Training in progress, step 375, checkpoint
Browse files- last-checkpoint/chat_template.jinja +4 -0
- last-checkpoint/config.json +66 -0
- last-checkpoint/generation_config.json +10 -0
- last-checkpoint/model.safetensors +3 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer_config.json +22 -0
- last-checkpoint/trainer_state.json +2659 -0
- last-checkpoint/training_args.bin +3 -0
last-checkpoint/chat_template.jinja
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '
|
| 2 |
+
' + message['content'] + '<|im_end|>' + '
|
| 3 |
+
'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
|
| 4 |
+
' }}{% endif %}
|
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Lfm2MoeForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"auto_map": {
|
| 6 |
+
"AutoConfig": "configuration_lfm2_moe.Lfm2MoeConfig",
|
| 7 |
+
"AutoModelForCausalLM": "modeling_lfm2_moe.Lfm2MoeForCausalLM"
|
| 8 |
+
},
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"conv_L_cache": 3,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"dtype": "bfloat16",
|
| 13 |
+
"eos_token_id": 7,
|
| 14 |
+
"hidden_size": 2048,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 7168,
|
| 17 |
+
"layer_types": [
|
| 18 |
+
"conv",
|
| 19 |
+
"conv",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"conv",
|
| 22 |
+
"conv",
|
| 23 |
+
"conv",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"conv",
|
| 26 |
+
"conv",
|
| 27 |
+
"conv",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"conv",
|
| 30 |
+
"conv",
|
| 31 |
+
"conv",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"conv",
|
| 34 |
+
"conv",
|
| 35 |
+
"conv",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"conv",
|
| 38 |
+
"conv",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"conv",
|
| 41 |
+
"conv"
|
| 42 |
+
],
|
| 43 |
+
"max_position_embeddings": 128000,
|
| 44 |
+
"model_type": "lfm2_moe",
|
| 45 |
+
"moe_intermediate_size": 1792,
|
| 46 |
+
"norm_eps": 1e-05,
|
| 47 |
+
"norm_topk_prob": true,
|
| 48 |
+
"num_attention_heads": 32,
|
| 49 |
+
"num_dense_layers": 2,
|
| 50 |
+
"num_experts": 32,
|
| 51 |
+
"num_experts_per_tok": 4,
|
| 52 |
+
"num_hidden_layers": 24,
|
| 53 |
+
"num_key_value_heads": 8,
|
| 54 |
+
"pad_token_id": 0,
|
| 55 |
+
"rope_parameters": {
|
| 56 |
+
"rope_theta": 1000000.0,
|
| 57 |
+
"rope_type": "default"
|
| 58 |
+
},
|
| 59 |
+
"routed_scaling_factor": 1.0,
|
| 60 |
+
"tie_word_embeddings": true,
|
| 61 |
+
"transformers_version": "5.0.0",
|
| 62 |
+
"unsloth_version": "2026.1.4",
|
| 63 |
+
"use_cache": false,
|
| 64 |
+
"use_expert_bias": true,
|
| 65 |
+
"vocab_size": 65536
|
| 66 |
+
}
|
last-checkpoint/generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
7
|
| 6 |
+
],
|
| 7 |
+
"max_length": 128000,
|
| 8 |
+
"pad_token_id": 0,
|
| 9 |
+
"transformers_version": "5.0.0"
|
| 10 |
+
}
|
last-checkpoint/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b6b72cf77930f6e49871825a7bc1f922793bcfcb9c02699bc735b610817fdcc3
|
| 3 |
+
size 16680154224
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9acc139e8aea2be68ada96539cfea414f4a12fcb36183ae93da05ec2f5de6f8
|
| 3 |
+
size 16957053431
|
last-checkpoint/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b7c72b81b0d822b6aa4a72bce960bb779e90a2897b5fc0a4b59dcb48b9b3ae5
|
| 3 |
+
size 14645
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:363d5d1b00939e1cc84fa67cef40f428e133b876ac240cbbacddf195b8308ea8
|
| 3 |
+
size 1465
|
