": 0.3431719825186919,
|
| 5 |
+
"train_runtime": 983.7331,
|
| 6 |
+
"train_samples": 16347,
|
| 7 |
+
"train_samples_per_second": 49.852,
|
| 8 |
+
"train_steps_per_second": 1.558
|
| 9 |
+
}
|
136_128_e3_3e-5/trainer_state.json
ADDED
|
@@ -0,0 +1,2185 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1533,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.009784735812133072,
|
| 14 |
+
"grad_norm": 2.357011079788208,
|
| 15 |
+
"learning_rate": 1.5584415584415587e-06,
|
| 16 |
+
"loss": 1.5533,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.019569471624266144,
|
| 21 |
+
"grad_norm": 1.8989530801773071,
|
| 22 |
+
"learning_rate": 3.5064935064935066e-06,
|
| 23 |
+
"loss": 1.4827,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.029354207436399216,
|
| 28 |
+
"grad_norm": 1.1365606784820557,
|
| 29 |
+
"learning_rate": 5.4545454545454545e-06,
|
| 30 |
+
"loss": 1.4367,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03913894324853229,
|
| 35 |
+
"grad_norm": 0.6406207084655762,
|
| 36 |
+
"learning_rate": 7.402597402597403e-06,
|
| 37 |
+
"loss": 1.3748,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04892367906066536,
|
| 42 |
+
"grad_norm": 0.5725207328796387,
|
| 43 |
+
"learning_rate": 9.35064935064935e-06,
|
| 44 |
+
"loss": 1.349,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05870841487279843,
|
| 49 |
+
"grad_norm": 0.43812453746795654,
|
| 50 |
+
"learning_rate": 1.12987012987013e-05,
|
| 51 |
+
"loss": 1.3655,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0684931506849315,
|
| 56 |
+
"grad_norm": 0.39907610416412354,
|
| 57 |
+
"learning_rate": 1.3246753246753247e-05,
|
| 58 |
+
"loss": 1.3333,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07827788649706457,
|
| 63 |
+
"grad_norm": 0.5294650197029114,
|
| 64 |
+
"learning_rate": 1.5194805194805194e-05,
|
| 65 |
+
"loss": 1.2902,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.08806262230919765,
|
| 70 |
+
"grad_norm": 0.3841898739337921,
|
| 71 |
+
"learning_rate": 1.7142857142857142e-05,
|
| 72 |
+
"loss": 1.3133,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.09784735812133072,
|
| 77 |
+
"grad_norm": 0.42246174812316895,
|
| 78 |
+
"learning_rate": 1.909090909090909e-05,
|
| 79 |
+
"loss": 1.2572,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.10763209393346379,
|
| 84 |
+
"grad_norm": 0.5118032693862915,
|
| 85 |
+
"learning_rate": 2.103896103896104e-05,
|
| 86 |
+
"loss": 1.2707,
|
| 87 |
+
"step": 55
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.11741682974559686,
|
| 91 |
+
"grad_norm": 0.3847714960575104,
|
| 92 |
+
"learning_rate": 2.298701298701299e-05,
|
| 93 |
+
"loss": 1.2148,
|
| 94 |
+
"step": 60
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.12720156555772993,
|
| 98 |
+
"grad_norm": 0.4106954336166382,
|
| 99 |
+
"learning_rate": 2.493506493506494e-05,
|
| 100 |
+
"loss": 1.2517,
|
| 101 |
+
"step": 65
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.136986301369863,
|
| 105 |
+
"grad_norm": 0.44203218817710876,
|
| 106 |
+
"learning_rate": 2.6883116883116883e-05,
|
| 107 |
+
"loss": 1.2317,
|
| 108 |
+
"step": 70
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.14677103718199608,
|
| 112 |
+
"grad_norm": 0.434898316860199,
|
| 113 |
+
"learning_rate": 2.8831168831168832e-05,
|
| 114 |
+
"loss": 1.2063,
|
| 115 |
+
"step": 75
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.15655577299412915,
|
| 119 |
+
"grad_norm": 0.44484278559684753,
|
| 120 |
+
"learning_rate": 2.999986033178712e-05,
|
| 121 |
+
"loss": 1.2038,
|
| 122 |
+
"step": 80
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.16634050880626222,
|
| 126 |
+
"grad_norm": 0.542182981967926,
|
| 127 |
+
"learning_rate": 2.999828909426247e-05,
|
| 128 |
+
"loss": 1.2007,
|
| 129 |
+
"step": 85
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1761252446183953,
|
| 133 |
+
"grad_norm": 0.5208318829536438,
|
| 134 |
+
"learning_rate": 2.9994972217431796e-05,
|
| 135 |
+
"loss": 1.1604,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.18590998043052837,
|
| 140 |
+
"grad_norm": 0.45145025849342346,
|
| 141 |
+
"learning_rate": 2.9989910087343828e-05,
|
| 142 |
+
"loss": 1.0982,
|
| 143 |
+
"step": 95
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.19569471624266144,
|
| 147 |
+
"grad_norm": 0.4416584074497223,
|
| 148 |
+
"learning_rate": 2.998310329317599e-05,
|
| 149 |
+
"loss": 1.1221,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.2054794520547945,
|
| 154 |
+
"grad_norm": 0.5566976070404053,
|
| 155 |
+
"learning_rate": 2.997455262716581e-05,
|
| 156 |
+
"loss": 1.0932,
|
| 157 |
+
"step": 105
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.21526418786692758,
|
| 161 |
+
"grad_norm": 0.5525748133659363,
|
| 162 |
+
"learning_rate": 2.9964259084518718e-05,
|
| 163 |
+
"loss": 1.0812,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.22504892367906065,
|
| 168 |
+
"grad_norm": 0.5754551291465759,
|
| 169 |
+
"learning_rate": 2.995222386329219e-05,
|
| 170 |
+
"loss": 1.0515,
|
| 171 |
+
"step": 115
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.23483365949119372,
|
| 175 |
+
"grad_norm": 0.5301942825317383,
|
| 176 |
+
"learning_rate": 2.9938448364256364e-05,
|
| 177 |
+
"loss": 1.0605,
|
| 178 |
+
"step": 120
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.2446183953033268,
|
| 182 |
+
"grad_norm": 0.6082203388214111,
|
| 183 |
+
"learning_rate": 2.992293419073095e-05,
|
| 184 |
+
"loss": 1.0602,
|
| 185 |
+
"step": 125
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.25440313111545987,
|
| 189 |
+
"grad_norm": 0.5788697600364685,
|
| 190 |
+
"learning_rate": 2.9905683148398642e-05,
|
| 191 |
+
"loss": 1.0093,
|
| 192 |
+
"step": 130
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.26418786692759294,
|
| 196 |
+
"grad_norm": 0.5942112803459167,
|
| 197 |
+
"learning_rate": 2.9886697245094958e-05,
|
| 198 |
+
"loss": 0.9631,
|
| 199 |
+
"step": 135
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.273972602739726,
|
| 203 |
+
"grad_norm": 0.6265203356742859,
|
| 204 |
+
"learning_rate": 2.9865978690574552e-05,
|
| 205 |
+
"loss": 0.9768,
|
| 206 |
+
"step": 140
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2837573385518591,
|
| 210 |
+
"grad_norm": 0.6244428753852844,
|
| 211 |
+
"learning_rate": 2.9843529896254007e-05,
|
| 212 |
+
"loss": 0.9517,
|
| 213 |
+
"step": 145
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.29354207436399216,
|
| 217 |
+
"grad_norm": 0.6594113707542419,
|
| 218 |
+
"learning_rate": 2.98193534749312e-05,
|
| 219 |
+
"loss": 0.975,
|
| 220 |
+
"step": 150
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.30332681017612523,
|
| 224 |
+
"grad_norm": 0.6485689878463745,
|
| 225 |
+
"learning_rate": 2.9793452240481162e-05,
|
| 226 |
+
"loss": 0.958,
|
| 227 |
+
"step": 155
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.3131115459882583,
|
| 231 |
+
"grad_norm": 0.7403001189231873,
|
| 232 |
+
"learning_rate": 2.9765829207528606e-05,
|
| 233 |
+
"loss": 0.9482,
|
| 234 |
+
"step": 160
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.32289628180039137,
|
| 238 |
+
"grad_norm": 0.7286673188209534,
|
| 239 |
+
"learning_rate": 2.9736487591097046e-05,
|
| 240 |
+
"loss": 0.8585,
|
| 241 |
+
"step": 165
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.33268101761252444,
|
| 245 |
+
"grad_norm": 0.7843438386917114,
|
| 246 |
+
"learning_rate": 2.9705430806234603e-05,
|
| 247 |
+
"loss": 0.8909,
|
| 248 |
+
"step": 170
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3424657534246575,
|
| 252 |
+
"grad_norm": 0.709354043006897,
|
| 253 |
+
"learning_rate": 2.967266246761652e-05,
|
| 254 |
+
"loss": 0.9095,
|
| 255 |
+
"step": 175
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3522504892367906,
|
| 259 |
+
"grad_norm": 0.7203502058982849,
|
| 260 |
+
"learning_rate": 2.9638186389124475e-05,
|
| 261 |
+
"loss": 0.8583,
|
| 262 |
+
"step": 180
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.36203522504892366,
|
| 266 |
+
"grad_norm": 0.7121981382369995,
|
| 267 |
+
"learning_rate": 2.9602006583402662e-05,
|
| 268 |
+
"loss": 0.8251,
|
| 269 |
+
"step": 185
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.37181996086105673,
|
| 273 |
+
"grad_norm": 0.8489810228347778,
|
| 274 |
+
"learning_rate": 2.956412726139078e-05,
|
| 275 |
+
"loss": 0.8325,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3816046966731898,
|
| 280 |
+
"grad_norm": 0.849193811416626,
|
| 281 |
+
"learning_rate": 2.9524552831833924e-05,
|
| 282 |
+
"loss": 0.8476,
|
| 283 |
+
"step": 195
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3913894324853229,
|
| 287 |
+
"grad_norm": 0.8722940683364868,
