Upload folder using huggingface_hub
Browse files- chat_template.jinja +16 -0
- checkpoint-17314/config.json +32 -0
- checkpoint-17314/generation_config.json +12 -0
- checkpoint-17314/model.safetensors +3 -0
- checkpoint-17314/optimizer.pt +3 -0
- checkpoint-17314/rng_state.pth +3 -0
- checkpoint-17314/scheduler.pt +3 -0
- checkpoint-17314/trainer_state.json +2464 -0
- checkpoint-17314/training_args.bin +3 -0
- config.json +32 -0
- generation_config.json +12 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +10 -0
- training_args.bin +3 -0
chat_template.jinja
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{% if message['role'] == 'user' %}{% if message.get('input') not in [None, '', 'none', 'nan'] %}Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
|
| 2 |
+
|
| 3 |
+
### Instruction:
|
| 4 |
+
{{ message['content'] }}
|
| 5 |
+
|
| 6 |
+
### Input:
|
| 7 |
+
{{ message['input'] }}
|
| 8 |
+
|
| 9 |
+
### Response:
|
| 10 |
+
{% else %}Below is an instruction that describes a task. Write a response that appropriately completes the request.
|
| 11 |
+
|
| 12 |
+
### Instruction:
|
| 13 |
+
{{ message['content'] }}
|
| 14 |
+
|
| 15 |
+
### Response:
|
| 16 |
+
{% endif %}{% elif message['role'] == 'assistant' %}{{ message['content'] }}{% endif %}{% endfor %}
|
checkpoint-17314/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 3,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 1,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 2304,
|
| 15 |
+
"max_position_embeddings": 1024,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 11,
|
| 20 |
+
"num_key_value_heads": 4,
|
| 21 |
+
"pad_token_id": 2,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-06,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"rope_theta": 10000.0,
|
| 26 |
+
"rope_type": "default"
|
| 27 |
+
},
|
| 28 |
+
"tie_word_embeddings": true,
|
| 29 |
+
"transformers_version": "5.10.2",
|
| 30 |
+
"use_cache": false,
|
| 31 |
+
"vocab_size": 25000
|
| 32 |
+
}
|
checkpoint-17314/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 3,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
1
|
| 6 |
+
],
|
| 7 |
+
"output_attentions": false,
|
| 8 |
+
"output_hidden_states": false,
|
| 9 |
+
"pad_token_id": 2,
|
| 10 |
+
"transformers_version": "5.10.2",
|
| 11 |
+
"use_cache": true
|
| 12 |
+
}
|
checkpoint-17314/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28fe6ca1fc2dbd08af9535569184d2529aeff6e121396752aec4bd7d78ceec22
|
| 3 |
+
size 379658184
|
checkpoint-17314/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ba94517756b54ec02e83397c6bacf8c3b2d3df4cb0552973f8456b6bbd49151
|
| 3 |
+
size 307603467
|
checkpoint-17314/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
|
| 3 |
+
size 14645
|
checkpoint-17314/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7adc318c68ef9d58b0b3c979cc39f761a6e8e928326716cb786d5c3f05bbf666
|
| 3 |
+
size 1465
|
checkpoint-17314/trainer_state.json
ADDED
|
@@ -0,0 +1,2464 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 17314,
|
| 3 |
+
"best_metric": 1.643274188041687,
|
| 4 |
+
"best_model_checkpoint": "./gros-michel-instruct/checkpoint-17314",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 17314,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.002887836432944438,
|
| 14 |
+
"grad_norm": 1.909358024597168,
|
| 15 |
+
"learning_rate": 2.8242074927953894e-06,
|
| 16 |
+
"loss": 2.1586764526367186,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.005775672865888876,
|
| 21 |
+
"grad_norm": 1.4766672849655151,
|
| 22 |
+
"learning_rate": 5.706051873198848e-06,
|
| 23 |
+
"loss": 2.118319396972656,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.008663509298833314,
|
| 28 |
+
"grad_norm": 1.7509185075759888,
|
| 29 |
+
"learning_rate": 8.587896253602305e-06,
|
| 30 |
+
"loss": 2.0512954711914064,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.011551345731777752,
|
| 35 |
+
"grad_norm": 1.505385398864746,
|
| 36 |
+
"learning_rate": 1.1469740634005764e-05,
|
| 37 |
+
"loss": 2.0283775329589844,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.01443918216472219,
|
| 42 |
+
"grad_norm": 1.4554438591003418,
|
| 43 |
+
"learning_rate": 1.4351585014409224e-05,
|
| 44 |
+
"loss": 2.006014404296875,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.017327018597666628,
|
| 49 |
+
"grad_norm": 1.4109344482421875,
|
| 50 |
+
"learning_rate": 1.723342939481268e-05,
|
| 51 |
+
"loss": 1.9948684692382812,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.020214855030611066,
|
| 56 |
+
"grad_norm": 1.5211414098739624,
|
| 57 |
+
"learning_rate": 1.9999999314322948e-05,
|
| 58 |
+
"loss": 1.9837632751464844,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.023102691463555505,
|
| 63 |
+
"grad_norm": 1.3641057014465332,
|
| 64 |
+
"learning_rate": 1.9999536485886737e-05,
|
| 65 |
+
"loss": 1.9829933166503906,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.025990527896499943,
|
| 70 |
+
"grad_norm": 1.4043707847595215,
|
| 71 |
+
"learning_rate": 1.999821660697185e-05,
|
| 72 |
+
"loss": 1.9271340942382813,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.02887836432944438,
|
| 77 |
+
"grad_norm": 1.5856163501739502,
|
| 78 |
+
"learning_rate": 1.999603979070382e-05,
|
| 79 |
+
"loss": 1.9438902282714843,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.03176620076238882,
|
| 84 |
+
"grad_norm": 2.086683511734009,
|
| 85 |
+
"learning_rate": 1.9993006223655434e-05,
|
| 86 |
+
"loss": 1.92769287109375,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.034654037195333255,
|
| 91 |
+
"grad_norm": 1.7868231534957886,
|
| 92 |
+
"learning_rate": 1.998911616583075e-05,
|
| 93 |
+
"loss": 1.9024320983886718,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.0375418736282777,
|
| 98 |
+
"grad_norm": 1.5367263555526733,
|
| 99 |
+
"learning_rate": 1.998436995064282e-05,
|
| 100 |
+
"loss": 1.9243475341796874,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.04042971006122213,
|
| 105 |
+
"grad_norm": 1.9622865915298462,
|
| 106 |
+
"learning_rate": 1.99787679848851e-05,
|
| 107 |
+
"loss": 1.91025146484375,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.04331754649416657,
|
| 112 |
+
"grad_norm": 1.790409803390503,
|
| 113 |
+
"learning_rate": 1.9972310748696582e-05,
|
| 114 |
+
"loss": 1.9063694763183594,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.04620538292711101,
|
| 119 |
+
"grad_norm": 1.5865241289138794,
|
| 120 |
+
"learning_rate": 1.996499879552066e-05,
|
| 121 |
+
"loss": 1.8904580688476562,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.049093219360055444,
|
| 126 |
+
"grad_norm": 1.7693233489990234,
|
| 127 |
+
"learning_rate": 1.9956832752057676e-05,
|
| 128 |
+
"loss": 1.9070089721679688,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.051981055792999886,
|
| 133 |
+
"grad_norm": 1.8749171495437622,
|
| 134 |
+
"learning_rate": 1.994781331821122e-05,
|
| 135 |
+
"loss": 1.851161651611328,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.05486889222594432,
|
| 140 |
+
"grad_norm": 1.4991623163223267,
|
| 141 |
+
"learning_rate": 1.9937941267028136e-05,
|
| 142 |
+
"loss": 1.8636807250976561,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.05775672865888876,
|
| 147 |
+
"grad_norm": 1.7407857179641724,
|
| 148 |
+
"learning_rate": 1.9927217444632257e-05,
|
| 149 |
+
"loss": 1.8693643188476563,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.0606445650918332,
|
| 154 |
+
"grad_norm": 1.7285146713256836,
|
| 155 |
+
"learning_rate": 1.9915642770151898e-05,
|
| 156 |
+
"loss": 1.8531916809082032,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.06353240152477764,
|
| 161 |
+
"grad_norm": 1.7245564460754395,
|
| 162 |
+
"learning_rate": 1.9903218235641078e-05,
|
| 163 |
+
"loss": 1.8473806762695313,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.06642023795772208,
|
| 168 |
+
"grad_norm": 1.6716848611831665,
|
| 169 |
+
"learning_rate": 1.9889944905994475e-05,
|
| 170 |
+
"loss": 1.8670359802246095,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.06930807439066651,
|
| 175 |
+
"grad_norm": 1.8723669052124023,
|
| 176 |
+
"learning_rate": 1.9875823918856178e-05,
|
| 177 |
+
"loss": 1.861584930419922,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.07219591082361095,
|
| 182 |
+
"grad_norm": 1.6046726703643799,
|
| 183 |
+
"learning_rate": 1.9860856484522163e-05,
|
| 184 |
+
"loss": 1.8729725646972657,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.0750837472565554,
|
| 189 |
+
"grad_norm": 1.5935516357421875,
|
| 190 |
+
"learning_rate": 1.9845043885836565e-05,
|
| 191 |
+
"loss": 1.8364366149902345,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.07797158368949983,
|
| 196 |
+
"grad_norm": 1.8055092096328735,
|
| 197 |
+
"learning_rate": 1.9828387478081732e-05,
|
| 198 |
+
"loss": 1.8610630798339844,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.08085942012244426,
|
| 203 |
+
"grad_norm": 1.951491117477417,
|
| 204 |
+
"learning_rate": 1.981088868886206e-05,
|
| 205 |
+
"loss": 1.8550961303710938,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.0837472565553887,
|
| 210 |
+
"grad_norm": 1.7456669807434082,
|
| 211 |
+
"learning_rate": 1.9792549017981637e-05,
|
| 212 |
+
"loss": 1.8567230224609375,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.08663509298833313,
|
| 217 |
+
"grad_norm": 1.577018141746521,
|
| 218 |
+
"learning_rate": 1.9773370037315688e-05,
|
| 219 |
+
"loss": 1.8586839294433595,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.08952292942127758,
|
| 224 |
+
"grad_norm": 1.6036415100097656,
|
| 225 |
+
"learning_rate": 1.9753353390675866e-05,
|
| 226 |
+
"loss": 1.8475460815429687,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.09241076585422202,
|
| 231 |
+
"grad_norm": 1.7602298259735107,
|
| 232 |
+
"learning_rate": 1.9732500793669348e-05,
|
| 233 |
+
"loss": 1.8480513000488281,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.09529860228716645,
|
| 238 |
+
"grad_norm": 1.6979150772094727,
|
| 239 |
+
"learning_rate": 1.97108140335518e-05,
|
| 240 |
+
"loss": 1.8358657836914063,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.09818643872011089,
|
| 245 |
+
"grad_norm": 1.7246454954147339,
|
| 246 |
+
"learning_rate": 1.9688294969074192e-05,
|
| 247 |
+
"loss": 1.81704833984375,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.10107427515305534,
|
| 252 |
+
"grad_norm": 1.552440881729126,
|
| 253 |
+
"learning_rate": 1.9664945530323485e-05,
|
| 254 |
+
"loss": 1.7836668395996094,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.10396211158599977,
|
| 259 |
+