|
| 288 |
+
"learning_rate": 2.9483287900769436e-05,
|
| 289 |
+
"loss": 0.8101,
|
| 290 |
+
"step": 200
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.40117416829745595,
|
| 294 |
+
"grad_norm": 0.9100476503372192,
|
| 295 |
+
"learning_rate": 2.9440337270990835e-05,
|
| 296 |
+
"loss": 0.7866,
|
| 297 |
+
"step": 205
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.410958904109589,
|
| 301 |
+
"grad_norm": 0.9058657884597778,
|
| 302 |
+
"learning_rate": 2.9395705941488814e-05,
|
| 303 |
+
"loss": 0.8068,
|
| 304 |
+
"step": 210
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.4207436399217221,
|
| 308 |
+
"grad_norm": 0.9936424493789673,
|
| 309 |
+
"learning_rate": 2.9349399106869406e-05,
|
| 310 |
+
"loss": 0.7925,
|
| 311 |
+
"step": 215
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.43052837573385516,
|
| 315 |
+
"grad_norm": 0.9257177710533142,
|
| 316 |
+
"learning_rate": 2.9301422156749404e-05,
|
| 317 |
+
"loss": 0.7624,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.44031311154598823,
|
| 322 |
+
"grad_norm": 0.9148401618003845,
|
| 323 |
+
"learning_rate": 2.925178067512904e-05,
|
| 324 |
+
"loss": 0.7557,
|
| 325 |
+
"step": 225
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.4500978473581213,
|
| 329 |
+
"grad_norm": 0.8784103989601135,
|
| 330 |
+
"learning_rate": 2.9200480439742092e-05,
|
| 331 |
+
"loss": 0.7359,
|
| 332 |
+
"step": 230
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.4598825831702544,
|
| 336 |
+
"grad_norm": 0.8627538084983826,
|
| 337 |
+
"learning_rate": 2.9147527421383417e-05,
|
| 338 |
+
"loss": 0.7662,
|
| 339 |
+
"step": 235
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.46966731898238745,
|
| 343 |
+
"grad_norm": 0.898265540599823,
|
| 344 |
+
"learning_rate": 2.9092927783213996e-05,
|
| 345 |
+
"loss": 0.7214,
|
| 346 |
+
"step": 240
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.4794520547945205,
|
| 350 |
+
"grad_norm": 0.9602180123329163,
|
| 351 |
+
"learning_rate": 2.9036687880043634e-05,
|
| 352 |
+
"loss": 0.7156,
|
| 353 |
+
"step": 245
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.4892367906066536,
|
| 357 |
+
"grad_norm": 1.0888434648513794,
|
| 358 |
+
"learning_rate": 2.8978814257591304e-05,
|
| 359 |
+
"loss": 0.6845,
|
| 360 |
+
"step": 250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.49902152641878667,
|
| 364 |
+
"grad_norm": 0.8821093440055847,
|
| 365 |
+
"learning_rate": 2.891931365172331e-05,
|
| 366 |
+
"loss": 0.6605,
|
| 367 |
+
"step": 255
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5088062622309197,
|
| 371 |
+
"grad_norm": 0.9522275328636169,
|
| 372 |
+
"learning_rate": 2.8858192987669303e-05,
|
| 373 |
+
"loss": 0.7019,
|
| 374 |
+
"step": 260
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.5185909980430529,
|
| 378 |
+
"grad_norm": 1.0097681283950806,
|
| 379 |
+
"learning_rate": 2.8795459379216254e-05,
|
| 380 |
+
"loss": 0.6589,
|
| 381 |
+
"step": 265
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.5283757338551859,
|
| 385 |
+
"grad_norm": 0.9757689237594604,
|
| 386 |
+
"learning_rate": 2.8731120127880496e-05,
|
| 387 |
+
"loss": 0.6713,
|
| 388 |
+
"step": 270
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.538160469667319,
|
| 392 |
+
"grad_norm": 1.1266475915908813,
|
| 393 |
+
"learning_rate": 2.8665182722057893e-05,
|
| 394 |
+
"loss": 0.6949,
|
| 395 |
+
"step": 275
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.547945205479452,
|
| 399 |
+
"grad_norm": 1.1574108600616455,
|
| 400 |
+
"learning_rate": 2.8597654836152277e-05,
|
| 401 |
+
"loss": 0.6465,
|
| 402 |
+
"step": 280
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.5577299412915852,
|
| 406 |
+
"grad_norm": 1.0349950790405273,
|
| 407 |
+
"learning_rate": 2.852854432968224e-05,
|
| 408 |
+
"loss": 0.6615,
|
| 409 |
+
"step": 285
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5675146771037182,
|
| 413 |
+
"grad_norm": 1.0373854637145996,
|
| 414 |
+
"learning_rate": 2.8457859246366348e-05,
|
| 415 |
+
"loss": 0.6342,
|
| 416 |
+
"step": 290
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5772994129158513,
|
| 420 |
+
"grad_norm": 0.994081974029541,
|
| 421 |
+
"learning_rate": 2.8385607813186968e-05,
|
| 422 |
+
"loss": 0.5868,
|
| 423 |
+
"step": 295
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5870841487279843,
|
| 427 |
+
"grad_norm": 1.0934957265853882,
|
| 428 |
+
"learning_rate": 2.83117984394327e-05,
|
| 429 |
+
"loss": 0.6362,
|
| 430 |
+
"step": 300
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5968688845401174,
|
| 434 |
+
"grad_norm": 1.1121666431427002,
|
| 435 |
+
"learning_rate": 2.823643971571967e-05,
|
| 436 |
+
"loss": 0.5943,
|
| 437 |
+
"step": 305
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.6066536203522505,
|
| 441 |
+
"grad_norm": 1.0695664882659912,
|
| 442 |
+
"learning_rate": 2.815954041299163e-05,
|
| 443 |
+
"loss": 0.5985,
|
| 444 |
+
"step": 310
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6164383561643836,
|
| 448 |
+
"grad_norm": 1.299292802810669,
|
| 449 |
+
"learning_rate": 2.808110948149915e-05,
|
| 450 |
+
"loss": 0.6022,
|
| 451 |
+
"step": 315
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6262230919765166,
|
| 455 |
+
"grad_norm": 1.195971965789795,
|
| 456 |
+
"learning_rate": 2.800115604975789e-05,
|
| 457 |
+
"loss": 0.6159,
|
| 458 |
+
"step": 320
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.6360078277886497,
|
| 462 |
+
"grad_norm": 1.0570621490478516,
|
| 463 |
+
"learning_rate": 2.7919689423486136e-05,
|
| 464 |
+
"loss": 0.6024,
|
| 465 |
+
"step": 325
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.6457925636007827,
|
| 469 |
+
"grad_norm": 1.0627111196517944,
|
| 470 |
+
"learning_rate": 2.7836719084521714e-05,
|
| 471 |
+
"loss": 0.5783,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.6555772994129159,
|
| 476 |
+
"grad_norm": 0.9751792550086975,
|
| 477 |
+
"learning_rate": 2.775225468971842e-05,
|
| 478 |
+
"loss": 0.5847,
|
| 479 |
+
"step": 335
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6653620352250489,
|
| 483 |
+
"grad_norm": 1.2819820642471313,
|
| 484 |
+
"learning_rate": 2.7666306069822057e-05,
|
| 485 |
+
"loss": 0.5672,
|
| 486 |
+
"step": 340
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.675146771037182,
|
| 490 |
+
"grad_norm": 1.0742990970611572,
|
| 491 |
+
"learning_rate": 2.757888322832624e-05,
|
| 492 |
+
"loss": 0.5275,
|
| 493 |
+
"step": 345
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.684931506849315,
|
| 497 |
+
"grad_norm": 1.2804126739501953,
|
| 498 |
+
"learning_rate": 2.7489996340308096e-05,
|
| 499 |
+
"loss": 0.5626,
|
| 500 |
+
"step": 350
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6947162426614482,
|
| 504 |
+
"grad_norm": 1.052629828453064,
|
| 505 |
+
"learning_rate": 2.7399655751244005e-05,
|
| 506 |
+
"loss": 0.593,
|
| 507 |
+
"step": 355
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.7045009784735812,
|
| 511 |
+
"grad_norm": 1.1923245191574097,
|
| 512 |
+
"learning_rate": 2.7307871975805486e-05,
|
| 513 |
+
"loss": 0.5062,
|
| 514 |
+
"step": 360
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.7142857142857143,
|
| 518 |
+
"grad_norm": 1.086018681526184,
|
| 519 |
+
"learning_rate": 2.721465569663541e-05,
|
| 520 |
+
"loss": 0.5254,
|
| 521 |
+
"step": 365
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.7240704500978473,
|
| 525 |
+
"grad_norm": 1.1433733701705933,
|
| 526 |
+
"learning_rate": 2.7120017763104648e-05,
|
| 527 |
+
"loss": 0.4838,
|
| 528 |
+
"step": 370
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.7338551859099804,
|
| 532 |
+
"grad_norm": 1.1854525804519653,
|
| 533 |
+
"learning_rate": 2.7023969190049337e-05,
|
| 534 |
+
"loss": 0.5365,
|
| 535 |
+
"step": 375
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.7436399217221135,
|
| 539 |
+
"grad_norm": 1.0598149299621582,
|
| 540 |
+
"learning_rate": 2.6926521156488857e-05,
|
| 541 |
+
"loss": 0.5169,
|
| 542 |
+
"step": 380
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.7534246575342466,
|
| 546 |
+
"grad_norm": 1.4364194869995117,
|
| 547 |
+
"learning_rate": 2.6827685004324706e-05,
|
| 548 |
+
"loss": 0.5201,
|
| 549 |
+
"step": 385
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.7632093933463796,
|
| 553 |
+
"grad_norm": 1.0421122312545776,
|
| 554 |
+
"learning_rate": 2.672747223702045e-05,
|
| 555 |
+
"loss": 0.4953,
|
| 556 |
+
"step": 390
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.7729941291585127,
|
| 560 |
+
"grad_norm": 1.3497486114501953,
|
| 561 |
+
"learning_rate": 2.662589451826282e-05,
|
| 562 |
+
"loss": 0.4934,
|
| 563 |
+
"step": 395
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7827788649706457,
|
| 567 |
+
"grad_norm": 1.1607416868209839,
|
| 568 |
+
"learning_rate": 2.652296367060421e-05,
|
| 569 |
+
"loss": 0.4541,
|
| 570 |
+
"step": 400
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.7925636007827789,
|
| 574 |
+
"grad_norm": 1.1762253046035767,
|
| 575 |
+
"learning_rate": 2.641869167408662e-05,
|
| 576 |
+
"loss": 0.4578,
|
| 577 |
+
"step": 405
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.8023483365949119,
|
| 581 |
+
"grad_norm": 1.2061841487884521,
|
| 582 |
+
"learning_rate": 2.6313090664847358e-05,
|
| 583 |
+
"loss": 0.493,
|
| 584 |
+
"step": 410
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.812133072407045,
|
| 588 |
+
"grad_norm": 1.2003624439239502,
|
| 589 |
+
"learning_rate": 2.6206172933706478e-05,
|
| 590 |
+
"loss": 0.477,
|
| 591 |
+
"step": 415
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.821917808219178,
|
| 595 |
+
"grad_norm": 1.216618299484253,
|
| 596 |
+
"learning_rate": 2.6097950924736287e-05,
|
| 597 |
+
"loss": 0.437,
|
| 598 |
+
"step": 420
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.8317025440313112,
|
| 602 |
+
"grad_norm": 1.0385233163833618,
|
| 603 |
+
"learning_rate": 2.598843723381299e-05,
|
| 604 |
+
"loss": 0.4873,
|
| 605 |
+
"step": 425
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.8414872798434442,
|
| 609 |
+
"grad_norm": 1.1734546422958374,
|
| 610 |
+
"learning_rate": 2.587764460715065e-05,
|
| 611 |
+
"loss": 0.4655,
|
| 612 |
+
"step": 430
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.8512720156555773,
|
| 616 |
+
"grad_norm": 1.1633533239364624,
|
| 617 |
+
"learning_rate": 2.5765585939817677e-05,
|
| 618 |
+
"loss": 0.4118,
|
| 619 |
+
"step": 435
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.8610567514677103,
|
| 623 |
+
"grad_norm": 1.3289645910263062,
|
| 624 |
+
"learning_rate": 2.565227427423598e-05,
|
| 625 |
+
"loss": 0.4202,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.8708414872798435,
|
| 630 |
+
"grad_norm": 1.2137118577957153,
|
| 631 |
+
"learning_rate": 2.5537722798662973e-05,
|
| 632 |
+
"loss": 0.446,
|
| 633 |
+
"step": 445
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8806262230919765,
|
| 637 |
+
"grad_norm": 1.246453046798706,
|
| 638 |
+
"learning_rate": 2.5421944845656585e-05,
|
| 639 |
+
"loss": 0.4376,
|
| 640 |
+
"step": 450
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8904109589041096,
|
| 644 |
+
"grad_norm": 1.1725696325302124,
|
| 645 |
+
"learning_rate": 2.5304953890523526e-05,
|
| 646 |
+
"loss": 0.4214,
|
| 647 |
+
"step": 455
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9001956947162426,
|
| 651 |
+
"grad_norm": 1.1813075542449951,
|
| 652 |
+
"learning_rate": 2.5186763549750878e-05,
|
| 653 |
+
"loss": 0.4149,
|
| 654 |
+
"step": 460
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.9099804305283757,
|
| 658 |
+
"grad_norm": 1.2469724416732788,
|
| 659 |
+
"learning_rate": 2.5067387579421288e-05,
|
| 660 |
+
"loss": 0.3934,
|
| 661 |
+
"step": 465
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.9197651663405088,
|
| 665 |
+
"grad_norm": 1.105574131011963,
|
| 666 |
+
"learning_rate": 2.494683987361193e-05,
|
| 667 |
+
"loss": 0.4008,
|
| 668 |
+
"step": 470
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.9295499021526419,
|
| 672 |
+
"grad_norm": 1.1743494272232056,
|
| 673 |
+
"learning_rate": 2.482513446277735e-05,
|
| 674 |
+
"loss": 0.3838,
|
| 675 |
+
"step": 475
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.9393346379647749,
|
| 679 |
+
"grad_norm": 1.2398494482040405,
|
| 680 |
+
"learning_rate": 2.47022855121165e-05,
|
| 681 |
+
"loss": 0.3989,
|
| 682 |
+
"step": 480
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.949119373776908,
|
| 686 |
+
"grad_norm": 1.1253057718276978,
|
| 687 |
+
"learning_rate": 2.4578307319924043e-05,
|
| 688 |
+
"loss": 0.3662,
|
| 689 |
+
"step": 485
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.958904109589041,
|
| 693 |
+
"grad_norm": 1.233214259147644,
|
| 694 |
+
"learning_rate": 2.4453214315926203e-05,
|
| 695 |
+
"loss": 0.369,
|
| 696 |
+
"step": 490
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.9686888454011742,
|
| 700 |
+
"grad_norm": 1.2067475318908691,
|
| 701 |
+
"learning_rate": 2.4327021059601287e-05,
|
| 702 |
+
"loss": 0.3557,
|
| 703 |
+
"step": 495
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.9784735812133072,
|
| 707 |
+
"grad_norm": 1.2306338548660278,
|
| 708 |
+
"learning_rate": 2.4199742238485114e-05,
|
| 709 |
+
"loss": 0.3581,
|
| 710 |
+
"step": 500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.9882583170254403,
|
| 714 |
+
"grad_norm": 1.1564475297927856,
|
| 715 |
+
"learning_rate": 2.4071392666461565e-05,
|
| 716 |
+
"loss": 0.3688,
|
| 717 |
+
"step": 505
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.9980430528375733,
|
| 721 |
+
"grad_norm": 1.1837538480758667,
|
| 722 |
+
"learning_rate": 2.394198728203838e-05,
|
| 723 |
+
"loss": 0.3668,
|
| 724 |
+
"step": 510
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.0078277886497065,
|
| 728 |
+
"grad_norm": 1.1428850889205933,
|
| 729 |
+
"learning_rate": 2.3811541146608507e-05,
|
| 730 |
+
"loss": 0.3424,
|
| 731 |
+
"step": 515
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.0176125244618395,
|
| 735 |
+
"grad_norm": 1.4234540462493896,
|
| 736 |
+
"learning_rate": 2.368006944269709e-05,
|
| 737 |
+
"loss": 0.2864,
|
| 738 |
+
"step": 520
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.0273972602739727,
|
| 742 |
+
"grad_norm": 1.1882015466690063,
|
| 743 |
+
"learning_rate": 2.354758747219439e-05,
|
| 744 |
+
"loss": 0.2787,
|
| 745 |
+
"step": 525
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.0371819960861057,
|
| 749 |
+
"grad_norm": 1.0477839708328247,
|
| 750 |
+
"learning_rate": 2.341411065457484e-05,
|
| 751 |
+
"loss": 0.2824,
|
| 752 |
+
"step": 530
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.0469667318982387,
|
| 756 |
+
"grad_norm": 1.2897940874099731,
|
| 757 |
+
"learning_rate": 2.3279654525102356e-05,
|
| 758 |
+
"loss": 0.2933,
|
| 759 |
+
"step": 535
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.0567514677103718,
|
| 763 |
+
"grad_norm": 1.2501660585403442,
|
| 764 |
+
"learning_rate": 2.314423473302218e-05,
|
| 765 |
+
"loss": 0.2996,
|
| 766 |
+
"step": 540
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.0665362035225048,
|
| 770 |
+
"grad_norm": 1.0286684036254883,
|
| 771 |
+
"learning_rate": 2.3007867039739537e-05,
|
| 772 |
+
"loss": 0.3156,
|
| 773 |
+
"step": 545
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.076320939334638,
|
| 777 |
+
"grad_norm": 1.1917754411697388,
|
| 778 |
+
"learning_rate": 2.287056731698512e-05,
|
| 779 |
+
"loss": 0.2666,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.086105675146771,
|
| 784 |
+
"grad_norm": 1.2268648147583008,
|
| 785 |
+
"learning_rate": 2.2732351544967824e-05,
|
| 786 |
+
"loss": 0.2839,
|
| 787 |
+
"step": 555
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.095890410958904,
|
| 791 |
+
"grad_norm": 1.0929343700408936,
|
| 792 |
+
"learning_rate": 2.259323581051481e-05,
|
| 793 |
+
"loss": 0.2883,
|
| 794 |
+
"step": 560
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.1056751467710373,
|
| 798 |
+
"grad_norm": 1.4470988512039185,
|
| 799 |
+
"learning_rate": 2.245323630519917e-05,
|
| 800 |
+
"loss": 0.2626,
|
| 801 |
+
"step": 565
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.1154598825831703,
|
| 805 |
+
"grad_norm": 1.1646130084991455,
|
| 806 |
+
"learning_rate": 2.23123693234554e-05,
|
| 807 |
+
"loss": 0.286,
|
| 808 |
+
"step": 570
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.1252446183953033,
|
| 812 |
+
"grad_norm": 1.3293094635009766,
|
| 813 |
+
"learning_rate": 2.2170651260682925e-05,
|
| 814 |
+
"loss": 0.2728,
|
| 815 |
+
"step": 575
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.1350293542074363,
|
| 819 |
+
"grad_norm": 0.9581043720245361,
|
| 820 |
+
"learning_rate": 2.2028098611337828e-05,
|
| 821 |
+
"loss": 0.2873,
|
| 822 |
+
"step": 580
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.1448140900195694,
|
| 826 |
+
"grad_norm": 1.1518635749816895,
|
| 827 |
+
"learning_rate": 2.1884727967013065e-05,
|
| 828 |
+
"loss": 0.2497,
|
| 829 |
+
"step": 585
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.1545988258317026,
|
| 833 |
+
"grad_norm": 1.1648210287094116,
|
| 834 |
+
"learning_rate": 2.174055601450742e-05,
|
| 835 |
+
"loss": 0.2715,
|
| 836 |
+
"step": 590
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.1643835616438356,
|
| 840 |
+