"grad_norm": 1.9719949960708618,
|
| 260 |
+
"learning_rate": 1.964076771855721e-05,
|
| 261 |
+
"loss": 1.8269293212890625,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.10684994801894421,
|
| 266 |
+
"grad_norm": 1.613662838935852,
|
| 267 |
+
"learning_rate": 1.961576360603193e-05,
|
| 268 |
+
"loss": 1.8393894958496093,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.10973778445188864,
|
| 273 |
+
"grad_norm": 1.637629508972168,
|
| 274 |
+
"learning_rate": 1.9589935335825635e-05,
|
| 275 |
+
"loss": 1.8150407409667968,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.11262562088483308,
|
| 280 |
+
"grad_norm": 1.6869436502456665,
|
| 281 |
+
"learning_rate": 1.956328512165408e-05,
|
| 282 |
+
"loss": 1.7969142150878907,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.11551345731777753,
|
| 287 |
+
"grad_norm": 1.6456196308135986,
|
| 288 |
+
"learning_rate": 1.953581524768101e-05,
|
| 289 |
+
"loss": 1.78811279296875,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.11840129375072196,
|
| 294 |
+
"grad_norm": 2.124149799346924,
|
| 295 |
+
"learning_rate": 1.950752806832243e-05,
|
| 296 |
+
"loss": 1.817532501220703,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.1212891301836664,
|
| 301 |
+
"grad_norm": 1.615490198135376,
|
| 302 |
+
"learning_rate": 1.9478426008044767e-05,
|
| 303 |
+
"loss": 1.8148939514160156,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.12417696661661083,
|
| 308 |
+
"grad_norm": 1.6563453674316406,
|
| 309 |
+
"learning_rate": 1.944851156115711e-05,
|
| 310 |
+
"loss": 1.8007850646972656,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.12706480304955528,
|
| 315 |
+
"grad_norm": 1.5598517656326294,
|
| 316 |
+
"learning_rate": 1.9417787291597408e-05,
|
| 317 |
+
"loss": 1.798876495361328,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.12995263948249972,
|
| 322 |
+
"grad_norm": 1.5592163801193237,
|
| 323 |
+
"learning_rate": 1.9386255832712705e-05,
|
| 324 |
+
"loss": 1.8308999633789063,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.13284047591544415,
|
| 329 |
+
"grad_norm": 1.8133422136306763,
|
| 330 |
+
"learning_rate": 1.9353919887033468e-05,
|
| 331 |
+
"loss": 1.7812022399902343,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.13572831234838859,
|
| 336 |
+
"grad_norm": 1.6968121528625488,
|
| 337 |
+
"learning_rate": 1.9320782226041927e-05,
|
| 338 |
+
"loss": 1.7900765991210938,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.13861614878133302,
|
| 343 |
+
"grad_norm": 1.5911802053451538,
|
| 344 |
+
"learning_rate": 1.928684568993455e-05,
|
| 345 |
+
"loss": 1.8047982788085937,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.14150398521427746,
|
| 350 |
+
"grad_norm": 1.6581611633300781,
|
| 351 |
+
"learning_rate": 1.9252113187378613e-05,
|
| 352 |
+
"loss": 1.7713580322265625,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1443918216472219,
|
| 357 |
+
"grad_norm": 1.5770312547683716,
|
| 358 |
+
"learning_rate": 1.921658769526289e-05,
|
| 359 |
+
"loss": 1.806968231201172,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.14727965808016633,
|
| 364 |
+
"grad_norm": 1.7962005138397217,
|
| 365 |
+
"learning_rate": 1.918027225844252e-05,
|
| 366 |
+
"loss": 1.78536865234375,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.1501674945131108,
|
| 371 |
+
"grad_norm": 1.8649613857269287,
|
| 372 |
+
"learning_rate": 1.914316998947802e-05,
|
| 373 |
+
"loss": 1.7859089660644532,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.15305533094605522,
|
| 378 |
+
"grad_norm": 1.658275842666626,
|
| 379 |
+
"learning_rate": 1.910528406836853e-05,
|
| 380 |
+
"loss": 1.784400634765625,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.15594316737899966,
|
| 385 |
+
"grad_norm": 1.4610999822616577,
|
| 386 |
+
"learning_rate": 1.906661774227924e-05,
|
| 387 |
+
"loss": 1.8047561645507812,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.1588310038119441,
|
| 392 |
+
"grad_norm": 1.8046791553497314,
|
| 393 |
+
"learning_rate": 1.902717432526309e-05,
|
| 394 |
+
"loss": 1.82025146484375,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.16171884024488853,
|
| 399 |
+
"grad_norm": 1.575330138206482,
|
| 400 |
+
"learning_rate": 1.898695719797672e-05,
|
| 401 |
+
"loss": 1.7557150268554687,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.16460667667783296,
|
| 406 |
+
"grad_norm": 1.871414303779602,
|
| 407 |
+
"learning_rate": 1.894596980739073e-05,
|
| 408 |
+
"loss": 1.7949139404296874,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.1674945131107774,
|
| 413 |
+
"grad_norm": 1.556754469871521,
|
| 414 |
+
"learning_rate": 1.8904215666494212e-05,
|
| 415 |
+
"loss": 1.8172926330566406,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.17038234954372183,
|
| 420 |
+
"grad_norm": 1.737238883972168,
|
| 421 |
+
"learning_rate": 1.8861698353993697e-05,
|
| 422 |
+
"loss": 1.7613954162597656,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.17327018597666627,
|
| 427 |
+
"grad_norm": 1.4752652645111084,
|
| 428 |
+
"learning_rate": 1.8818421514006393e-05,
|
| 429 |
+
"loss": 1.7767857360839843,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.17615802240961073,
|
| 434 |
+
"grad_norm": 1.6746442317962646,
|
| 435 |
+
"learning_rate": 1.877438885574787e-05,
|
| 436 |
+
"loss": 1.778104248046875,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.17904585884255517,
|
| 441 |
+
"grad_norm": 1.6609421968460083,
|
| 442 |
+
"learning_rate": 1.872960415321415e-05,
|
| 443 |
+
"loss": 1.7814427185058594,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.1819336952754996,
|
| 448 |
+
"grad_norm": 1.736544132232666,
|
| 449 |
+
"learning_rate": 1.8684071244858214e-05,
|
| 450 |
+
"loss": 1.8073533630371095,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.18482153170844404,
|
| 455 |
+
"grad_norm": 1.9827585220336914,
|
| 456 |
+
"learning_rate": 1.8637794033261046e-05,
|
| 457 |
+
"loss": 1.7557388305664063,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.18770936814138847,
|
| 462 |
+
"grad_norm": 1.789551854133606,
|
| 463 |
+
"learning_rate": 1.859077648479713e-05,
|
| 464 |
+
"loss": 1.7680348205566405,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.1905972045743329,
|
| 469 |
+
"grad_norm": 1.7813619375228882,
|
| 470 |
+
"learning_rate": 1.8543022629294497e-05,
|
| 471 |
+
"loss": 1.7958369445800781,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.19348504100727734,
|
| 476 |
+
"grad_norm": 1.8258658647537231,
|
| 477 |
+
"learning_rate": 1.849453655968933e-05,
|
| 478 |
+
"loss": 1.787459716796875,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.19637287744022178,
|
| 483 |
+
"grad_norm": 1.5540802478790283,
|
| 484 |
+
"learning_rate": 1.8445322431675167e-05,
|
| 485 |
+
"loss": 1.7415863037109376,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.1992607138731662,
|
| 490 |
+
"grad_norm": 2.128023386001587,
|
| 491 |
+
"learning_rate": 1.839538446334672e-05,
|
| 492 |
+
"loss": 1.7830693054199218,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.20214855030611067,
|
| 497 |
+
"grad_norm": 1.5497944355010986,
|
| 498 |
+
"learning_rate": 1.834472693483835e-05,
|
| 499 |
+
"loss": 1.752845001220703,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.2050363867390551,
|
| 504 |
+
"grad_norm": 1.634918212890625,
|
| 505 |
+
"learning_rate": 1.8293354187957205e-05,
|
| 506 |
+
"loss": 1.7226042175292968,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.20792422317199954,
|
| 511 |
+
"grad_norm": 1.4494820833206177,
|
| 512 |
+
"learning_rate": 1.8241270625811105e-05,
|
| 513 |
+
"loss": 1.7640054321289063,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.21081205960494398,
|
| 518 |
+
"grad_norm": 1.5471206903457642,
|
| 519 |
+
"learning_rate": 1.818848071243116e-05,
|
| 520 |
+
"loss": 1.7560072326660157,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.21369989603788841,
|
| 525 |
+
"grad_norm": 1.6115829944610596,
|
| 526 |
+
"learning_rate": 1.8134988972389133e-05,
|
| 527 |
+
"loss": 1.7349012756347657,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.21658773247083285,
|
| 532 |
+
"grad_norm": 1.8919874429702759,
|
| 533 |
+
"learning_rate": 1.808079999040968e-05,
|
| 534 |
+
"loss": 1.751190185546875,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.21947556890377729,
|
| 539 |
+
"grad_norm": 1.5235891342163086,
|
| 540 |
+
"learning_rate": 1.8025918410977373e-05,
|
| 541 |
+
"loss": 1.7529731750488282,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.22236340533672172,
|
| 546 |
+
"grad_norm": 1.9161391258239746,
|
| 547 |
+
"learning_rate": 1.797034893793863e-05,
|
| 548 |
+
"loss": 1.7648738098144532,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.22525124176966616,
|
| 553 |
+
"grad_norm": 1.6345319747924805,
|
| 554 |
+
"learning_rate": 1.791409633409857e-05,
|
| 555 |
+
"loss": 1.7473922729492188,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.22813907820261062,
|
| 560 |
+
"grad_norm": 1.5785596370697021,
|
| 561 |
+
"learning_rate": 1.7857165420812765e-05,
|
| 562 |
+
"loss": 1.7222659301757812,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.23102691463555505,
|
| 567 |
+
"grad_norm": 1.901938557624817,
|
| 568 |
+
"learning_rate": 1.7799561077574033e-05,
|
| 569 |
+
"loss": 1.767135467529297,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.2339147510684995,
|
| 574 |
+
"grad_norm": 1.7114841938018799,
|
| 575 |
+
"learning_rate": 1.7741288241594223e-05,
|
| 576 |
+
"loss": 1.7547673034667968,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.23680258750144392,
|
| 581 |
+
"grad_norm": 1.4892417192459106,
|
| 582 |
+
"learning_rate": 1.7682351907381034e-05,
|
| 583 |
+
"loss": 1.7259805297851563,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.23969042393438836,
|
| 588 |
+
"grad_norm": 1.4946736097335815,
|
| 589 |
+
"learning_rate": 1.7622757126309945e-05,
|
| 590 |
+
"loss": 1.7445753479003907,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.2425782603673328,
|
| 595 |
+
"grad_norm": 1.6325931549072266,
|
| 596 |
+
"learning_rate": 1.7562509006191277e-05,
|
| 597 |
+
"loss": 1.7440658569335938,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.24546609680027723,
|
| 602 |
+
"grad_norm": 1.5260684490203857,
|
| 603 |
+
"learning_rate": 1.75016127108324e-05,
|
| 604 |
+
"loss": 1.768995361328125,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.24835393323322166,
|
| 609 |
+
"grad_norm": 1.7039450407028198,
|
| 610 |
+
"learning_rate": 1.744007345959514e-05,
|
| 611 |
+
"loss": 1.7270556640625,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.2512417696661661,