"grad_norm": 1.4971526861190796,
|
| 841 |
+
"learning_rate": 2.1595599533883306e-05,
|
| 842 |
+
"loss": 0.2834,
|
| 843 |
+
"step": 595
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.1741682974559686,
|
| 847 |
+
"grad_norm": 1.225674033164978,
|
| 848 |
+
"learning_rate": 2.144987539651376e-05,
|
| 849 |
+
"loss": 0.2445,
|
| 850 |
+
"step": 600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.1839530332681019,
|
| 854 |
+
"grad_norm": 1.2057493925094604,
|
| 855 |
+
"learning_rate": 2.130340056311879e-05,
|
| 856 |
+
"loss": 0.246,
|
| 857 |
+
"step": 605
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.1937377690802349,
|
| 861 |
+
"grad_norm": 1.2546255588531494,
|
| 862 |
+
"learning_rate": 2.1156192081791355e-05,
|
| 863 |
+
"loss": 0.2638,
|
| 864 |
+
"step": 610
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.203522504892368,
|
| 868 |
+
"grad_norm": 1.2434742450714111,
|
| 869 |
+
"learning_rate": 2.100826708601311e-05,
|
| 870 |
+
"loss": 0.2425,
|
| 871 |
+
"step": 615
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.213307240704501,
|
| 875 |
+
"grad_norm": 1.1534274816513062,
|
| 876 |
+
"learning_rate": 2.0859642792660284e-05,
|
| 877 |
+
"loss": 0.2145,
|
| 878 |
+
"step": 620
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.223091976516634,
|
| 882 |
+
"grad_norm": 1.1731568574905396,
|
| 883 |
+
"learning_rate": 2.0710336499999814e-05,
|
| 884 |
+
"loss": 0.241,
|
| 885 |
+
"step": 625
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.2328767123287672,
|
| 889 |
+
"grad_norm": 1.1680022478103638,
|
| 890 |
+
"learning_rate": 2.0560365585676017e-05,
|
| 891 |
+
"loss": 0.2347,
|
| 892 |
+
"step": 630
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.2426614481409002,
|
| 896 |
+
"grad_norm": 1.3656913042068481,
|
| 897 |
+
"learning_rate": 2.0409747504687987e-05,
|
| 898 |
+
"loss": 0.2659,
|
| 899 |
+
"step": 635
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.2524461839530332,
|
| 903 |
+
"grad_norm": 1.0687566995620728,
|
| 904 |
+
"learning_rate": 2.0258499787358067e-05,
|
| 905 |
+
"loss": 0.2321,
|
| 906 |
+
"step": 640
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.2622309197651664,
|
| 910 |
+
"grad_norm": 1.3821700811386108,
|
| 911 |
+
"learning_rate": 2.010664003729149e-05,
|
| 912 |
+
"loss": 0.2437,
|
| 913 |
+
"step": 645
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.2720156555772995,
|
| 917 |
+
"grad_norm": 1.1575123071670532,
|
| 918 |
+
"learning_rate": 1.995418592932751e-05,
|
| 919 |
+
"loss": 0.2356,
|
| 920 |
+
"step": 650
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.2818003913894325,
|
| 924 |
+
"grad_norm": 1.2188833951950073,
|
| 925 |
+
"learning_rate": 1.9801155207482218e-05,
|
| 926 |
+
"loss": 0.2491,
|
| 927 |
+
"step": 655
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.2915851272015655,
|
| 931 |
+
"grad_norm": 1.2536076307296753,
|
| 932 |
+
"learning_rate": 1.9647565682883372e-05,
|
| 933 |
+
"loss": 0.207,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.3013698630136985,
|
| 938 |
+
"grad_norm": 1.0576915740966797,
|
| 939 |
+
"learning_rate": 1.9493435231697333e-05,
|
| 940 |
+
"loss": 0.2472,
|
| 941 |
+
"step": 665
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.3111545988258317,
|
| 945 |
+
"grad_norm": 1.011817455291748,
|
| 946 |
+
"learning_rate": 1.933878179304852e-05,
|
| 947 |
+
"loss": 0.2189,
|
| 948 |
+
"step": 670
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.3209393346379648,
|
| 952 |
+
"grad_norm": 1.4796313047409058,
|
| 953 |
+
"learning_rate": 1.918362336693142e-05,
|
| 954 |
+
"loss": 0.2223,
|
| 955 |
+
"step": 675
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.3307240704500978,
|
| 959 |
+
"grad_norm": 1.1362252235412598,
|
| 960 |
+
"learning_rate": 1.9027978012115652e-05,
|
| 961 |
+
"loss": 0.2007,
|
| 962 |
+
"step": 680
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.340508806262231,
|
| 966 |
+
"grad_norm": 1.2390251159667969,
|
| 967 |
+
"learning_rate": 1.8871863844044083e-05,
|
| 968 |
+
"loss": 0.2248,
|
| 969 |
+
"step": 685
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.350293542074364,
|
| 973 |
+
"grad_norm": 1.0809073448181152,
|
| 974 |
+
"learning_rate": 1.871529903272439e-05,
|
| 975 |
+
"loss": 0.2086,
|
| 976 |
+
"step": 690
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.360078277886497,
|
| 980 |
+
"grad_norm": 1.238761305809021,
|
| 981 |
+
"learning_rate": 1.8558301800614272e-05,
|
| 982 |
+
"loss": 0.1983,
|
| 983 |
+
"step": 695
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.36986301369863,
|
| 987 |
+
"grad_norm": 1.0757040977478027,
|
| 988 |
+
"learning_rate": 1.8400890420500546e-05,
|
| 989 |
+
"loss": 0.2186,
|
| 990 |
+
"step": 700
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.379647749510763,
|
| 994 |
+
"grad_norm": 1.2721891403198242,
|
| 995 |
+
"learning_rate": 1.8243083213372412e-05,
|
| 996 |
+
"loss": 0.2002,
|
| 997 |
+
"step": 705
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.3894324853228963,
|
| 1001 |
+
"grad_norm": 1.081621766090393,
|
| 1002 |
+
"learning_rate": 1.808489854628905e-05,
|
| 1003 |
+
"loss": 0.2161,
|
| 1004 |
+
"step": 710
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.3992172211350293,
|
| 1008 |
+
"grad_norm": 1.2029180526733398,
|
| 1009 |
+
"learning_rate": 1.7926354830241928e-05,
|
| 1010 |
+
"loss": 0.2186,
|
| 1011 |
+
"step": 715
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.4090019569471623,
|
| 1015 |
+
"grad_norm": 1.0653958320617676,
|
| 1016 |
+
"learning_rate": 1.7767470518011934e-05,
|
| 1017 |
+
"loss": 0.1946,
|
| 1018 |
+
"step": 720
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.4187866927592956,
|
| 1022 |
+
"grad_norm": 1.1065515279769897,
|
| 1023 |
+
"learning_rate": 1.7608264102021693e-05,
|
| 1024 |
+
"loss": 0.2064,
|
| 1025 |
+
"step": 725
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.4285714285714286,
|
| 1029 |
+
"grad_norm": 1.1080173254013062,
|
| 1030 |
+
"learning_rate": 1.744875411218324e-05,
|
| 1031 |
+
"loss": 0.1793,
|
| 1032 |
+
"step": 730
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.4383561643835616,
|
| 1036 |
+
"grad_norm": 1.3316106796264648,
|
| 1037 |
+
"learning_rate": 1.7288959113741334e-05,
|
| 1038 |
+
"loss": 0.1797,
|
| 1039 |
+
"step": 735
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.4481409001956946,
|
| 1043 |
+
"grad_norm": 1.0640560388565063,
|
| 1044 |
+
"learning_rate": 1.712889770511265e-05,
|
| 1045 |
+
"loss": 0.2125,
|
| 1046 |
+
"step": 740
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.4579256360078277,
|
| 1050 |
+
"grad_norm": 1.031713604927063,
|
| 1051 |
+
"learning_rate": 1.6968588515721166e-05,
|
| 1052 |
+
"loss": 0.1917,
|
| 1053 |
+
"step": 745
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.467710371819961,
|
| 1057 |
+
"grad_norm": 1.1430633068084717,
|
| 1058 |
+
"learning_rate": 1.6808050203829845e-05,
|
| 1059 |
+
"loss": 0.1721,
|
| 1060 |
+
"step": 750
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.477495107632094,
|
| 1064 |
+
"grad_norm": 1.0728070735931396,
|
| 1065 |
+
"learning_rate": 1.6647301454369068e-05,
|
| 1066 |
+
"loss": 0.175,
|
| 1067 |
+
"step": 755
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.487279843444227,
|
| 1071 |
+
"grad_norm": 1.2282472848892212,
|
| 1072 |
+
"learning_rate": 1.6486360976761866e-05,
|
| 1073 |
+
"loss": 0.1841,
|
| 1074 |
+
"step": 760
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.4970645792563602,
|
| 1078 |
+
"grad_norm": 1.169406771659851,
|
| 1079 |
+
"learning_rate": 1.632524750274636e-05,
|
| 1080 |
+
"loss": 0.1496,
|
| 1081 |
+
"step": 765
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.5068493150684932,
|
| 1085 |
+
"grad_norm": 1.085239052772522,
|
| 1086 |
+
"learning_rate": 1.6163979784195594e-05,
|
| 1087 |
+
"loss": 0.1893,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.5166340508806262,
|
| 1092 |
+
"grad_norm": 1.020064115524292,
|
| 1093 |
+
"learning_rate": 1.6002576590935022e-05,
|
| 1094 |
+
"loss": 0.1769,
|
| 1095 |
+
"step": 775
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.5264187866927594,
|
| 1099 |
+
"grad_norm": 1.0755733251571655,
|
| 1100 |
+
"learning_rate": 1.5841056708557876e-05,
|
| 1101 |
+
"loss": 0.1691,
|
| 1102 |
+
"step": 780
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.5362035225048922,
|
| 1106 |
+
"grad_norm": 1.065834879875183,
|
| 1107 |
+
"learning_rate": 1.567943893623877e-05,
|
| 1108 |
+
"loss": 0.1785,
|
| 1109 |
+
"step": 785
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.5459882583170255,
|
| 1113 |
+
"grad_norm": 1.0913126468658447,
|
| 1114 |
+