|
| 616 |
+
"grad_norm": 1.8408640623092651,
|
| 617 |
+
"learning_rate": 1.7377896526948457e-05,
|
| 618 |
+
"loss": 1.744080810546875,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.25412960609911056,
|
| 623 |
+
"grad_norm": 1.5456106662750244,
|
| 624 |
+
"learning_rate": 1.731508724201635e-05,
|
| 625 |
+
"loss": 1.7196807861328125,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.257017442532055,
|
| 630 |
+
"grad_norm": 1.5989595651626587,
|
| 631 |
+
"learning_rate": 1.7251650988121117e-05,
|
| 632 |
+
"loss": 1.718225860595703,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.25990527896499943,
|
| 637 |
+
"grad_norm": 1.7446736097335815,
|
| 638 |
+
"learning_rate": 1.7187593202321943e-05,
|
| 639 |
+
"loss": 1.7268605041503906,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.26279311539794387,
|
| 644 |
+
"grad_norm": 1.6288601160049438,
|
| 645 |
+
"learning_rate": 1.712291937494892e-05,
|
| 646 |
+
"loss": 1.7300091552734376,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.2656809518308883,
|
| 651 |
+
"grad_norm": 1.5699272155761719,
|
| 652 |
+
"learning_rate": 1.7057635049132444e-05,
|
| 653 |
+
"loss": 1.7404280090332032,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.26856878826383274,
|
| 658 |
+
"grad_norm": 1.5675526857376099,
|
| 659 |
+
"learning_rate": 1.699174582032814e-05,
|
| 660 |
+
"loss": 1.743017120361328,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.27145662469677717,
|
| 665 |
+
"grad_norm": 1.9028033018112183,
|
| 666 |
+
"learning_rate": 1.692525733583729e-05,
|
| 667 |
+
"loss": 1.7313507080078125,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.2743444611297216,
|
| 672 |
+
"grad_norm": 1.679025650024414,
|
| 673 |
+
"learning_rate": 1.6858175294322765e-05,
|
| 674 |
+
"loss": 1.7050982666015626,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.27723229756266604,
|
| 679 |
+
"grad_norm": 1.9859976768493652,
|
| 680 |
+
"learning_rate": 1.6790505445320652e-05,
|
| 681 |
+
"loss": 1.7545135498046875,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.2801201339956105,
|
| 686 |
+
"grad_norm": 1.5824297666549683,
|
| 687 |
+
"learning_rate": 1.672225358874743e-05,
|
| 688 |
+
"loss": 1.7135137939453124,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.2830079704285549,
|
| 693 |
+
"grad_norm": 1.75406813621521,
|
| 694 |
+
"learning_rate": 1.665342557440288e-05,
|
| 695 |
+
"loss": 1.7267703247070312,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.28589580686149935,
|
| 700 |
+
"grad_norm": 1.463829517364502,
|
| 701 |
+
"learning_rate": 1.6584027301468702e-05,
|
| 702 |
+
"loss": 1.766334991455078,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.2887836432944438,
|
| 707 |
+
"grad_norm": 1.6549428701400757,
|
| 708 |
+
"learning_rate": 1.6514064718002907e-05,
|
| 709 |
+
"loss": 1.7155796813964843,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.2916714797273882,
|
| 714 |
+
"grad_norm": 1.7096177339553833,
|
| 715 |
+
"learning_rate": 1.6443543820429994e-05,
|
| 716 |
+
"loss": 1.7424166870117188,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.29455931616033265,
|
| 721 |
+
"grad_norm": 1.8001513481140137,
|
| 722 |
+
"learning_rate": 1.6372470653027028e-05,
|
| 723 |
+
"loss": 1.7000042724609374,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.29744715259327714,
|
| 728 |
+
"grad_norm": 1.6341148614883423,
|
| 729 |
+
"learning_rate": 1.6300851307405578e-05,
|
| 730 |
+
"loss": 1.7590242004394532,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.3003349890262216,
|
| 735 |
+
"grad_norm": 1.5808595418930054,
|
| 736 |
+
"learning_rate": 1.6228691921989615e-05,
|
| 737 |
+
"loss": 1.7320175170898438,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.303222825459166,
|
| 742 |
+
"grad_norm": 1.6536704301834106,
|
| 743 |
+
"learning_rate": 1.6155998681489387e-05,
|
| 744 |
+
"loss": 1.713394775390625,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.30611066189211045,
|
| 749 |
+
"grad_norm": 1.7273539304733276,
|
| 750 |
+
"learning_rate": 1.608277781637135e-05,
|
| 751 |
+
"loss": 1.7214227294921876,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.3089984983250549,
|
| 756 |
+
"grad_norm": 1.4707372188568115,
|
| 757 |
+
"learning_rate": 1.6009035602324127e-05,
|
| 758 |
+
"loss": 1.7233995056152345,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.3118863347579993,
|
| 763 |
+
"grad_norm": 1.5928140878677368,
|
| 764 |
+
"learning_rate": 1.5934778359720675e-05,
|
| 765 |
+
"loss": 1.6906736755371095,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.31477417119094375,
|
| 770 |
+
"grad_norm": 1.8241453170776367,
|
| 771 |
+
"learning_rate": 1.5860012453076524e-05,
|
| 772 |
+
"loss": 1.7230400085449218,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.3176620076238882,
|
| 777 |
+
"grad_norm": 1.4234002828598022,
|
| 778 |
+
"learning_rate": 1.5784744290504303e-05,
|
| 779 |
+
"loss": 1.7349647521972655,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.3205498440568326,
|
| 784 |
+
"grad_norm": 1.699881911277771,
|
| 785 |
+
"learning_rate": 1.5708980323164516e-05,
|
| 786 |
+
"loss": 1.6921821594238282,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.32343768048977706,
|
| 791 |
+
"grad_norm": 1.8189245462417603,
|
| 792 |
+
"learning_rate": 1.5632727044712593e-05,
|
| 793 |
+
"loss": 1.715100555419922,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.3263255169227215,
|
| 798 |
+
"grad_norm": 1.4830021858215332,
|
| 799 |
+
"learning_rate": 1.5555990990742357e-05,
|
| 800 |
+
"loss": 1.6999215698242187,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.3292133533556659,
|
| 805 |
+
"grad_norm": 1.5961987972259521,
|
| 806 |
+
"learning_rate": 1.547877873822583e-05,
|
| 807 |
+
"loss": 1.6873838806152344,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.33210118978861036,
|
| 812 |
+
"grad_norm": 1.5166821479797363,
|
| 813 |
+
"learning_rate": 1.5401096904949557e-05,
|
| 814 |
+
"loss": 1.7406321716308595,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.3349890262215548,
|
| 819 |
+
"grad_norm": 1.5214062929153442,
|
| 820 |
+
"learning_rate": 1.532295214894739e-05,
|
| 821 |
+
"loss": 1.717274169921875,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.33787686265449923,
|
| 826 |
+
"grad_norm": 1.6674859523773193,
|
| 827 |
+
"learning_rate": 1.5244351167929823e-05,
|
| 828 |
+
"loss": 1.6962802124023437,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.34076469908744367,
|
| 833 |
+
"grad_norm": 1.6426726579666138,
|
| 834 |
+
"learning_rate": 1.5165300698709965e-05,
|
| 835 |
+
"loss": 1.6908042907714844,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.3436525355203881,
|
| 840 |
+
"grad_norm": 1.5097322463989258,
|
| 841 |
+
"learning_rate": 1.5085807516626107e-05,
|
| 842 |
+
"loss": 1.696060028076172,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.34654037195333254,
|
| 847 |
+
"grad_norm": 1.791212797164917,
|
| 848 |
+
"learning_rate": 1.5005878434961032e-05,
|
| 849 |
+
"loss": 1.692390899658203,
|
| 850 |
+
"step": 6000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.34942820838627703,
|
| 854 |
+
"grad_norm": 1.6023569107055664,
|
| 855 |
+
"learning_rate": 1.4925520304358045e-05,
|
| 856 |
+
"loss": 1.7093226623535156,
|
| 857 |
+
"step": 6050
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.35231604481922146,
|
| 861 |
+
"grad_norm": 1.5880030393600464,
|
| 862 |
+
"learning_rate": 1.4844740012233819e-05,
|
| 863 |
+
"loss": 1.7326803588867188,
|
| 864 |
+
"step": 6100
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.3552038812521659,
|
| 868 |
+
"grad_norm": 1.981419324874878,
|
| 869 |
+
"learning_rate": 1.4763544482188076e-05,
|
| 870 |
+
"loss": 1.707506103515625,
|
| 871 |
+
"step": 6150
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.35809171768511033,
|
| 875 |
+
"grad_norm": 2.412752866744995,
|
| 876 |
+
"learning_rate": 1.4681940673410169e-05,
|
| 877 |
+
"loss": 1.6915376281738281,
|
| 878 |
+
"step": 6200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.36097955411805477,
|
| 882 |
+
"grad_norm": 1.6970546245574951,
|
| 883 |
+
"learning_rate": 1.4599935580082634e-05,
|
| 884 |
+
"loss": 1.7359632873535156,
|
| 885 |
+
"step": 6250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.3638673905509992,
|
| 889 |
+
"grad_norm": 1.7609550952911377,
|
| 890 |
+
"learning_rate": 1.4517536230781697e-05,
|
| 891 |
+
"loss": 1.6990045166015626,
|
| 892 |
+
"step": 6300
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.36675522698394364,
|
| 896 |
+
"grad_norm": 2.1967782974243164,
|
| 897 |
+
"learning_rate": 1.4434749687874892e-05,
|
| 898 |
+
"loss": 1.7320846557617187,
|
| 899 |
+
"step": 6350
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.3696430634168881,
|
| 903 |
+
"grad_norm": 1.9859265089035034,
|
| 904 |
+
"learning_rate": 1.435158304691572e-05,
|
| 905 |
+
"loss": 1.688117218017578,
|
| 906 |
+
"step": 6400
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.3725308998498325,
|
| 910 |
+
"grad_norm": 1.7201696634292603,
|
| 911 |
+
"learning_rate": 1.4268043436035535e-05,
|
| 912 |
+
"loss": 1.7054281616210938,
|
| 913 |
+
"step": 6450
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.37541873628277694,
|
| 917 |
+
"grad_norm": 1.6569297313690186,
|
| 918 |
+
"learning_rate": 1.4184138015332561e-05,
|
| 919 |
+
"loss": 1.6909075927734376,
|
| 920 |
+
"step": 6500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.3783065727157214,
|
| 924 |
+
"grad_norm": 1.3564033508300781,
|
| 925 |
+
"learning_rate": 1.409987397625823e-05,
|
| 926 |
+
"loss": 1.7042794799804688,
|
| 927 |
+
"step": 6550
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.3811944091486658,
|
| 931 |
+
"grad_norm": 1.5804102420806885,
|
| 932 |
+
"learning_rate": 1.4015258541000805e-05,
|
| 933 |
+
"loss": 1.7144258117675781,
|
| 934 |
+
"step": 6600
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.38408224558161025,
|
| 938 |
+
"grad_norm": 1.8997526168823242,
|
| 939 |
+
"learning_rate": 1.3930298961866368e-05,
|
| 940 |
+
"loss": 1.7227467346191405,
|
| 941 |
+
"step": 6650
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.3869700820145547,
|
| 945 |
+
"grad_norm": 1.5830646753311157,
|
| 946 |
+
"learning_rate": 1.384500252065724e-05,
|
| 947 |
+
"loss": 1.6636395263671875,
|
| 948 |
+
"step": 6700
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.3898579184474991,
|
| 952 |
+
"grad_norm": 1.9107835292816162,
|
| 953 |
+
"learning_rate": 1.3759376528047854e-05,
|
| 954 |
+
"loss": 1.6942121887207031,
|
| 955 |
+
"step": 6750
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.39274575488044355,
|
| 959 |
+
"grad_norm": 2.012664556503296,
|
| 960 |
+
"learning_rate": 1.3673428322958177e-05,
|
| 961 |
+
"loss": 1.6895285034179688,
|
| 962 |
+
"step": 6800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.395633591313388,
|
| 966 |
+
"grad_norm": 1.5067923069000244,
|
| 967 |
+
"learning_rate": 1.3587165271924691e-05,
|
| 968 |
+
"loss": 1.7128817749023437,
|
| 969 |
+
"step": 6850
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.3985214277463324,
|
| 973 |
+
"grad_norm": 1.6675416231155396,
|
| 974 |
+
"learning_rate": 1.3500594768469013e-05,
|
| 975 |
+
"loss": 1.7297770690917968,
|
| 976 |
+
"step": 6900
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.40140926417927686,
|
| 980 |
+
"grad_norm": 1.8157047033309937,
|
| 981 |
+
"learning_rate": 1.34137242324642e-05,
|
| 982 |
+
"loss": 1.6969735717773438,
|
| 983 |
+
"step": 6950
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.40429710061222135,
|
| 987 |
+
"grad_norm": 1.7043299674987793,
|
| 988 |
+
"learning_rate": 1.3326561109498828e-05,
|
| 989 |
+
"loss": 1.7162074279785156,
|
| 990 |
+
"step": 7000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.4071849370451658,
|
| 994 |
+
"grad_norm": 1.5321038961410522,
|
| 995 |
+
"learning_rate": 1.3239112870238783e-05,
|
| 996 |
+
"loss": 1.6754794311523438,
|
| 997 |
+
"step": 7050
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.4100727734781102,
|
| 1001 |
+
"grad_norm": 1.6292834281921387,
|
| 1002 |
+
"learning_rate": 1.3151387009787014e-05,
|
| 1003 |
+
"loss": 1.7081124877929688,
|
| 1004 |
+
"step": 7100
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.41296060991105465,
|
| 1008 |
+
"grad_norm": 1.5183711051940918,
|
| 1009 |
+
"learning_rate": 1.30633910470411e-05,
|
| 1010 |
+
"loss": 1.7077902221679688,
|
| 1011 |
+
"step": 7150
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.4158484463439991,
|
| 1015 |
+
"grad_norm": 1.9996466636657715,
|
| 1016 |
+
"learning_rate": 1.2975132524048823e-05,
|
| 1017 |
+
"loss": 1.7082980346679688,
|
| 1018 |
+
"step": 7200
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.4187362827769435,
|
| 1022 |
+
"grad_norm": 1.5458447933197021,
|
| 1023 |
+
"learning_rate": 1.2886619005361754e-05,
|
| 1024 |
+
"loss": 1.6622166442871094,
|
| 1025 |
+
"step": 7250
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.42162411920988796,
|
| 1029 |
+
"grad_norm": 1.5174431800842285,
|
| 1030 |
+
"learning_rate": 1.279785807738688e-05,
|
| 1031 |
+
"loss": 1.6897288513183595,
|
| 1032 |
+
"step": 7300
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.4245119556428324,
|
| 1036 |
+
"grad_norm": 1.8351675271987915,
|
| 1037 |
+
"learning_rate": 1.2708857347736414e-05,
|
| 1038 |
+
"loss": 1.6853300476074218,
|
| 1039 |
+
"step": 7350
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.42739979207577683,
|
| 1043 |
+
"grad_norm": 1.715381383895874,
|
| 1044 |
+
"learning_rate": 1.2619624444575724e-05,
|
| 1045 |
+
"loss": 1.6880805969238282,
|
| 1046 |
+
"step": 7400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.43028762850872126,
|
| 1050 |
+
"grad_norm": 1.6147445440292358,
|
| 1051 |
+
"learning_rate": 1.2530167015969548e-05,
|
| 1052 |
+
"loss": 1.7018417358398437,
|
| 1053 |
+
"step": 7450
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.4331754649416657,
|
| 1057 |
+
"grad_norm": 1.7816251516342163,
|
| 1058 |
+
"learning_rate": 1.244049272922647e-05,
|
| 1059 |
+
"loss": 1.6919480895996093,
|
| 1060 |
+
"step": 7500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.43606330137461013,
|
| 1064 |
+
"grad_norm": 1.6341052055358887,
|
| 1065 |
+
"learning_rate": 1.2350609270241789e-05,
|
| 1066 |
+
"loss": 1.6707876586914063,
|
| 1067 |
+
"step": 7550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.43895113780755457,
|
| 1071 |
+
"grad_norm": 1.948175072669983,
|
| 1072 |
+
"learning_rate": 1.2260524342838735e-05,
|
| 1073 |
+
"loss": 1.6909373474121094,
|
| 1074 |
+
"step": 7600
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.441838974240499,
|
| 1078 |
+
"grad_norm": 1.8076790571212769,
|
| 1079 |
+
"learning_rate": 1.2170245668108212e-05,
|
| 1080 |
+
"loss": 1.6883811950683594,
|
| 1081 |
+
"step": 7650
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.44472681067344344,
|
| 1085 |
+
"grad_norm": 1.4878610372543335,
|
| 1086 |
+
"learning_rate": 1.2079780983747012e-05,
|
| 1087 |
+
"loss": 1.6800541687011719,
|
| 1088 |
+
"step": 7700
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.4476146471063879,
|
| 1092 |
+
"grad_norm": 1.5750727653503418,
|
| 1093 |
+
"learning_rate": 1.1989138043394625e-05,
|
| 1094 |
+
"loss": 1.669432373046875,
|
| 1095 |
+
"step": 7750
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.4505024835393323,
|
| 1099 |
+
"grad_norm": 1.8376322984695435,
|
| 1100 |
+
"learning_rate": 1.1898324615968703e-05,
|
| 1101 |
+
"loss": 1.651328125,
|
| 1102 |
+
"step": 7800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.45339031997227675,
|
| 1106 |
+
"grad_norm": 1.835090160369873,
|
| 1107 |
+
"learning_rate": 1.1807348484999157e-05,
|
| 1108 |
+
"loss": 1.7042234802246095,
|
| 1109 |
+
"step": 7850
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.45627815640522124,
|
| 1113 |
+
"grad_norm": 1.8628439903259277,
|
| 1114 |
+
"learning_rate": 1.171621744796108e-05,
|
| 1115 |
+
"loss": 1.6629547119140624,
|
| 1116 |
+
"step": 7900
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.45916599283816567,
|
| 1120 |
+
"grad_norm": 1.8912602663040161,
|
| 1121 |
+
"learning_rate": 1.1624939315606398e-05,
|
| 1122 |
+
"loss": 1.6662156677246094,
|
| 1123 |
+
"step": 7950
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.4620538292711101,
|
| 1127 |
+
"grad_norm": 1.5730589628219604,
|
| 1128 |
+
"learning_rate": 1.1533521911294442e-05,
|
| 1129 |
+
"loss": 1.6824444580078124,
|
| 1130 |
+
"step": 8000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.46494166570405454,
|
| 1134 |
+
"grad_norm": 1.9254348278045654,
|
| 1135 |
+
"learning_rate": 1.1441973070321391e-05,
|
| 1136 |
+
"loss": 1.682872314453125,
|
| 1137 |
+
"step": 8050
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.467829502136999,
|
| 1141 |
+
"grad_norm": 1.5994994640350342,
|
| 1142 |
+
"learning_rate": 1.135030063924875e-05,
|
| 1143 |
+
"loss": 1.6739805603027345,
|
| 1144 |
+
"step": 8100
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.4707173385699434,
|
| 1148 |
+
"grad_norm": 1.4216588735580444,
|
| 1149 |
+
"learning_rate": 1.1258512475230807e-05,
|
| 1150 |
+
"loss": 1.663856201171875,
|
| 1151 |
+
"step": 8150
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.47360517500288785,
|
| 1155 |
+
"grad_norm": 1.8356016874313354,
|
| 1156 |
+
"learning_rate": 1.1166616445341192e-05,
|
| 1157 |
+
"loss": 1.6873770141601563,
|
| 1158 |
+
"step": 8200
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.4764930114358323,
|
| 1162 |
+
"grad_norm": 1.7545536756515503,
|
| 1163 |
+
"learning_rate": 1.107462042589863e-05,
|
| 1164 |
+
"loss": 1.6732991027832032,
|
| 1165 |
+
"step": 8250
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.4793808478687767,
|
| 1169 |
+
"grad_norm": 1.6197389364242554,
|
| 1170 |
+
"learning_rate": 1.0982532301791847e-05,
|
| 1171 |
+
"loss": 1.6640838623046874,
|
| 1172 |
+
"step": 8300
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.48226868430172115,
|
| 1176 |
+
"grad_norm": 1.6120935678482056,
|
| 1177 |
+
"learning_rate": 1.089035996580377e-05,
|
| 1178 |
+
"loss": 1.6712326049804687,
|
| 1179 |
+
"step": 8350
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.4851565207346656,
|
| 1183 |
+
"grad_norm": 1.6320757865905762,
|
| 1184 |
+
"learning_rate": 1.0798111317935036e-05,
|
| 1185 |
+
"loss": 1.6833267211914062,
|
| 1186 |
+
"step": 8400
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.48804435716761,
|
| 1190 |
+
"grad_norm": 1.7512632608413696,
|
| 1191 |
+
"learning_rate": 1.0705794264726914e-05,
|
| 1192 |
+
"loss": 1.6716848754882812,
|
| 1193 |
+
"step": 8450
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.49093219360055446,
|
| 1197 |
+
"grad_norm": 1.664544939994812,
|
| 1198 |
+
"learning_rate": 1.0613416718583616e-05,
|
| 1199 |
+
"loss": 1.6609603881835937,
|
| 1200 |
+
"step": 8500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.4938200300334989,
|
| 1204 |
+
"grad_norm": 1.5058132410049438,
|
| 1205 |
+
"learning_rate": 1.052098659709415e-05,
|
| 1206 |
+
"loss": 1.7054307556152344,
|
| 1207 |
+
"step": 8550
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.4967078664664433,
|
| 1211 |
+
"grad_norm": 1.7243521213531494,
|
| 1212 |
+
"learning_rate": 1.0428511822353692e-05,
|
| 1213 |
+
"loss": 1.6992800903320313,
|
| 1214 |
+
"step": 8600
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.49959570289938776,
|
| 1218 |
+
"grad_norm": 2.327094793319702,
|
| 1219 |
+
"learning_rate": 1.0336000320284632e-05,
|
| 1220 |
+
"loss": 1.6802761840820313,
|
| 1221 |
+
"step": 8650
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.5024835393323323,
|
| 1225 |
+
"grad_norm": 1.8326010704040527,
|
| 1226 |
+
"learning_rate": 1.0243460019957196e-05,
|
| 1227 |
+
"loss": 1.6687158203125,
|
| 1228 |
+
"step": 8700
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.5053713757652767,
|
| 1232 |
+
"grad_norm": 1.5392510890960693,
|
| 1233 |
+
"learning_rate": 1.0150898852909895e-05,
|
| 1234 |
+
"loss": 1.7017506408691405,
|
| 1235 |
+
"step": 8750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.5082592121982211,
|
| 1239 |
+
"grad_norm": 1.5009840726852417,
|
| 1240 |
+
"learning_rate": 1.00583247524697e-05,
|
| 1241 |
+
"loss": 1.6612632751464844,
|
| 1242 |
+
"step": 8800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.5111470486311656,
|
| 1246 |
+
"grad_norm": 1.4728184938430786,
|
| 1247 |
+
"learning_rate": 9.965745653072096e-06,
|
| 1248 |
+
"loss": 1.6834872436523438,
|
| 1249 |
+
"step": 8850
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.51403488506411,
|
| 1253 |
+
"grad_norm": 1.6140389442443848,
|
| 1254 |
+
"learning_rate": 9.873169489581012e-06,
|
| 1255 |
+
"loss": 1.6660243225097657,
|
| 1256 |
+
"step": 8900
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.5169227214970544,
|
| 1260 |
+
"grad_norm": 2.108262777328491,
|
| 1261 |
+
"learning_rate": 9.780604196608765e-06,
|
| 1262 |
+
"loss": 1.7060279846191406,
|
| 1263 |
+
"step": 8950
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.5198105579299989,
|
| 1267 |
+
"grad_norm": 1.3609580993652344,
|
| 1268 |
+