"learning_rate": 1.551774208454562e-05,
|
| 1115 |
+
"loss": 0.1867,
|
| 1116 |
+
"step": 790
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.5557729941291585,
|
| 1120 |
+
"grad_norm": 1.1648215055465698,
|
| 1121 |
+
"learning_rate": 1.5355984973250366e-05,
|
| 1122 |
+
"loss": 0.1907,
|
| 1123 |
+
"step": 795
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.5655577299412915,
|
| 1127 |
+
"grad_norm": 1.4390522241592407,
|
| 1128 |
+
"learning_rate": 1.519418642913849e-05,
|
| 1129 |
+
"loss": 0.163,
|
| 1130 |
+
"step": 800
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.5753424657534247,
|
| 1134 |
+
"grad_norm": 1.01279878616333,
|
| 1135 |
+
"learning_rate": 1.5032365283817825e-05,
|
| 1136 |
+
"loss": 0.1625,
|
| 1137 |
+
"step": 805
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.5851272015655578,
|
| 1141 |
+
"grad_norm": 1.1234655380249023,
|
| 1142 |
+
"learning_rate": 1.4870540371526718e-05,
|
| 1143 |
+
"loss": 0.167,
|
| 1144 |
+
"step": 810
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.5949119373776908,
|
| 1148 |
+
"grad_norm": 1.0970678329467773,
|
| 1149 |
+
"learning_rate": 1.4708730526941975e-05,
|
| 1150 |
+
"loss": 0.1441,
|
| 1151 |
+
"step": 815
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.604696673189824,
|
| 1155 |
+
"grad_norm": 0.9171003103256226,
|
| 1156 |
+
"learning_rate": 1.454695458298667e-05,
|
| 1157 |
+
"loss": 0.1464,
|
| 1158 |
+
"step": 820
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.6144814090019568,
|
| 1162 |
+
"grad_norm": 1.074061393737793,
|
| 1163 |
+
"learning_rate": 1.4385231368638216e-05,
|
| 1164 |
+
"loss": 0.146,
|
| 1165 |
+
"step": 825
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.62426614481409,
|
| 1169 |
+
"grad_norm": 0.9906405806541443,
|
| 1170 |
+
"learning_rate": 1.4223579706736857e-05,
|
| 1171 |
+
"loss": 0.157,
|
| 1172 |
+
"step": 830
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.634050880626223,
|
| 1176 |
+
"grad_norm": 0.9814542531967163,
|
| 1177 |
+
"learning_rate": 1.4062018411794925e-05,
|
| 1178 |
+
"loss": 0.1428,
|
| 1179 |
+
"step": 835
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.643835616438356,
|
| 1183 |
+
"grad_norm": 0.9765394330024719,
|
| 1184 |
+
"learning_rate": 1.3900566287806982e-05,
|
| 1185 |
+
"loss": 0.1421,
|
| 1186 |
+
"step": 840
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.6536203522504893,
|
| 1190 |
+
"grad_norm": 1.194342851638794,
|
| 1191 |
+
"learning_rate": 1.3739242126061288e-05,
|
| 1192 |
+
"loss": 0.1843,
|
| 1193 |
+
"step": 845
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.6634050880626223,
|
| 1197 |
+
"grad_norm": 0.910889744758606,
|
| 1198 |
+
"learning_rate": 1.3578064702952668e-05,
|
| 1199 |
+
"loss": 0.1454,
|
| 1200 |
+
"step": 850
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.6731898238747553,
|
| 1204 |
+
"grad_norm": 1.0159302949905396,
|
| 1205 |
+
"learning_rate": 1.3417052777797151e-05,
|
| 1206 |
+
"loss": 0.1542,
|
| 1207 |
+
"step": 855
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.6829745596868886,
|
| 1211 |
+
"grad_norm": 0.915496289730072,
|
| 1212 |
+
"learning_rate": 1.3256225090648585e-05,
|
| 1213 |
+
"loss": 0.1564,
|
| 1214 |
+
"step": 860
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.6927592954990214,
|
| 1218 |
+
"grad_norm": 0.9835331439971924,
|
| 1219 |
+
"learning_rate": 1.3095600360117506e-05,
|
| 1220 |
+
"loss": 0.1325,
|
| 1221 |
+
"step": 865
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.7025440313111546,
|
| 1225 |
+
"grad_norm": 1.0963622331619263,
|
| 1226 |
+
"learning_rate": 1.293519728119248e-05,
|
| 1227 |
+
"loss": 0.1462,
|
| 1228 |
+
"step": 870
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.7123287671232876,
|
| 1232 |
+
"grad_norm": 1.078162431716919,
|
| 1233 |
+
"learning_rate": 1.2775034523064206e-05,
|
| 1234 |
+
"loss": 0.1378,
|
| 1235 |
+
"step": 875
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.7221135029354206,
|
| 1239 |
+
"grad_norm": 1.0844110250473022,
|
| 1240 |
+
"learning_rate": 1.261513072695265e-05,
|
| 1241 |
+
"loss": 0.1501,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.7318982387475539,
|
| 1246 |
+
"grad_norm": 0.9779272079467773,
|
| 1247 |
+
"learning_rate": 1.2455504503937369e-05,
|
| 1248 |
+
"loss": 0.1358,
|
| 1249 |
+
"step": 885
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.741682974559687,
|
| 1253 |
+
"grad_norm": 0.9694347381591797,
|
| 1254 |
+
"learning_rate": 1.2296174432791415e-05,
|
| 1255 |
+
"loss": 0.1341,
|
| 1256 |
+
"step": 890
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.75146771037182,
|
| 1260 |
+
"grad_norm": 1.135961890220642,
|
| 1261 |
+
"learning_rate": 1.2137159057818942e-05,
|
| 1262 |
+
"loss": 0.1358,
|
| 1263 |
+
"step": 895
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.7612524461839532,
|
| 1267 |
+
"grad_norm": 1.0626426935195923,
|
| 1268 |
+
"learning_rate": 1.1978476886696868e-05,
|
| 1269 |
+
"loss": 0.1271,
|
| 1270 |
+
"step": 900
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.771037181996086,
|
| 1274 |
+
"grad_norm": 0.8748185634613037,
|
| 1275 |
+
"learning_rate": 1.1820146388320764e-05,
|
| 1276 |
+
"loss": 0.132,
|
| 1277 |
+
"step": 905
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.7808219178082192,
|
| 1281 |
+
"grad_norm": 0.8870331048965454,
|
| 1282 |
+
"learning_rate": 1.1662185990655285e-05,
|
| 1283 |
+
"loss": 0.1312,
|
| 1284 |
+
"step": 910
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.7906066536203522,
|
| 1288 |
+
"grad_norm": 1.1639935970306396,
|
| 1289 |
+
"learning_rate": 1.1504614078589353e-05,
|
| 1290 |
+
"loss": 0.1351,
|
| 1291 |
+
"step": 915
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.8003913894324852,
|
| 1295 |
+
"grad_norm": 0.8740416765213013,
|
| 1296 |
+
"learning_rate": 1.1347448991796335e-05,
|
| 1297 |
+
"loss": 0.1148,
|
| 1298 |
+
"step": 920
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.8101761252446185,
|
| 1302 |
+
"grad_norm": 1.1621346473693848,
|
| 1303 |
+
"learning_rate": 1.1190709022599546e-05,
|
| 1304 |
+
"loss": 0.1408,
|
| 1305 |
+
"step": 925
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.8199608610567515,
|
| 1309 |
+
"grad_norm": 0.9756697416305542,
|
| 1310 |
+
"learning_rate": 1.1034412413843166e-05,
|
| 1311 |
+
"loss": 0.1203,
|
| 1312 |
+
"step": 930
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.8297455968688845,
|
| 1316 |
+
"grad_norm": 1.061224341392517,
|
| 1317 |
+
"learning_rate": 1.0878577356769008e-05,
|
| 1318 |
+
"loss": 0.1278,
|
| 1319 |
+
"step": 935
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.8395303326810177,
|
| 1323 |
+
"grad_norm": 0.939208447933197,
|
| 1324 |
+
"learning_rate": 1.0723221988899235e-05,
|
| 1325 |
+
"loss": 0.1263,
|
| 1326 |
+
"step": 940
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.8493150684931505,
|
| 1330 |
+
"grad_norm": 0.83660888671875,
|
| 1331 |
+
"learning_rate": 1.0568364391925357e-05,
|
| 1332 |
+
"loss": 0.1108,
|
| 1333 |
+
"step": 945
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.8590998043052838,
|
| 1337 |
+
"grad_norm": 0.9582011699676514,
|
| 1338 |
+
"learning_rate": 1.041402258960371e-05,
|
| 1339 |
+
"loss": 0.1171,
|
| 1340 |
+
"step": 950
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.8688845401174168,
|
| 1344 |
+
"grad_norm": 0.9707714319229126,
|
| 1345 |
+
"learning_rate": 1.0260214545657688e-05,
|
| 1346 |
+
"loss": 0.1003,
|
| 1347 |
+
"step": 955
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.8786692759295498,
|
| 1351 |
+
"grad_norm": 0.9569882154464722,
|
| 1352 |
+
"learning_rate": 1.0106958161686963e-05,
|
| 1353 |
+
"loss": 0.1144,
|
| 1354 |
+
"step": 960
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.888454011741683,
|
| 1358 |
+
"grad_norm": 0.8673170208930969,
|
| 1359 |
+
"learning_rate": 9.954271275083926e-06,
|
| 1360 |
+
"loss": 0.1238,
|
| 1361 |
+
"step": 965
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.898238747553816,
|
| 1365 |
+
"grad_norm": 1.0494937896728516,
|
| 1366 |
+
"learning_rate": 9.8021716569576e-06,
|
| 1367 |
+
"loss": 0.1172,
|
| 1368 |
+
"step": 970
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.908023483365949,
|
| 1372 |
+
"grad_norm": 1.2174538373947144,
|
| 1373 |
+
"learning_rate": 9.650677010065305e-06,
|
| 1374 |
+
"loss": 0.1143,
|
| 1375 |
+
"step": 975
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.9178082191780823,
|
| 1379 |
+
"grad_norm": 0.9288437962532043,
|
| 1380 |
+
"learning_rate": 9.4998049667522e-06,
|
| 1381 |
+
"loss": 0.1049,
|
| 1382 |
+
"step": 980
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.927592954990215,
|
| 1386 |
+
"grad_norm": 0.9887030124664307,
|
| 1387 |
+
"learning_rate": 9.349573086899111e-06,
|
| 1388 |
+
"loss": 0.1045,
|
| 1389 |
+
"step": 985
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.9373776908023483,
|
| 1393 |
+
"grad_norm": 1.1209849119186401,
|
| 1394 |
+
"learning_rate": 9.199998855878719e-06,
|
| 1395 |
+
"loss": 0.114,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.9471624266144814,
|
| 1400 |
+
"grad_norm": 0.9360877871513367,
|
| 1401 |
+
"learning_rate": 9.051099682520475e-06,
|
| 1402 |
+
"loss": 0.1134,
|
| 1403 |
+
"step": 995
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.9569471624266144,
|
| 1407 |
+
"grad_norm": 0.776114284992218,
|
| 1408 |
+
"learning_rate": 8.902892897084382e-06,
|
| 1409 |
+
"loss": 0.1056,
|
| 1410 |
+
"step": 1000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.9667318982387476,
|
| 1414 |
+
"grad_norm": 0.9445653557777405,
|
| 1415 |
+
"learning_rate": 8.755395749243958e-06,
|
| 1416 |
+
"loss": 0.0993,
|
| 1417 |
+
"step": 1005
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.9765166340508806,
|
| 1421 |
+
"grad_norm": 1.2114028930664062,
|
| 1422 |
+
"learning_rate": 8.608625406078526e-06,
|
| 1423 |
+
"loss": 0.1092,
|
| 1424 |
+
"step": 1010
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.9863013698630136,
|
| 1428 |
+
"grad_norm": 1.0555222034454346,
|
| 1429 |
+
"learning_rate": 8.46259895007519e-06,
|
| 1430 |
+
"loss": 0.0942,
|
| 1431 |
+
"step": 1015
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.9960861056751469,
|
| 1435 |
+
"grad_norm": 0.8842493891716003,
|
| 1436 |
+
"learning_rate": 8.317333377140606e-06,
|
| 1437 |
+
"loss": 0.106,
|
| 1438 |
+
"step": 1020
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 2.0058708414872797,
|
| 1442 |
+
"grad_norm": 0.6439769864082336,
|
| 1443 |
+
"learning_rate": 8.172845594622825e-06,
|
| 1444 |
+
"loss": 0.0835,
|
| 1445 |
+
"step": 1025
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 2.015655577299413,
|
| 1449 |
+
"grad_norm": 0.8429094552993774,
|
| 1450 |
+
"learning_rate": 8.029152419343472e-06,
|
| 1451 |
+
"loss": 0.0847,
|
| 1452 |
+
"step": 1030
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 2.025440313111546,
|
| 1456 |
+
"grad_norm": 0.7041258811950684,
|
| 1457 |
+
"learning_rate": 7.88627057564045e-06,
|
| 1458 |
+
"loss": 0.0769,
|
| 1459 |
+
"step": 1035
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 2.035225048923679,
|
| 1463 |
+
"grad_norm": 0.8132873773574829,
|
| 1464 |
+
"learning_rate": 7.744216693421403e-06,
|
| 1465 |
+
"loss": 0.0825,
|
| 1466 |
+
"step": 1040
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 2.045009784735812,
|
| 1470 |
+
"grad_norm": 0.7724251747131348,
|
| 1471 |
+
"learning_rate": 7.603007306228181e-06,
|
| 1472 |
+
"loss": 0.0923,
|
| 1473 |
+
"step": 1045
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 2.0547945205479454,
|
| 1477 |
+
"grad_norm": 0.881170392036438,
|
| 1478 |
+
"learning_rate": 7.462658849312507e-06,
|
| 1479 |
+
"loss": 0.0783,
|
| 1480 |
+
"step": 1050
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 2.064579256360078,
|
| 1484 |
+
"grad_norm": 0.7710109949111938,
|
| 1485 |
+
"learning_rate": 7.3231876577230875e-06,
|
| 1486 |
+
"loss": 0.0815,
|
| 1487 |
+
"step": 1055
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 2.0743639921722115,
|
| 1491 |
+
"grad_norm": 0.9433515667915344,
|
| 1492 |
+
"learning_rate": 7.1846099644044006e-06,
|
| 1493 |
+
"loss": 0.0867,
|
| 1494 |
+
"step": 1060
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 2.0841487279843443,
|
| 1498 |
+
"grad_norm": 0.7743017077445984,
|
| 1499 |
+
"learning_rate": 7.046941898307346e-06,
|
| 1500 |
+
"loss": 0.0769,
|
| 1501 |
+
"step": 1065
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 2.0939334637964775,
|
| 1505 |
+
"grad_norm": 0.847722053527832,
|
| 1506 |
+
"learning_rate": 6.9101994825120144e-06,
|
| 1507 |
+
"loss": 0.0802,
|
| 1508 |
+
"step": 1070
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 2.1037181996086107,
|
| 1512 |
+
"grad_norm": 0.6796703934669495,
|
| 1513 |
+
"learning_rate": 6.774398632362769e-06,
|
| 1514 |
+
"loss": 0.0838,
|
| 1515 |
+
"step": 1075
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 2.1135029354207435,
|
| 1519 |
+
"grad_norm": 0.6822458505630493,
|
| 1520 |
+
"learning_rate": 6.639555153615878e-06,
|
| 1521 |
+
"loss": 0.0854,
|
| 1522 |
+
"step": 1080
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 2.1232876712328768,
|
| 1526 |
+
"grad_norm": 0.7536410093307495,
|
| 1527 |
+
"learning_rate": 6.505684740599881e-06,
|
| 1528 |
+
"loss": 0.0707,
|
| 1529 |
+
"step": 1085
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 2.1330724070450096,
|
| 1533 |
+
"grad_norm": 0.6132102608680725,
|
| 1534 |
+
"learning_rate": 6.3728029743889586e-06,
|
| 1535 |
+
"loss": 0.08,
|
| 1536 |
+
"step": 1090
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 2.142857142857143,
|
| 1540 |
+
"grad_norm": 0.8666600584983826,
|
| 1541 |
+
"learning_rate": 6.240925320989441e-06,
|
| 1542 |
+
"loss": 0.0902,
|
| 1543 |
+
"step": 1095
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 2.152641878669276,
|
| 1547 |
+
"grad_norm": 0.8916414380073547,
|
| 1548 |
+
"learning_rate": 6.11006712953975e-06,
|
| 1549 |
+
"loss": 0.0834,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 2.162426614481409,
|
| 1554 |
+
"grad_norm": 0.696893572807312,
|
| 1555 |
+
"learning_rate": 5.980243630523911e-06,
|
| 1556 |
+
"loss": 0.0714,
|
| 1557 |
+
"step": 1105
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 2.172211350293542,
|
| 1561 |
+
"grad_norm": 0.8130199313163757,
|
| 1562 |
+
"learning_rate": 5.851469933998907e-06,
|
| 1563 |
+
"loss": 0.0823,
|
| 1564 |
+
"step": 1110
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 2.1819960861056753,
|
| 1568 |
+
"grad_norm": 0.6856159567832947,
|
| 1569 |
+
"learning_rate": 5.723761027836002e-06,
|
| 1570 |
+
"loss": 0.0832,
|
| 1571 |
+
"step": 1115
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 2.191780821917808,
|
| 1575 |
+
"grad_norm": 0.6900305151939392,
|
| 1576 |
+
"learning_rate": 5.597131775976343e-06,
|
| 1577 |
+
"loss": 0.0797,
|
| 1578 |
+
"step": 1120
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 2.2015655577299413,
|
| 1582 |
+
"grad_norm": 1.0851941108703613,
|
| 1583 |
+
"learning_rate": 5.471596916700933e-06,
|
| 1584 |
+
"loss": 0.071,
|
| 1585 |
+
"step": 1125
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 2.2113502935420746,
|
| 1589 |
+
"grad_norm": 0.7255151867866516,
|
| 1590 |
+
"learning_rate": 5.347171060915278e-06,
|
| 1591 |
+
"loss": 0.0683,
|
| 1592 |
+
"step": 1130
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.2211350293542074,
|
| 1596 |
+
"grad_norm": 0.8086680173873901,
|
| 1597 |
+
"learning_rate": 5.223868690448817e-06,
|
| 1598 |
+
"loss": 0.0815,
|
| 1599 |
+
"step": 1135
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 2.2309197651663406,
|
| 1603 |
+
"grad_norm": 0.7399266958236694,
|
| 1604 |
+
"learning_rate": 5.101704156369398e-06,
|
| 1605 |
+
"loss": 0.0763,
|
| 1606 |
+
"step": 1140
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 2.2407045009784734,
|
| 1610 |
+
"grad_norm": 0.6829865574836731,
|
| 1611 |
+
"learning_rate": 4.980691677312974e-06,
|
| 1612 |
+
"loss": 0.0782,
|
| 1613 |
+
"step": 1145
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 2.2504892367906066,
|
| 1617 |
+
"grad_norm": 0.7899296879768372,
|
| 1618 |
+
"learning_rate": 4.860845337828684e-06,
|
| 1619 |
+
"loss": 0.0826,
|
| 1620 |
+
"step": 1150
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 2.26027397260274,
|
| 1624 |
+
"grad_norm": 0.9397788643836975,
|
| 1625 |
+
"learning_rate": 4.742179086739612e-06,
|
| 1626 |
+
"loss": 0.0732,
|
| 1627 |
+
"step": 1155
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 2.2700587084148727,
|
| 1631 |
+
"grad_norm": 0.5895610451698303,
|
| 1632 |
+
"learning_rate": 4.624706735519239e-06,
|
| 1633 |
+
"loss": 0.0719,
|
| 1634 |
+
"step": 1160
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.279843444227006,
|
| 1638 |
+
"grad_norm": 0.903843879699707,
|
| 1639 |
+
"learning_rate": 4.50844195668397e-06,
|
| 1640 |
+
"loss": 0.0656,
|
| 1641 |
+
"step": 1165
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 2.2896281800391387,
|
| 1645 |
+
"grad_norm": 0.8454123139381409,
|
| 1646 |
+
"learning_rate": 4.393398282201788e-06,
|
| 1647 |
+
"loss": 0.0772,
|
| 1648 |
+
"step": 1170
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 2.299412915851272,
|
| 1652 |
+
"grad_norm": 0.6164603233337402,
|
| 1653 |
+
"learning_rate": 4.279589101917285e-06,
|
| 1654 |
+
"loss": 0.0686,