"learning_rate": 9.688057707835949e-06,
|
| 1269 |
+
"loss": 1.678179473876953,
|
| 1270 |
+
"step": 9000
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.5226983943629433,
|
| 1274 |
+
"grad_norm": 1.5883218050003052,
|
| 1275 |
+
"learning_rate": 9.59553795533148e-06,
|
| 1276 |
+
"loss": 1.688070068359375,
|
| 1277 |
+
"step": 9050
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.5255862307958877,
|
| 1281 |
+
"grad_norm": 1.6165927648544312,
|
| 1282 |
+
"learning_rate": 9.503052868872727e-06,
|
| 1283 |
+
"loss": 1.6642349243164063,
|
| 1284 |
+
"step": 9100
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.5284740672288322,
|
| 1288 |
+
"grad_norm": 1.9804484844207764,
|
| 1289 |
+
"learning_rate": 9.410610375265874e-06,
|
| 1290 |
+
"loss": 1.6736724853515625,
|
| 1291 |
+
"step": 9150
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.5313619036617766,
|
| 1295 |
+
"grad_norm": 1.6085772514343262,
|
| 1296 |
+
"learning_rate": 9.318218397666513e-06,
|
| 1297 |
+
"loss": 1.6785069274902344,
|
| 1298 |
+
"step": 9200
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.534249740094721,
|
| 1302 |
+
"grad_norm": 1.6080623865127563,
|
| 1303 |
+
"learning_rate": 9.225884854900547e-06,
|
| 1304 |
+
"loss": 1.658829803466797,
|
| 1305 |
+
"step": 9250
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.5371375765276655,
|
| 1309 |
+
"grad_norm": 1.6201704740524292,
|
| 1310 |
+
"learning_rate": 9.133617660785505e-06,
|
| 1311 |
+
"loss": 1.68093017578125,
|
| 1312 |
+
"step": 9300
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.5400254129606099,
|
| 1316 |
+
"grad_norm": 1.7423920631408691,
|
| 1317 |
+
"learning_rate": 9.041424723452221e-06,
|
| 1318 |
+
"loss": 1.6248587036132813,
|
| 1319 |
+
"step": 9350
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.5429132493935543,
|
| 1323 |
+
"grad_norm": 1.4537217617034912,
|
| 1324 |
+
"learning_rate": 8.949313944667067e-06,
|
| 1325 |
+
"loss": 1.6704637145996093,
|
| 1326 |
+
"step": 9400
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.5458010858264988,
|
| 1330 |
+
"grad_norm": 1.5950026512145996,
|
| 1331 |
+
"learning_rate": 8.857293219154667e-06,
|
| 1332 |
+
"loss": 1.675209503173828,
|
| 1333 |
+
"step": 9450
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.5486889222594432,
|
| 1337 |
+
"grad_norm": 1.5893745422363281,
|
| 1338 |
+
"learning_rate": 8.765370433921295e-06,
|
| 1339 |
+
"loss": 1.66578857421875,
|
| 1340 |
+
"step": 9500
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.5515767586923876,
|
| 1344 |
+
"grad_norm": 1.8272515535354614,
|
| 1345 |
+
"learning_rate": 8.673553467578835e-06,
|
| 1346 |
+
"loss": 1.6775022888183593,
|
| 1347 |
+
"step": 9550
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.5544645951253321,
|
| 1351 |
+
"grad_norm": 1.5789238214492798,
|
| 1352 |
+
"learning_rate": 8.581850189669547e-06,
|
| 1353 |
+
"loss": 1.6983125305175781,
|
| 1354 |
+
"step": 9600
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.5573524315582765,
|
| 1358 |
+
"grad_norm": 1.7056528329849243,
|
| 1359 |
+
"learning_rate": 8.490268459991572e-06,
|
| 1360 |
+
"loss": 1.678738555908203,
|
| 1361 |
+
"step": 9650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.560240267991221,
|
| 1365 |
+
"grad_norm": 1.8825808763504028,
|
| 1366 |
+
"learning_rate": 8.398816127925262e-06,
|
| 1367 |
+
"loss": 1.6820941162109375,
|
| 1368 |
+
"step": 9700
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.5631281044241654,
|
| 1372 |
+
"grad_norm": 1.5640108585357666,
|
| 1373 |
+
"learning_rate": 8.307501031760433e-06,
|
| 1374 |
+
"loss": 1.6734996032714844,
|
| 1375 |
+
"step": 9750
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.5660159408571098,
|
| 1379 |
+
"grad_norm": 1.4816759824752808,
|
| 1380 |
+
"learning_rate": 8.216330998024534e-06,
|
| 1381 |
+
"loss": 1.6495098876953125,
|
| 1382 |
+
"step": 9800
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.5689037772900543,
|
| 1386 |
+
"grad_norm": 1.7909116744995117,
|
| 1387 |
+
"learning_rate": 8.125313840811879e-06,
|
| 1388 |
+
"loss": 1.6810971069335938,
|
| 1389 |
+
"step": 9850
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.5717916137229987,
|
| 1393 |
+
"grad_norm": 1.6357197761535645,
|
| 1394 |
+
"learning_rate": 8.034457361113859e-06,
|
| 1395 |
+
"loss": 1.6776667785644532,
|
| 1396 |
+
"step": 9900
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.5746794501559431,
|
| 1400 |
+
"grad_norm": 1.789825201034546,
|
| 1401 |
+
"learning_rate": 7.94376934615037e-06,
|
| 1402 |
+
"loss": 1.668673553466797,
|
| 1403 |
+
"step": 9950
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.5775672865888876,
|
| 1407 |
+
"grad_norm": 1.774862289428711,
|
| 1408 |
+
"learning_rate": 7.85325756870235e-06,
|
| 1409 |
+
"loss": 1.6840528869628906,
|
| 1410 |
+
"step": 10000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.580455123021832,
|
| 1414 |
+
"grad_norm": 1.657164216041565,
|
| 1415 |
+
"learning_rate": 7.762929786445603e-06,
|
| 1416 |
+
"loss": 1.6432334899902343,
|
| 1417 |
+
"step": 10050
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.5833429594547764,
|
| 1421 |
+
"grad_norm": 1.9835896492004395,
|
| 1422 |
+
"learning_rate": 7.67279374128588e-06,
|
| 1423 |
+
"loss": 1.6822531127929687,
|
| 1424 |
+
"step": 10100
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.5862307958877209,
|
| 1428 |
+
"grad_norm": 1.8748596906661987,
|
| 1429 |
+
"learning_rate": 7.582857158695322e-06,
|
| 1430 |
+
"loss": 1.6691751098632812,
|
| 1431 |
+
"step": 10150
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.5891186323206653,
|
| 1435 |
+
"grad_norm": 1.7453851699829102,
|
| 1436 |
+
"learning_rate": 7.493127747050353e-06,
|
| 1437 |
+
"loss": 1.6669338989257811,
|
| 1438 |
+
"step": 10200
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.5920064687536098,
|
| 1442 |
+
"grad_norm": 1.8414359092712402,
|
| 1443 |
+
"learning_rate": 7.4036131969709544e-06,
|
| 1444 |
+
"loss": 1.672958526611328,
|
| 1445 |
+
"step": 10250
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.5948943051865543,
|
| 1449 |
+
"grad_norm": 1.6471014022827148,
|
| 1450 |
+
"learning_rate": 7.314321180661549e-06,
|
| 1451 |
+
"loss": 1.6692904663085937,
|
| 1452 |
+
"step": 10300
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.5977821416194987,
|
| 1456 |
+
"grad_norm": 1.7643845081329346,
|
| 1457 |
+
"learning_rate": 7.225259351253394e-06,
|
| 1458 |
+
"loss": 1.6480662536621093,
|
| 1459 |
+
"step": 10350
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.6006699780524432,
|
| 1463 |
+
"grad_norm": 1.8969323635101318,
|
| 1464 |
+
"learning_rate": 7.136435342148671e-06,
|
| 1465 |
+
"loss": 1.6646434020996095,
|
| 1466 |
+
"step": 10400
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.6035578144853876,
|
| 1470 |
+
"grad_norm": 1.6020116806030273,
|
| 1471 |
+
"learning_rate": 7.0478567663662e-06,
|
| 1472 |
+
"loss": 1.658534698486328,
|
| 1473 |
+
"step": 10450
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.606445650918332,
|
| 1477 |
+
"grad_norm": 1.6735584735870361,
|
| 1478 |
+
"learning_rate": 6.959531215888961e-06,
|
| 1479 |
+
"loss": 1.6774485778808594,
|
| 1480 |
+
"step": 10500
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.6093334873512765,
|
| 1484 |
+
"grad_norm": 1.7870824337005615,
|
| 1485 |
+
"learning_rate": 6.871466261013374e-06,
|
| 1486 |
+
"loss": 1.662366180419922,
|
| 1487 |
+
"step": 10550
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.6122213237842209,
|
| 1491 |
+
"grad_norm": 1.663075566291809,
|
| 1492 |
+
"learning_rate": 6.783669449700476e-06,
|
| 1493 |
+
"loss": 1.6842955017089845,
|
| 1494 |
+
"step": 10600
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.6151091602171653,
|
| 1498 |
+
"grad_norm": 1.731175184249878,
|
| 1499 |
+
"learning_rate": 6.696148306928978e-06,
|
| 1500 |
+
"loss": 1.6783932495117186,
|
| 1501 |
+
"step": 10650
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.6179969966501098,
|
| 1505 |
+
"grad_norm": 1.6345709562301636,
|
| 1506 |
+
"learning_rate": 6.608910334050301e-06,
|
| 1507 |
+
"loss": 1.6980265808105468,
|
| 1508 |
+
"step": 10700
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.6208848330830542,
|
| 1512 |
+
"grad_norm": 1.572339415550232,
|
| 1513 |
+
"learning_rate": 6.5219630081456685e-06,
|
| 1514 |
+
"loss": 1.6494500732421875,
|
| 1515 |
+
"step": 10750
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.6237726695159986,
|
| 1519 |
+
"grad_norm": 1.6059499979019165,
|
| 1520 |
+
"learning_rate": 6.435313781385223e-06,
|
| 1521 |
+
"loss": 1.6419541931152344,
|
| 1522 |
+
"step": 10800
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.6266605059489431,
|
| 1526 |
+
"grad_norm": 1.8062158823013306,
|
| 1527 |
+
"learning_rate": 6.348970080389327e-06,
|
| 1528 |
+
"loss": 1.670931396484375,
|
| 1529 |
+
"step": 10850
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.6295483423818875,
|
| 1533 |
+
"grad_norm": 1.4668166637420654,
|
| 1534 |
+
"learning_rate": 6.2629393055920264e-06,
|
| 1535 |
+
"loss": 1.6773101806640625,
|
| 1536 |
+
"step": 10900
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.6324361788148319,
|
| 1540 |
+
"grad_norm": 1.4954174757003784,
|
| 1541 |
+
"learning_rate": 6.177228830606774e-06,
|
| 1542 |
+
"loss": 1.6458819580078126,
|
| 1543 |
+
"step": 10950
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.6353240152477764,
|
| 1547 |
+
"grad_norm": 1.474352240562439,
|
| 1548 |
+
"learning_rate": 6.091846001594433e-06,
|
| 1549 |
+
"loss": 1.6466766357421876,
|
| 1550 |
+
"step": 11000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.6382118516807208,
|
| 1554 |
+
"grad_norm": 1.7969880104064941,
|
| 1555 |
+
"learning_rate": 6.006798136633656e-06,
|
| 1556 |
+
"loss": 1.68860107421875,
|
| 1557 |
+
"step": 11050
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.6410996881136652,
|
| 1561 |
+
"grad_norm": 1.7214688062667847,
|
| 1562 |
+
"learning_rate": 5.92209252509365e-06,
|
| 1563 |
+
"loss": 1.679088134765625,
|
| 1564 |
+
"step": 11100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.6439875245466097,
|
| 1568 |
+
"grad_norm": 1.8291304111480713,
|
| 1569 |
+
"learning_rate": 5.837736427009426e-06,
|
| 1570 |
+
"loss": 1.619277801513672,
|
| 1571 |
+
"step": 11150
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.6468753609795541,
|
| 1575 |
+
"grad_norm": 2.1492888927459717,
|
| 1576 |
+
"learning_rate": 5.753737072459527e-06,
|
| 1577 |
+
"loss": 1.6094837951660157,
|
| 1578 |
+
"step": 11200