|
| 1655 |
+
"step": 1175
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 2.309197651663405,
|
| 1659 |
+
"grad_norm": 0.5521309971809387,
|
| 1660 |
+
"learning_rate": 4.1670276619932185e-06,
|
| 1661 |
+
"loss": 0.0704,
|
| 1662 |
+
"step": 1180
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 2.318982387475538,
|
| 1666 |
+
"grad_norm": 0.8986679911613464,
|
| 1667 |
+
"learning_rate": 4.0557270633688155e-06,
|
| 1668 |
+
"loss": 0.0766,
|
| 1669 |
+
"step": 1185
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 2.328767123287671,
|
| 1673 |
+
"grad_norm": 0.7122253775596619,
|
| 1674 |
+
"learning_rate": 3.94570026023495e-06,
|
| 1675 |
+
"loss": 0.0729,
|
| 1676 |
+
"step": 1190
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.3385518590998045,
|
| 1680 |
+
"grad_norm": 0.6210092306137085,
|
| 1681 |
+
"learning_rate": 3.83696005852642e-06,
|
| 1682 |
+
"loss": 0.068,
|
| 1683 |
+
"step": 1195
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.3483365949119372,
|
| 1687 |
+
"grad_norm": 0.8375857472419739,
|
| 1688 |
+
"learning_rate": 3.7295191144314914e-06,
|
| 1689 |
+
"loss": 0.0693,
|
| 1690 |
+
"step": 1200
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 2.3581213307240705,
|
| 1694 |
+
"grad_norm": 0.6256863474845886,
|
| 1695 |
+
"learning_rate": 3.623389932918811e-06,
|
| 1696 |
+
"loss": 0.0702,
|
| 1697 |
+
"step": 1205
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.3679060665362037,
|
| 1701 |
+
"grad_norm": 0.6554362773895264,
|
| 1702 |
+
"learning_rate": 3.518584866282009e-06,
|
| 1703 |
+
"loss": 0.0681,
|
| 1704 |
+
"step": 1210
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.3776908023483365,
|
| 1708 |
+
"grad_norm": 0.7649148106575012,
|
| 1709 |
+
"learning_rate": 3.4151161127020003e-06,
|
| 1710 |
+
"loss": 0.0662,
|
| 1711 |
+
"step": 1215
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.3874755381604698,
|
| 1715 |
+
"grad_norm": 0.6647529006004333,
|
| 1716 |
+
"learning_rate": 3.3129957148272532e-06,
|
| 1717 |
+
"loss": 0.0777,
|
| 1718 |
+
"step": 1220
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.3972602739726026,
|
| 1722 |
+
"grad_norm": 0.6368157267570496,
|
| 1723 |
+
"learning_rate": 3.2122355583721557e-06,
|
| 1724 |
+
"loss": 0.0669,
|
| 1725 |
+
"step": 1225
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.407045009784736,
|
| 1729 |
+
"grad_norm": 0.5931635499000549,
|
| 1730 |
+
"learning_rate": 3.1128473707336463e-06,
|
| 1731 |
+
"loss": 0.0699,
|
| 1732 |
+
"step": 1230
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.416829745596869,
|
| 1736 |
+
"grad_norm": 0.6239747405052185,
|
| 1737 |
+
"learning_rate": 3.0148427196262678e-06,
|
| 1738 |
+
"loss": 0.0703,
|
| 1739 |
+
"step": 1235
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.426614481409002,
|
| 1743 |
+
"grad_norm": 0.7887265086174011,
|
| 1744 |
+
"learning_rate": 2.9182330117358098e-06,
|
| 1745 |
+
"loss": 0.077,
|
| 1746 |
+
"step": 1240
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 2.436399217221135,
|
| 1750 |
+
"grad_norm": 0.6641157865524292,
|
| 1751 |
+
"learning_rate": 2.823029491391711e-06,
|
| 1752 |
+
"loss": 0.0739,
|
| 1753 |
+
"step": 1245
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 2.446183953033268,
|
| 1757 |
+
"grad_norm": 0.5395591259002686,
|
| 1758 |
+
"learning_rate": 2.7292432392583077e-06,
|
| 1759 |
+
"loss": 0.0735,
|
| 1760 |
+
"step": 1250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.455968688845401,
|
| 1764 |
+
"grad_norm": 0.6540674567222595,
|
| 1765 |
+
"learning_rate": 2.636885171045193e-06,
|
| 1766 |
+
"loss": 0.0723,
|
| 1767 |
+
"step": 1255
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 2.4657534246575343,
|
| 1771 |
+
"grad_norm": 0.7001171112060547,
|
| 1772 |
+
"learning_rate": 2.545966036236728e-06,
|
| 1773 |
+
"loss": 0.0682,
|
| 1774 |
+
"step": 1260
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.475538160469667,
|
| 1778 |
+
"grad_norm": 1.1719061136245728,
|
| 1779 |
+
"learning_rate": 2.456496416840928e-06,
|
| 1780 |
+
"loss": 0.0689,
|
| 1781 |
+
"step": 1265
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.4853228962818004,
|
| 1785 |
+
"grad_norm": 0.655920147895813,
|
| 1786 |
+
"learning_rate": 2.3684867261578236e-06,
|
| 1787 |
+
"loss": 0.0805,
|
| 1788 |
+
"step": 1270
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.4951076320939336,
|
| 1792 |
+
"grad_norm": 0.621269941329956,
|
| 1793 |
+
"learning_rate": 2.281947207567473e-06,
|
| 1794 |
+
"loss": 0.068,
|
| 1795 |
+
"step": 1275
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.5048923679060664,
|
| 1799 |
+
"grad_norm": 0.6041397452354431,
|
| 1800 |
+
"learning_rate": 2.196887933337733e-06,
|
| 1801 |
+
"loss": 0.0621,
|
| 1802 |
+
"step": 1280
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.5146771037181996,
|
| 1806 |
+
"grad_norm": 0.672671914100647,
|
| 1807 |
+
"learning_rate": 2.1133188034519597e-06,
|
| 1808 |
+
"loss": 0.0693,
|
| 1809 |
+
"step": 1285
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.524461839530333,
|
| 1813 |
+
"grad_norm": 0.6204409003257751,
|
| 1814 |
+
"learning_rate": 2.0312495444567645e-06,
|
| 1815 |
+
"loss": 0.0556,
|
| 1816 |
+
"step": 1290
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.5342465753424657,
|
| 1820 |
+
"grad_norm": 0.5381624698638916,
|
| 1821 |
+
"learning_rate": 1.9506897083299335e-06,
|
| 1822 |
+
"loss": 0.0717,
|
| 1823 |
+
"step": 1295
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.544031311154599,
|
| 1827 |
+
"grad_norm": 0.6148281693458557,
|
| 1828 |
+
"learning_rate": 1.8716486713686949e-06,
|
| 1829 |
+
"loss": 0.067,
|
| 1830 |
+
"step": 1300
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.553816046966732,
|
| 1834 |
+
"grad_norm": 0.6378031373023987,
|
| 1835 |
+
"learning_rate": 1.7941356330984116e-06,
|
| 1836 |
+
"loss": 0.0654,
|
| 1837 |
+
"step": 1305
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.563600782778865,
|
| 1841 |
+
"grad_norm": 0.4585803747177124,
|
| 1842 |
+
"learning_rate": 1.718159615201853e-06,
|
| 1843 |
+
"loss": 0.0693,
|
| 1844 |
+
"step": 1310
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.573385518590998,
|
| 1848 |
+
"grad_norm": 0.6192830204963684,
|
| 1849 |
+
"learning_rate": 1.643729460469176e-06,
|
| 1850 |
+
"loss": 0.0602,
|
| 1851 |
+
"step": 1315
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.583170254403131,
|
| 1855 |
+
"grad_norm": 0.5830395817756653,
|
| 1856 |
+
"learning_rate": 1.5708538317687148e-06,
|
| 1857 |
+
"loss": 0.065,
|
| 1858 |
+
"step": 1320
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.592954990215264,
|
| 1862 |
+
"grad_norm": 0.5701336860656738,
|
| 1863 |
+
"learning_rate": 1.4995412110387163e-06,
|
| 1864 |
+
"loss": 0.061,
|
| 1865 |
+
"step": 1325
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.602739726027397,
|
| 1869 |
+
"grad_norm": 0.4947655200958252,
|
| 1870 |
+
"learning_rate": 1.4297998983001327e-06,
|
| 1871 |
+
"loss": 0.0636,
|
| 1872 |
+
"step": 1330
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.6125244618395302,
|
| 1876 |
+
"grad_norm": 0.7686696648597717,
|
| 1877 |
+
"learning_rate": 1.3616380106905974e-06,
|
| 1878 |
+
"loss": 0.0698,
|
| 1879 |
+
"step": 1335
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.6223091976516635,
|
| 1883 |
+
"grad_norm": 0.523448646068573,
|
| 1884 |
+
"learning_rate": 1.2950634815196633e-06,
|
| 1885 |
+
"loss": 0.0747,
|
| 1886 |
+
"step": 1340
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.6320939334637963,
|
| 1890 |
+
"grad_norm": 0.47885662317276,
|
| 1891 |
+
"learning_rate": 1.2300840593454621e-06,
|
| 1892 |
+
"loss": 0.0604,
|
| 1893 |
+
"step": 1345
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.6418786692759295,
|
| 1897 |
+
"grad_norm": 0.6009384989738464,
|
| 1898 |
+
"learning_rate": 1.1667073070728462e-06,
|
| 1899 |
+
"loss": 0.0679,
|
| 1900 |
+
"step": 1350
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.6516634050880628,
|
| 1904 |
+
"grad_norm": 0.43924564123153687,
|
| 1905 |
+
"learning_rate": 1.104940601073161e-06,
|
| 1906 |
+
"loss": 0.0585,
|
| 1907 |
+
"step": 1355
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.6614481409001955,
|
| 1911 |
+
"grad_norm": 0.5072720646858215,
|
| 1912 |
+
"learning_rate": 1.044791130325702e-06,
|
| 1913 |
+
"loss": 0.0672,
|
| 1914 |
+
"step": 1360
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.671232876712329,
|
| 1918 |
+
"grad_norm": 0.5864505171775818,
|
| 1919 |
+
"learning_rate": 9.862658955810045e-07,
|
| 1920 |
+
"loss": 0.07,
|
| 1921 |
+
"step": 1365
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.681017612524462,