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.6497631974124985,
|
| 1582 |
+
"grad_norm": 1.6656548976898193,
|
| 1583 |
+
"learning_rate": 5.670101660946362e-06,
|
| 1584 |
+
"loss": 1.677728271484375,
|
| 1585 |
+
"step": 11250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.652651033845443,
|
| 1589 |
+
"grad_norm": 1.4686917066574097,
|
| 1590 |
+
"learning_rate": 5.586837360779143e-06,
|
| 1591 |
+
"loss": 1.675196990966797,
|
| 1592 |
+
"step": 11300
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.6555388702783874,
|
| 1596 |
+
"grad_norm": 1.8649028539657593,
|
| 1597 |
+
"learning_rate": 5.5039513084594876e-06,
|
| 1598 |
+
"loss": 1.6791043090820312,
|
| 1599 |
+
"step": 11350
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.6584267067113319,
|
| 1603 |
+
"grad_norm": 1.7350457906723022,
|
| 1604 |
+
"learning_rate": 5.421450608069759e-06,
|
| 1605 |
+
"loss": 1.6814950561523438,
|
| 1606 |
+
"step": 11400
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.6613145431442763,
|
| 1610 |
+
"grad_norm": 1.964842438697815,
|
| 1611 |
+
"learning_rate": 5.33934233066419e-06,
|
| 1612 |
+
"loss": 1.6668353271484375,
|
| 1613 |
+
"step": 11450
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.6642023795772207,
|
| 1617 |
+
"grad_norm": 1.8737123012542725,
|
| 1618 |
+
"learning_rate": 5.257633513662819e-06,
|
| 1619 |
+
"loss": 1.6469189453125,
|
| 1620 |
+
"step": 11500
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.6670902160101652,
|
| 1624 |
+
"grad_norm": 1.659156322479248,
|
| 1625 |
+
"learning_rate": 5.176331160248329e-06,
|
| 1626 |
+
"loss": 1.6729913330078126,
|
| 1627 |
+
"step": 11550
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.6699780524431096,
|
| 1631 |
+
"grad_norm": 1.7961658239364624,
|
| 1632 |
+
"learning_rate": 5.095442238765793e-06,
|
| 1633 |
+
"loss": 1.6672465515136718,
|
| 1634 |
+
"step": 11600
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.672865888876054,
|
| 1638 |
+
"grad_norm": 1.4658139944076538,
|
| 1639 |
+
"learning_rate": 5.014973682125461e-06,
|
| 1640 |
+
"loss": 1.6669802856445313,
|
| 1641 |
+
"step": 11650
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.6757537253089985,
|
| 1645 |
+
"grad_norm": 1.4296271800994873,
|
| 1646 |
+
"learning_rate": 4.93493238720851e-06,
|
| 1647 |
+
"loss": 1.7058442687988282,
|
| 1648 |
+
"step": 11700
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.6786415617419429,
|
| 1652 |
+
"grad_norm": 1.7735282182693481,
|
| 1653 |
+
"learning_rate": 4.855325214275928e-06,
|
| 1654 |
+
"loss": 1.6220843505859375,
|
| 1655 |
+
"step": 11750
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.6815293981748873,
|
| 1659 |
+
"grad_norm": 1.6961135864257812,
|
| 1660 |
+
"learning_rate": 4.776158986380543e-06,
|
| 1661 |
+
"loss": 1.6413137817382812,
|
| 1662 |
+
"step": 11800
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.6844172346078318,
|
| 1666 |
+
"grad_norm": 1.6736559867858887,
|
| 1667 |
+
"learning_rate": 4.697440488782214e-06,
|
| 1668 |
+
"loss": 1.648203125,
|
| 1669 |
+
"step": 11850
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.6873050710407762,
|
| 1673 |
+
"grad_norm": 1.733680009841919,
|
| 1674 |
+
"learning_rate": 4.619176468366274e-06,
|
| 1675 |
+
"loss": 1.660599822998047,
|
| 1676 |
+
"step": 11900
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.6901929074737206,
|
| 1680 |
+
"grad_norm": 1.8642600774765015,
|
| 1681 |
+
"learning_rate": 4.54137363306526e-06,
|
| 1682 |
+
"loss": 1.6600419616699218,
|
| 1683 |
+
"step": 11950
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.6930807439066651,
|
| 1687 |
+
"grad_norm": 1.8840606212615967,
|
| 1688 |
+
"learning_rate": 4.464038651283985e-06,
|
| 1689 |
+
"loss": 1.683453826904297,
|
| 1690 |
+
"step": 12000
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.6959685803396095,
|
| 1694 |
+
"grad_norm": 1.8018664121627808,
|
| 1695 |
+
"learning_rate": 4.3871781513279976e-06,
|
| 1696 |
+
"loss": 1.6748831176757812,
|
| 1697 |
+
"step": 12050
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.6988564167725541,
|
| 1701 |
+
"grad_norm": 1.8538159132003784,
|
| 1702 |
+
"learning_rate": 4.3107987208354725e-06,
|
| 1703 |
+
"loss": 1.6308975219726562,
|
| 1704 |
+
"step": 12100
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.7017442532054985,
|
| 1708 |
+
"grad_norm": 1.6661447286605835,
|
| 1709 |
+
"learning_rate": 4.234906906212585e-06,
|
| 1710 |
+
"loss": 1.655359344482422,
|
| 1711 |
+
"step": 12150
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.7046320896384429,
|
| 1715 |
+
"grad_norm": 1.5082441568374634,
|
| 1716 |
+
"learning_rate": 4.159509212072445e-06,
|
| 1717 |
+
"loss": 1.6447402954101562,
|
| 1718 |
+
"step": 12200
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.7075199260713874,
|
| 1722 |
+
"grad_norm": 1.7229399681091309,
|
| 1723 |
+
"learning_rate": 4.084612100677574e-06,
|
| 1724 |
+
"loss": 1.6748966979980469,
|
| 1725 |
+
"step": 12250
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.7104077625043318,
|
| 1729 |
+
"grad_norm": 1.8212205171585083,
|
| 1730 |
+
"learning_rate": 4.010221991386029e-06,
|
| 1731 |
+
"loss": 1.6603372192382813,
|
| 1732 |
+
"step": 12300
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.7132955989372762,
|
| 1736 |
+
"grad_norm": 1.7022404670715332,
|
| 1737 |
+
"learning_rate": 3.936345260101224e-06,
|
| 1738 |
+
"loss": 1.6737469482421874,
|
| 1739 |
+
"step": 12350
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.7161834353702207,
|
| 1743 |
+
"grad_norm": 1.7086529731750488,
|
| 1744 |
+
"learning_rate": 3.8629882387254405e-06,
|
| 1745 |
+
"loss": 1.6395367431640624,
|
| 1746 |
+
"step": 12400
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.7190712718031651,
|
| 1750 |
+
"grad_norm": 1.6520321369171143,
|
| 1751 |
+
"learning_rate": 3.7901572146171374e-06,
|
| 1752 |
+
"loss": 1.6600856018066406,
|
| 1753 |
+
"step": 12450
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.7219591082361095,
|
| 1757 |
+
"grad_norm": 1.7346488237380981,
|
| 1758 |
+
"learning_rate": 3.717858430052057e-06,
|
| 1759 |
+
"loss": 1.633073272705078,
|
| 1760 |
+
"step": 12500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.724846944669054,
|
| 1764 |
+
"grad_norm": 1.7913254499435425,
|
| 1765 |
+
"learning_rate": 3.6460980816882183e-06,
|
| 1766 |
+
"loss": 1.653193359375,
|
| 1767 |
+
"step": 12550
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.7277347811019984,
|
| 1771 |
+
"grad_norm": 1.533605694770813,
|
| 1772 |
+
"learning_rate": 3.574882320034797e-06,
|
| 1773 |
+
"loss": 1.6393182373046875,
|
| 1774 |
+
"step": 12600
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.7306226175349428,
|
| 1778 |
+
"grad_norm": 1.6388990879058838,
|
| 1779 |
+
"learning_rate": 3.5042172489249783e-06,
|
| 1780 |
+
"loss": 1.669947509765625,
|
| 1781 |
+
"step": 12650
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.7335104539678873,
|
| 1785 |
+
"grad_norm": 2.0770416259765625,
|
| 1786 |
+
"learning_rate": 3.4341089249927883e-06,
|
| 1787 |
+
"loss": 1.6865406799316407,
|
| 1788 |
+
"step": 12700
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.7363982904008317,
|
| 1792 |
+
"grad_norm": 2.1256344318389893,
|
| 1793 |
+
"learning_rate": 3.364563357154016e-06,
|
| 1794 |
+
"loss": 1.6530905151367188,
|
| 1795 |
+
"step": 12750
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.7392861268337761,
|
| 1799 |
+
"grad_norm": 1.5653146505355835,
|
| 1800 |
+
"learning_rate": 3.2955865060911698e-06,
|
| 1801 |
+
"loss": 1.6506643676757813,
|
| 1802 |
+
"step": 12800
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.7421739632667206,
|
| 1806 |
+
"grad_norm": 1.7314833402633667,
|
| 1807 |
+
"learning_rate": 3.2271842837425917e-06,
|
| 1808 |
+
"loss": 1.6328428649902345,
|
| 1809 |
+
"step": 12850
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.745061799699665,
|
| 1813 |
+
"grad_norm": 1.5500118732452393,
|
| 1814 |
+
"learning_rate": 3.159362552795767e-06,
|
| 1815 |
+
"loss": 1.603163299560547,
|
| 1816 |
+
"step": 12900
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.7479496361326095,
|
| 1820 |
+
"grad_norm": 1.451372742652893,
|
| 1821 |
+
"learning_rate": 3.092127126184834e-06,
|
| 1822 |
+
"loss": 1.6769163513183594,
|
| 1823 |
+
"step": 12950
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.7508374725655539,
|
| 1827 |
+
"grad_norm": 1.7614902257919312,
|
| 1828 |
+
"learning_rate": 3.0254837665923554e-06,
|
| 1829 |
+
"loss": 1.672865753173828,
|
| 1830 |
+
"step": 13000
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.7537253089984983,
|
| 1834 |
+
"grad_norm": 1.5754705667495728,
|
| 1835 |
+
"learning_rate": 2.9594381859554124e-06,
|
| 1836 |
+
"loss": 1.6698451232910156,
|
| 1837 |
+
"step": 13050
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.7566131454314428,
|
| 1841 |
+
"grad_norm": 1.782485008239746,
|
| 1842 |
+
"learning_rate": 2.893996044976034e-06,
|
| 1843 |
+
"loss": 1.6509432983398438,
|
| 1844 |
+
"step": 13100
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.7595009818643872,
|
| 1848 |
+
"grad_norm": 1.7711464166641235,
|
| 1849 |
+
"learning_rate": 2.829162952636033e-06,
|
| 1850 |
+
"loss": 1.7015487670898437,
|
| 1851 |
+
"step": 13150
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.7623888182973316,
|
| 1855 |
+
"grad_norm": 1.5038173198699951,
|
| 1856 |
+
"learning_rate": 2.7649444657162585e-06,
|
| 1857 |
+
"loss": 1.689712677001953,
|
| 1858 |
+
"step": 13200
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.7652766547302761,
|
| 1862 |
+
"grad_norm": 1.7243162393569946,
|
| 1863 |
+
"learning_rate": 2.701346088320318e-06,
|
| 1864 |
+
"loss": 1.648837127685547,
|
| 1865 |
+
"step": 13250
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.7681644911632205,
|
| 1869 |
+
"grad_norm": 1.6037522554397583,
|
| 1870 |
+
"learning_rate": 2.638373271402861e-06,
|
| 1871 |
+
"loss": 1.6385755920410157,
|
| 1872 |
+
"step": 13300
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.7710523275961649,
|
| 1876 |
+
"grad_norm": 1.5733518600463867,
|
| 1877 |
+
"learning_rate": 2.576031412302349e-06,
|
| 1878 |
+
"loss": 1.6177082824707032,
|
| 1879 |
+
"step": 13350
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.7739401640291094,
|
| 1883 |
+
"grad_norm": 1.5366418361663818,
|
| 1884 |
+
"learning_rate": 2.5143258542784623e-06,
|
| 1885 |
+
"loss": 1.6302444458007812,
|
| 1886 |
+
"step": 13400
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.7768280004620538,
|
| 1890 |
+
"grad_norm": 1.4973539113998413,
|
| 1891 |
+
"learning_rate": 2.4532618860541477e-06,
|
| 1892 |
+
"loss": 1.6360208129882812,
|
| 1893 |
+
"step": 13450
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.7797158368949982,
|
| 1897 |
+
"grad_norm": 1.562669038772583,
|
| 1898 |
+
"learning_rate": 2.3928447413623156e-06,
|
| 1899 |
+
"loss": 1.6490003967285156,
|
| 1900 |
+
"step": 13500
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.7826036733279427,
|
| 1904 |
+
"grad_norm": 1.7103418111801147,
|
| 1905 |
+
"learning_rate": 2.3330795984972664e-06,
|
| 1906 |
+
"loss": 1.6488714599609375,
|
| 1907 |
+
"step": 13550
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.7854915097608871,
|
| 1911 |
+
"grad_norm": 1.535810947418213,
|
| 1912 |
+
"learning_rate": 2.2739715798708596e-06,
|
| 1913 |
+
"loss": 1.6512696838378906,
|
| 1914 |
+
"step": 13600
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.7883793461938315,
|
| 1918 |
+
"grad_norm": 1.6145868301391602,
|
| 1919 |
+
"learning_rate": 2.2155257515734817e-06,
|
| 1920 |
+
"loss": 1.6339292907714844,
|
| 1921 |
+
"step": 13650
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.791267182626776,
|
| 1925 |
+
"grad_norm": 1.6797300577163696,
|
| 1926 |
+
"learning_rate": 2.157747122939834e-06,
|
| 1927 |
+
"loss": 1.648025665283203,
|
| 1928 |
+
"step": 13700
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.7941550190597204,
|
| 1932 |
+
"grad_norm": 1.6211791038513184,
|
| 1933 |
+
"learning_rate": 2.1006406461195927e-06,
|
| 1934 |
+
"loss": 1.6540110778808594,
|
| 1935 |
+
"step": 13750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.7970428554926648,
|
| 1939 |
+
"grad_norm": 1.4437707662582397,
|
| 1940 |
+
"learning_rate": 2.044211215652946e-06,
|
| 1941 |
+
"loss": 1.6965692138671875,
|
| 1942 |
+
"step": 13800
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.7999306919256093,
|
| 1946 |
+
"grad_norm": 1.5797706842422485,
|
| 1947 |
+
"learning_rate": 1.9884636680511214e-06,
|
| 1948 |
+
"loss": 1.644268798828125,
|
| 1949 |
+
"step": 13850
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.8028185283585537,
|
| 1953 |
+
"grad_norm": 1.5433852672576904,
|
| 1954 |
+
"learning_rate": 1.9334027813818245e-06,
|
| 1955 |
+
"loss": 1.646070556640625,
|
| 1956 |
+
"step": 13900
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.8057063647914983,
|
| 1960 |
+
"grad_norm": 1.6594672203063965,
|
| 1961 |
+
"learning_rate": 1.8790332748597195e-06,
|
| 1962 |
+
"loss": 1.6128846740722655,
|
| 1963 |
+
"step": 13950
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.8085942012244427,
|
| 1967 |
+
"grad_norm": 1.9353688955307007,
|
| 1968 |
+
"learning_rate": 1.8253598084419655e-06,
|
| 1969 |
+
"loss": 1.6358761596679687,
|
| 1970 |
+
"step": 14000
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.8114820376573871,
|
| 1974 |
+
"grad_norm": 1.5906012058258057,
|
| 1975 |
+
"learning_rate": 1.7723869824287987e-06,
|
| 1976 |
+
"loss": 1.59000732421875,
|
| 1977 |
+
"step": 14050
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.8143698740903316,
|
| 1981 |
+
"grad_norm": 1.9525619745254517,
|
| 1982 |
+
"learning_rate": 1.720119337069257e-06,
|
| 1983 |
+
"loss": 1.6441253662109374,
|
| 1984 |
+
"step": 14100
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.817257710523276,
|
| 1988 |
+
"grad_norm": 1.576789379119873,
|
| 1989 |
+
"learning_rate": 1.6685613521720334e-06,
|
| 1990 |
+
"loss": 1.640019989013672,
|
| 1991 |
+
"step": 14150
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.8201455469562204,
|
| 1995 |
+
"grad_norm": 1.3934725522994995,
|
| 1996 |
+
"learning_rate": 1.6177174467215162e-06,
|
| 1997 |
+
"loss": 1.627001953125,
|
| 1998 |
+
"step": 14200
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.8230333833891649,
|
| 2002 |
+
"grad_norm": 1.8330366611480713,
|
| 2003 |
+
"learning_rate": 1.5675919784990468e-06,
|
| 2004 |
+
"loss": 1.6685850524902344,
|
| 2005 |
+
"step": 14250
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.8259212198221093,
|
| 2009 |
+
"grad_norm": 1.5132921934127808,
|
| 2010 |
+
"learning_rate": 1.5181892437094148e-06,
|
| 2011 |
+
"loss": 1.6635725402832031,
|
| 2012 |
+
"step": 14300
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.8288090562550537,
|
| 2016 |
+
"grad_norm": 1.7452752590179443,
|
| 2017 |
+
"learning_rate": 1.4695134766126285e-06,
|
| 2018 |
+
"loss": 1.6459202575683594,
|
| 2019 |
+
"step": 14350
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.8316968926879982,
|
| 2023 |
+
"grad_norm": 1.9245837926864624,
|
| 2024 |
+
"learning_rate": 1.421568849161018e-06,
|
| 2025 |
+
"loss": 1.6429708862304688,
|
| 2026 |
+
"step": 14400
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.8345847291209426,
|
| 2030 |
+
"grad_norm": 1.7182401418685913,
|
| 2031 |
+
"learning_rate": 1.3743594706416453e-06,
|
| 2032 |
+
"loss": 1.6326852416992188,
|
| 2033 |
+
"step": 14450
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.837472565553887,
|
| 2037 |
+
"grad_norm": 1.8509873151779175,
|
| 2038 |
+
"learning_rate": 1.327889387324104e-06,
|
| 2039 |
+
"loss": 1.6436007690429688,
|
| 2040 |
+
"step": 14500
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.8403604019868315,
|
| 2044 |
+
"grad_norm": 2.129338502883911,
|
| 2045 |
+
"learning_rate": 1.2821625821137228e-06,
|
| 2046 |
+
"loss": 1.64162353515625,
|
| 2047 |
+
"step": 14550
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.8432482384197759,
|
| 2051 |
+
"grad_norm": 1.7505098581314087,
|
| 2052 |
+
"learning_rate": 1.2371829742101926e-06,
|
| 2053 |
+
"loss": 1.656988067626953,
|
| 2054 |
+
"step": 14600
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.8461360748527204,
|
| 2058 |
+
"grad_norm": 1.6397038698196411,
|
| 2059 |
+
"learning_rate": 1.1929544187716535e-06,
|
| 2060 |
+
"loss": 1.605113525390625,
|
| 2061 |
+
"step": 14650
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.8490239112856648,
|
| 2065 |
+
"grad_norm": 1.8759368658065796,
|
| 2066 |
+
"learning_rate": 1.149480706584275e-06,
|
| 2067 |
+
"loss": 1.6254893493652345,
|
| 2068 |
+
"step": 14700
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.8519117477186092,
|
| 2072 |
+
"grad_norm": 1.5237820148468018,
|
| 2073 |
+
"learning_rate": 1.1067655637373499e-06,
|
| 2074 |
+
"loss": 1.6461317443847656,
|
| 2075 |
+
"step": 14750
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.8547995841515537,
|
| 2079 |
+
"grad_norm": 1.7539292573928833,
|
| 2080 |
+
"learning_rate": 1.064812651303937e-06,
|
| 2081 |
+
"loss": 1.6243466186523436,
|
| 2082 |
+
"step": 14800
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.8576874205844981,
|
| 2086 |
+
"grad_norm": 1.9527074098587036,
|
| 2087 |
+
"learning_rate": 1.0236255650270743e-06,
|
| 2088 |
+
"loss": 1.6253079223632811,
|
| 2089 |
+
"step": 14850
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.8605752570174425,
|
| 2093 |
+
"grad_norm": 1.6070497035980225,
|
| 2094 |
+
"learning_rate": 9.832078350115814e-07,
|
| 2095 |
+
"loss": 1.6698310852050782,
|
| 2096 |
+
"step": 14900
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.863463093450387,
|
| 2100 |
+
"grad_norm": 1.7272017002105713,
|
| 2101 |
+
"learning_rate": 9.435629254215206e-07,
|
| 2102 |
+
"loss": 1.636328125,
|
| 2103 |
+
"step": 14950
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.8663509298833314,
|
| 2107 |
+
"grad_norm": 1.6818664073944092,
|
| 2108 |
+
"learning_rate": 9.046942341832631e-07,
|
| 2109 |
+
"loss": 1.6539004516601563,
|
| 2110 |
+
"step": 15000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.8692387663162758,
|
| 2114 |
+
"grad_norm": 1.534456491470337,
|
| 2115 |
+
"learning_rate": 8.666050926942627e-07,
|
| 2116 |
+
"loss": 1.6301126098632812,
|
| 2117 |
+
"step": 15050
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.8721266027492203,
|
| 2121 |
+
"grad_norm": 1.6540178060531616,
|
| 2122 |
+
"learning_rate": 8.292987655375351e-07,
|
| 2123 |
+
"loss": 1.6415965270996093,
|
| 2124 |
+
"step": 15100
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.8750144391821647,
|
| 2128 |
+
"grad_norm": 1.83139967918396,
|
| 2129 |
+
"learning_rate": 7.927784502018432e-07,
|
| 2130 |
+
"loss": 1.6059373474121095,
|
| 2131 |
+
"step": 15150
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.8779022756151091,
|
| 2135 |
+
"grad_norm": 1.5106669664382935,
|
| 2136 |
+
"learning_rate": 7.570472768076464e-07,
|
| 2137 |
+
"loss": 1.6205198669433594,
|
| 2138 |
+
"step": 15200
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.8807901120480536,
|
| 2142 |
+
"grad_norm": 1.7108333110809326,
|
| 2143 |
+
"learning_rate": 7.221083078388202e-07,
|
| 2144 |
+
"loss": 1.623634796142578,
|
| 2145 |
+
"step": 15250
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.883677948480998,
|
| 2149 |
+
"grad_norm": 1.694732666015625,
|
| 2150 |
+
"learning_rate": 6.879645378801769e-07,
|
| 2151 |
+
"loss": 1.647873992919922,
|
| 2152 |
+
"step": 15300
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.8865657849139424,
|
| 2156 |
+
"grad_norm": 1.911983609199524,
|
| 2157 |
+
"learning_rate": 6.546188933607978e-07,
|
| 2158 |
+
"loss": 1.6603366088867189,
|
| 2159 |
+
"step": 15350
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.8894536213468869,
|
| 2163 |
+
"grad_norm": 1.683427333831787,
|
| 2164 |
+
"learning_rate": 6.220742323032203e-07,
|
| 2165 |
+
"loss": 1.6439109802246095,
|
| 2166 |
+
"step": 15400
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.8923414577798313,
|
| 2170 |
+
"grad_norm": 1.729750633239746,
|
| 2171 |
+
"learning_rate": 5.90333344078462e-07,
|
| 2172 |
+
"loss": 1.661332550048828,
|
| 2173 |
+
"step": 15450
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.8952292942127758,
|
| 2177 |
+
"grad_norm": 1.9331262111663818,
|
| 2178 |
+
"learning_rate": 5.593989491669716e-07,
|
| 2179 |
+
"loss": 1.656004638671875,
|
| 2180 |
+
"step": 15500
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.8981171306457202,
|
| 2184 |
+
"grad_norm": 1.9529179334640503,
|
| 2185 |
+
"learning_rate": 5.292736989254388e-07,
|
| 2186 |
+
"loss": 1.6529193115234375,
|
| 2187 |
+
"step": 15550
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.9010049670786646,
|
| 2191 |
+
"grad_norm": 1.6848317384719849,
|
| 2192 |
+
"learning_rate": 4.999601753595562e-07,
|
| 2193 |
+
"loss": 1.6554107666015625,
|
| 2194 |
+
"step": 15600
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.903892803511609,
|
| 2198 |
+
"grad_norm": 1.4782564640045166,
|
| 2199 |
+
"learning_rate": 4.714608909027185e-07,
|
| 2200 |
+
"loss": 1.6237272644042968,
|
| 2201 |
+