|
| 1925 |
+
"grad_norm": 0.6745476722717285,
|
| 1926 |
+
"learning_rate": 9.293717085460223e-07,
|
| 1927 |
+
"loss": 0.0678,
|
| 1928 |
+
"step": 1370
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.690802348336595,
|
| 1932 |
+
"grad_norm": 0.6452853679656982,
|
| 1933 |
+
"learning_rate": 8.741151910913276e-07,
|
| 1934 |
+
"loss": 0.0602,
|
| 1935 |
+
"step": 1375
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.700587084148728,
|
| 1939 |
+
"grad_norm": 0.5218824148178101,
|
| 1940 |
+
"learning_rate": 8.20502774480395e-07,
|
| 1941 |
+
"loss": 0.0657,
|
| 1942 |
+
"step": 1380
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.7103718199608613,
|
| 1946 |
+
"grad_norm": 0.47543880343437195,
|
| 1947 |
+
"learning_rate": 7.685406986210653e-07,
|
| 1948 |
+
"loss": 0.061,
|
| 1949 |
+
"step": 1385
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.720156555772994,
|
| 1953 |
+
"grad_norm": 0.48542219400405884,
|
| 1954 |
+
"learning_rate": 7.182350113392883e-07,
|
| 1955 |
+
"loss": 0.055,
|
| 1956 |
+
"step": 1390
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.7299412915851273,
|
| 1960 |
+
"grad_norm": 0.5949373245239258,
|
| 1961 |
+
"learning_rate": 6.695915676752301e-07,
|
| 1962 |
+
"loss": 0.0667,
|
| 1963 |
+
"step": 1395
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.73972602739726,
|
| 1967 |
+
"grad_norm": 0.553185760974884,
|
| 1968 |
+
"learning_rate": 6.226160292018013e-07,
|
| 1969 |
+
"loss": 0.0639,
|
| 1970 |
+
"step": 1400
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.7495107632093934,
|
| 1974 |
+
"grad_norm": 0.6269693374633789,
|
| 1975 |
+
"learning_rate": 5.773138633657121e-07,
|
| 1976 |
+
"loss": 0.0629,
|
| 1977 |
+
"step": 1405
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.759295499021526,
|
| 1981 |
+
"grad_norm": 0.5973682999610901,
|
| 1982 |
+
"learning_rate": 5.336903428511236e-07,
|
| 1983 |
+
"loss": 0.0658,
|
| 1984 |
+
"step": 1410
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.7690802348336594,
|
| 1988 |
+
"grad_norm": 0.5076714754104614,
|
| 1989 |
+
"learning_rate": 4.917505449659615e-07,
|
| 1990 |
+
"loss": 0.0568,
|
| 1991 |
+
"step": 1415
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.7788649706457926,
|
| 1995 |
+
"grad_norm": 0.5836893320083618,
|
| 1996 |
+
"learning_rate": 4.514993510509674e-07,
|
| 1997 |
+
"loss": 0.0665,
|
| 1998 |
+
"step": 1420
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.7886497064579254,
|
| 2002 |
+
"grad_norm": 0.44945862889289856,
|
| 2003 |
+
"learning_rate": 4.129414459115782e-07,
|
| 2004 |
+
"loss": 0.0592,
|
| 2005 |
+
"step": 1425
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.7984344422700587,
|
| 2009 |
+
"grad_norm": 0.5773463845252991,
|
| 2010 |
+
"learning_rate": 3.760813172726457e-07,
|
| 2011 |
+
"loss": 0.0623,
|
| 2012 |
+
"step": 1430
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.808219178082192,
|
| 2016 |
+
"grad_norm": 0.6719569563865662,
|
| 2017 |
+
"learning_rate": 3.4092325525613087e-07,
|
| 2018 |
+
"loss": 0.0606,
|
| 2019 |
+
"step": 1435
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.8180039138943247,
|
| 2023 |
+
"grad_norm": 0.6693986058235168,
|
| 2024 |
+
"learning_rate": 3.0747135188176846e-07,
|
| 2025 |
+
"loss": 0.0826,
|
| 2026 |
+
"step": 1440
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.827788649706458,
|
| 2030 |
+
"grad_norm": 0.5172216892242432,
|
| 2031 |
+
"learning_rate": 2.7572950059080615e-07,
|
| 2032 |
+
"loss": 0.0581,
|
| 2033 |
+
"step": 1445
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.837573385518591,
|
| 2037 |
+
"grad_norm": 0.6774749755859375,
|
| 2038 |
+
"learning_rate": 2.457013957928472e-07,
|
| 2039 |
+
"loss": 0.0555,
|
| 2040 |
+
"step": 1450
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.847358121330724,
|
| 2044 |
+
"grad_norm": 0.49793991446495056,
|
| 2045 |
+
"learning_rate": 2.1739053243585937e-07,
|
| 2046 |
+
"loss": 0.0559,
|
| 2047 |
+
"step": 1455
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.857142857142857,
|
| 2051 |
+
"grad_norm": 0.6323301196098328,
|
| 2052 |
+
"learning_rate": 1.908002055994068e-07,
|
| 2053 |
+
"loss": 0.0672,
|
| 2054 |
+
"step": 1460
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.8669275929549904,
|
| 2058 |
+
"grad_norm": 0.5901110172271729,
|
| 2059 |
+
"learning_rate": 1.65933510111132e-07,
|
| 2060 |
+
"loss": 0.0585,
|
| 2061 |
+
"step": 1465
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.8767123287671232,
|
| 2065 |
+
"grad_norm": 0.6155837178230286,
|
| 2066 |
+
"learning_rate": 1.4279334018655222e-07,
|
| 2067 |
+
"loss": 0.0651,
|
| 2068 |
+
"step": 1470
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.8864970645792565,
|
| 2072 |
+
"grad_norm": 0.5353571176528931,
|
| 2073 |
+
"learning_rate": 1.2138238909221055e-07,
|
| 2074 |
+
"loss": 0.0607,
|
| 2075 |
+
"step": 1475
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.8962818003913893,
|
| 2079 |
+
"grad_norm": 0.5011776685714722,
|
| 2080 |
+
"learning_rate": 1.0170314883219678e-07,
|
| 2081 |
+
"loss": 0.0712,
|
| 2082 |
+
"step": 1480
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.9060665362035225,
|
| 2086 |
+
"grad_norm": 0.5339565873146057,
|
| 2087 |
+
"learning_rate": 8.375790985811759e-08,
|
| 2088 |
+
"loss": 0.0633,
|
| 2089 |
+
"step": 1485
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.9158512720156553,
|
| 2093 |
+
"grad_norm": 0.6261019706726074,
|
| 2094 |
+
"learning_rate": 6.754876080250161e-08,
|
| 2095 |
+
"loss": 0.0682,
|
| 2096 |
+
"step": 1490
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.9256360078277885,
|
| 2100 |
+
"grad_norm": 0.5683228969573975,
|
| 2101 |
+
"learning_rate": 5.307758823571374e-08,
|
| 2102 |
+
"loss": 0.0651,
|
| 2103 |
+
"step": 1495
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.935420743639922,
|
| 2107 |
+
"grad_norm": 0.567682147026062,
|
| 2108 |
+
"learning_rate": 4.034607644637089e-08,
|
| 2109 |
+
"loss": 0.0696,
|
| 2110 |
+
"step": 1500
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.9452054794520546,
|
| 2114 |
+
"grad_norm": 0.6584129333496094,
|
| 2115 |
+
"learning_rate": 2.935570724532044e-08,
|
| 2116 |
+
"loss": 0.0612,
|
| 2117 |
+
"step": 1505
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.954990215264188,
|
| 2121 |
+
"grad_norm": 0.5677846670150757,
|
| 2122 |
+
"learning_rate": 2.0107759793158154e-08,
|
| 2123 |
+
"loss": 0.0675,
|
| 2124 |
+
"step": 1510
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.964774951076321,
|
| 2128 |
+
"grad_norm": 0.5319307446479797,
|
| 2129 |
+
"learning_rate": 1.2603310451358984e-08,
|
| 2130 |
+
"loss": 0.0611,
|
| 2131 |
+
"step": 1515
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.974559686888454,
|
| 2135 |
+
"grad_norm": 0.6892783045768738,
|
| 2136 |
+
"learning_rate": 6.8432326569989325e-09,
|
| 2137 |
+
"loss": 0.0712,
|
| 2138 |
+
"step": 1520
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.984344422700587,
|
| 2142 |
+
"grad_norm": 0.6270133256912231,
|
| 2143 |
+
"learning_rate": 2.8281968210913665e-09,
|
| 2144 |
+
"loss": 0.0602,
|
| 2145 |
+
"step": 1525
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.9941291585127203,
|
| 2149 |
+
"grad_norm": 0.6447855234146118,
|
| 2150 |
+
"learning_rate": 5.586702505627761e-10,
|
| 2151 |
+
"loss": 0.0703,
|
| 2152 |
+
"step": 1530
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 3.0,
|
| 2156 |
+
"step": 1533,
|
| 2157 |
+
"total_flos": 2.0550851304557117e+18,
|
| 2158 |
+
"train_loss": 0.3431719825186919,
|
| 2159 |
+
"train_runtime": 983.7331,
|
| 2160 |
+
"train_samples_per_second": 49.852,
|
| 2161 |
+
"train_steps_per_second": 1.558
|
| 2162 |
+
}
|
| 2163 |
+
],
|
| 2164 |
+
"logging_steps": 5,
|
| 2165 |
+
"max_steps": 1533,
|
| 2166 |
+
"num_input_tokens_seen": 0,
|
| 2167 |
+
"num_train_epochs": 3,
|
| 2168 |
+
"save_steps": 20000,
|
| 2169 |
+
"stateful_callbacks": {
|
| 2170 |
+
"TrainerControl": {
|
| 2171 |
+
"args": {
|
| 2172 |
+
"should_epoch_stop": false,
|
| 2173 |
+
"should_evaluate": false,
|
| 2174 |
+
"should_log": false,
|
| 2175 |
+
"should_save": false,
|
| 2176 |
+
"should_training_stop": false
|
| 2177 |
+
},
|
| 2178 |
+
"attributes": {}
|
| 2179 |
+
}
|
| 2180 |
+
},
|
| 2181 |
+
"total_flos": 2.0550851304557117e+18,
|
| 2182 |
+
"train_batch_size": 2,
|
| 2183 |
+
"trial_name": null,
|
| 2184 |
+
"trial_params": null
|
| 2185 |
+
}
|
136_128_e3_3e-5/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b7bf4cce4d7a8d2b79e3170e563270d6585726d22dcee90c1f435d146d47b627
|
| 3 |
+
size 8273
|
136_128_e3_3e-5/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|