"step": 15650
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.9067806399445535,
|
| 2205 |
+
"grad_norm": 2.1200802326202393,
|
| 2206 |
+
"learning_rate": 4.437782882006858e-07,
|
| 2207 |
+
"loss": 1.6512266540527343,
|
| 2208 |
+
"step": 15700
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.9096684763774979,
|
| 2212 |
+
"grad_norm": 1.6821216344833374,
|
| 2213 |
+
"learning_rate": 4.169147399022233e-07,
|
| 2214 |
+
"loss": 1.6788369750976562,
|
| 2215 |
+
"step": 15750
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.9125563128104425,
|
| 2219 |
+
"grad_norm": 1.7959096431732178,
|
| 2220 |
+
"learning_rate": 3.9087254845574405e-07,
|
| 2221 |
+
"loss": 1.6628717041015626,
|
| 2222 |
+
"step": 15800
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.9154441492433869,
|
| 2226 |
+
"grad_norm": 1.8554561138153076,
|
| 2227 |
+
"learning_rate": 3.6565394591197233e-07,
|
| 2228 |
+
"loss": 1.6514073181152344,
|
| 2229 |
+
"step": 15850
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.9183319856763313,
|
| 2233 |
+
"grad_norm": 1.6167805194854736,
|
| 2234 |
+
"learning_rate": 3.412610937326333e-07,
|
| 2235 |
+
"loss": 1.6294755554199218,
|
| 2236 |
+
"step": 15900
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.9212198221092758,
|
| 2240 |
+
"grad_norm": 1.634292483329773,
|
| 2241 |
+
"learning_rate": 3.176960826051967e-07,
|
| 2242 |
+
"loss": 1.6554867553710937,
|
| 2243 |
+
"step": 15950
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.9241076585422202,
|
| 2247 |
+
"grad_norm": 1.8534597158432007,
|
| 2248 |
+
"learning_rate": 2.9496093226368594e-07,
|
| 2249 |
+
"loss": 1.6308354187011718,
|
| 2250 |
+
"step": 16000
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.9269954949751646,
|
| 2254 |
+
"grad_norm": 1.5817147493362427,
|
| 2255 |
+
"learning_rate": 2.7305759131557084e-07,
|
| 2256 |
+
"loss": 1.6643826293945312,
|
| 2257 |
+
"step": 16050
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.9298833314081091,
|
| 2261 |
+
"grad_norm": 1.7468533515930176,
|
| 2262 |
+
"learning_rate": 2.5198793707475246e-07,
|
| 2263 |
+
"loss": 1.6447023010253907,
|
| 2264 |
+
"step": 16100
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.9327711678410535,
|
| 2268 |
+
"grad_norm": 1.8023625612258911,
|
| 2269 |
+
"learning_rate": 2.3175377540065935e-07,
|
| 2270 |
+
"loss": 1.6464491271972657,
|
| 2271 |
+
"step": 16150
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.935659004273998,
|
| 2275 |
+
"grad_norm": 1.8202990293502808,
|
| 2276 |
+
"learning_rate": 2.1235684054347172e-07,
|
| 2277 |
+
"loss": 1.6422506713867187,
|
| 2278 |
+
"step": 16200
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.9385468407069424,
|
| 2282 |
+
"grad_norm": 1.6471933126449585,
|
| 2283 |
+
"learning_rate": 1.9379879499547782e-07,
|
| 2284 |
+
"loss": 1.6311958312988282,
|
| 2285 |
+
"step": 16250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.9414346771398868,
|
| 2289 |
+
"grad_norm": 1.604020595550537,
|
| 2290 |
+
"learning_rate": 1.7608122934858697e-07,
|
| 2291 |
+
"loss": 1.646822967529297,
|
| 2292 |
+
"step": 16300
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.9443225135728313,
|
| 2296 |
+
"grad_norm": 1.7205817699432373,
|
| 2297 |
+
"learning_rate": 1.5920566215799848e-07,
|
| 2298 |
+
"loss": 1.6669078063964844,
|
| 2299 |
+
"step": 16350
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.9472103500057757,
|
| 2303 |
+
"grad_norm": 2.0770716667175293,
|
| 2304 |
+
"learning_rate": 1.4317353981204708e-07,
|
| 2305 |
+
"loss": 1.6899119567871095,
|
| 2306 |
+
"step": 16400
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.9500981864387201,
|
| 2310 |
+
"grad_norm": 2.079603433609009,
|
| 2311 |
+
"learning_rate": 1.279862364082385e-07,
|
| 2312 |
+
"loss": 1.645362091064453,
|
| 2313 |
+
"step": 16450
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.9529860228716646,
|
| 2317 |
+
"grad_norm": 1.9399348497390747,
|
| 2318 |
+
"learning_rate": 1.1364505363547273e-07,
|
| 2319 |
+
"loss": 1.655186004638672,
|
| 2320 |
+
"step": 16500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.955873859304609,
|
| 2324 |
+
"grad_norm": 1.5795135498046875,
|
| 2325 |
+
"learning_rate": 1.0015122066247773e-07,
|
| 2326 |
+
"loss": 1.6845616149902343,
|
| 2327 |
+
"step": 16550
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.9587616957375534,
|
| 2331 |
+
"grad_norm": 1.7427446842193604,
|
| 2332 |
+
"learning_rate": 8.750589403246357e-08,
|
| 2333 |
+
"loss": 1.6390235900878907,
|
| 2334 |
+
"step": 16600
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.9616495321704979,
|
| 2338 |
+
"grad_norm": 1.4469588994979858,
|
| 2339 |
+
"learning_rate": 7.57101575639918e-08,
|
| 2340 |
+
"loss": 1.62392822265625,
|
| 2341 |
+
"step": 16650
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.9645373686034423,
|
| 2345 |
+
"grad_norm": 1.5478720664978027,
|
| 2346 |
+
"learning_rate": 6.47650222580798e-08,
|
| 2347 |
+
"loss": 1.6577781677246093,
|
| 2348 |
+
"step": 16700
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.9674252050363867,
|
| 2352 |
+
"grad_norm": 1.6617034673690796,
|
| 2353 |
+
"learning_rate": 5.4671426211557745e-08,
|
| 2354 |
+
"loss": 1.6417259216308593,
|
| 2355 |
+
"step": 16750
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.9703130414693312,
|
| 2359 |
+
"grad_norm": 1.6961334943771362,
|
| 2360 |
+
"learning_rate": 4.543023453665529e-08,
|
| 2361 |
+
"loss": 1.674486083984375,
|
| 2362 |
+
"step": 16800
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.9732008779022756,
|
| 2366 |
+
"grad_norm": 1.6244295835494995,
|
| 2367 |
+
"learning_rate": 3.704223928686079e-08,
|
| 2368 |
+
"loss": 1.642431182861328,
|
| 2369 |
+
"step": 16850
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.97608871433522,
|
| 2373 |
+
"grad_norm": 2.1322579383850098,
|
| 2374 |
+
"learning_rate": 2.9508159389031178e-08,
|
| 2375 |
+
"loss": 1.648513946533203,
|
| 2376 |
+
"step": 16900
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.9789765507681645,
|
| 2380 |
+
"grad_norm": 1.9750131368637085,
|
| 2381 |
+
"learning_rate": 2.2828640581776807e-08,
|
| 2382 |
+
"loss": 1.6566302490234375,
|
| 2383 |
+
"step": 16950
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.9818643872011089,
|
| 2387 |
+
"grad_norm": 1.7420287132263184,
|
| 2388 |
+
"learning_rate": 1.70042553601113e-08,
|
| 2389 |
+
"loss": 1.6620863342285157,
|
| 2390 |
+
"step": 17000
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.9847522236340533,
|
| 2394 |
+
"grad_norm": 1.6003963947296143,
|
| 2395 |
+
"learning_rate": 1.2035502926388554e-08,
|
| 2396 |
+
"loss": 1.6512625122070312,
|
| 2397 |
+
"step": 17050
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.9876400600669978,
|
| 2401 |
+
"grad_norm": 1.5864931344985962,
|
| 2402 |
+
"learning_rate": 7.922809147513644e-09,
|
| 2403 |
+
"loss": 1.6260646057128907,
|
| 2404 |
+
"step": 17100
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.9905278964999422,
|
| 2408 |
+
"grad_norm": 1.500195860862732,
|
| 2409 |
+
"learning_rate": 4.666526518440906e-09,
|
| 2410 |
+
"loss": 1.642152099609375,
|
| 2411 |
+
"step": 17150
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.9934157329328867,
|
| 2415 |
+
"grad_norm": 1.839897632598877,
|
| 2416 |
+
"learning_rate": 2.2669341319669914e-09,
|
| 2417 |
+
"loss": 1.6416644287109374,
|
| 2418 |
+
"step": 17200
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.9963035693658311,
|
| 2422 |
+
"grad_norm": 1.6722792387008667,
|
| 2423 |
+
"learning_rate": 7.242376548044494e-10,
|
| 2424 |
+
"loss": 1.6504067993164062,
|
| 2425 |
+
"step": 17250
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.9991914057987755,
|
| 2429 |
+
"grad_norm": 1.5650098323822021,
|
| 2430 |
+
"learning_rate": 3.8569309956937305e-11,
|
| 2431 |
+
"loss": 1.6279316711425782,
|
| 2432 |
+
"step": 17300
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 1.0,
|
| 2436 |
+
"eval_loss": 1.643274188041687,
|
| 2437 |
+
"eval_runtime": 140.814,
|
| 2438 |
+
"eval_samples_per_second": 80.305,
|
| 2439 |
+
"eval_steps_per_second": 1.257,
|
| 2440 |
+
"step": 17314
|
| 2441 |
+
}
|
| 2442 |
+
],
|
| 2443 |
+
"logging_steps": 50,
|
| 2444 |
+
"max_steps": 17314,
|
| 2445 |
+
"num_input_tokens_seen": 0,
|
| 2446 |
+
"num_train_epochs": 1,
|
| 2447 |
+
"save_steps": 500,
|
| 2448 |
+
"stateful_callbacks": {
|
| 2449 |
+
"TrainerControl": {
|
| 2450 |
+
"args": {
|
| 2451 |
+
"should_epoch_stop": false,
|
| 2452 |
+
"should_evaluate": false,
|
| 2453 |
+
"should_log": false,
|
| 2454 |
+
"should_save": true,
|
| 2455 |
+
"should_training_stop": true
|
| 2456 |
+
},
|
| 2457 |
+
"attributes": {}
|
| 2458 |
+
}
|
| 2459 |
+
},
|
| 2460 |
+
"total_flos": 1.2886352033071104e+17,
|
| 2461 |
+
"train_batch_size": 32,
|
| 2462 |
+
"trial_name": null,
|
| 2463 |
+
"trial_params": null
|
| 2464 |
+
}
|
checkpoint-17314/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:14b56fc7ae3eb024f6675ebfb2d39a01a3878288ebdce3082ddbbaeafc0475ae
|
| 3 |
+
size 5201
|
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 3,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 1,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 2304,
|
| 15 |
+
"max_position_embeddings": 1024,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 11,
|
| 20 |
+
"num_key_value_heads": 4,
|
| 21 |
+
"pad_token_id": 2,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-06,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"rope_theta": 10000.0,
|
| 26 |
+
"rope_type": "default"
|
| 27 |
+
},
|
| 28 |
+
"tie_word_embeddings": true,
|
| 29 |
+
"transformers_version": "5.10.2",
|
| 30 |
+
"use_cache": false,
|
| 31 |
+
"vocab_size": 25000
|
| 32 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 3,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
1
|
| 6 |
+
],
|
| 7 |
+
"output_attentions": false,
|
| 8 |
+
"output_hidden_states": false,
|
| 9 |
+
"pad_token_id": 2,
|
| 10 |
+
"transformers_version": "5.10.2",
|
| 11 |
+
"use_cache": true
|
| 12 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28fe6ca1fc2dbd08af9535569184d2529aeff6e121396752aec4bd7d78ceec22
|
| 3 |
+
size 379658184
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"eos_token": "<|im_end|>",
|
| 5 |
+
"is_local": true,
|
| 6 |
+
"local_files_only": false,
|
| 7 |
+
"model_max_length": 4096,
|
| 8 |
+
"pad_token": "<pad>",
|
| 9 |
+
"tokenizer_class": "TokenizersBackend"
|
| 10 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:14b56fc7ae3eb024f6675ebfb2d39a01a3878288ebdce3082ddbbaeafc0475ae
|
| 3 |
+
size 5201
|