last-checkpoint/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
last-checkpoint/tokenizer_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|startoftext|>",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "<|im_end|>",
|
| 6 |
+
"is_local": false,
|
| 7 |
+
"legacy": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 13 |
+
"model_specific_special_tokens": {},
|
| 14 |
+
"pad_token": "<|pad|>",
|
| 15 |
+
"padding_side": "right",
|
| 16 |
+
"sp_model_kwargs": {},
|
| 17 |
+
"spaces_between_special_tokens": false,
|
| 18 |
+
"tokenizer_class": "TokenizersBackend",
|
| 19 |
+
"unk_token": null,
|
| 20 |
+
"use_default_system_prompt": false,
|
| 21 |
+
"use_fast": true
|
| 22 |
+
}
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,2659 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.09488866396761134,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 375,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00025303643724696357,
|
| 14 |
+
"grad_norm": 20.5,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 2.7386393547058105,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0005060728744939271,
|
| 21 |
+
"grad_norm": 16.0,
|
| 22 |
+
"learning_rate": 0.00022988505747126436,
|
| 23 |
+
"loss": 2.6319832801818848,
|
| 24 |
+
"step": 2
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0007591093117408907,
|
| 28 |
+
"grad_norm": 23.625,
|
| 29 |
+
"learning_rate": 0.0004597701149425287,
|
| 30 |
+
"loss": 2.7548398971557617,
|
| 31 |
+
"step": 3
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0010121457489878543,
|
| 35 |
+
"grad_norm": 13.125,
|
| 36 |
+
"learning_rate": 0.0006896551724137932,
|
| 37 |
+
"loss": 2.396958589553833,
|
| 38 |
+
"step": 4
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0012651821862348178,
|
| 42 |
+
"grad_norm": 9.9375,
|
| 43 |
+
"learning_rate": 0.0009195402298850574,
|
| 44 |
+
"loss": 2.2186975479125977,
|
| 45 |
+
"step": 5
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0015182186234817814,
|
| 49 |
+
"grad_norm": 7.59375,
|
| 50 |
+
"learning_rate": 0.0011494252873563218,
|
| 51 |
+
"loss": 2.168830394744873,
|
| 52 |
+
"step": 6
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.001771255060728745,
|
| 56 |
+
"grad_norm": 4.84375,
|
| 57 |
+
"learning_rate": 0.0013793103448275863,
|
| 58 |
+
"loss": 1.9716055393218994,
|
| 59 |
+
"step": 7
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.0020242914979757085,
|
| 63 |
+
"grad_norm": 2.546875,
|
| 64 |
+
"learning_rate": 0.0016091954022988506,
|
| 65 |
+
"loss": 1.8040101528167725,
|
| 66 |
+
"step": 8
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.002277327935222672,
|
| 70 |
+
"grad_norm": 1.484375,
|
| 71 |
+
"learning_rate": 0.0018390804597701149,
|
| 72 |
+
"loss": 1.6331110000610352,
|
| 73 |
+
"step": 9
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0025303643724696357,
|
| 77 |
+
"grad_norm": 1.4296875,
|
| 78 |
+
"learning_rate": 0.0020689655172413794,
|
| 79 |
+
"loss": 1.559295892715454,
|
| 80 |
+
"step": 10
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.002783400809716599,
|
| 84 |
+
"grad_norm": 1.2265625,
|
| 85 |
+
"learning_rate": 0.0022988505747126436,
|
| 86 |
+
"loss": 1.4502407312393188,
|
| 87 |
+
"step": 11
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.003036437246963563,
|
| 91 |
+
"grad_norm": 0.90625,
|
| 92 |
+
"learning_rate": 0.0025287356321839084,
|
| 93 |
+
"loss": 1.6704494953155518,
|
| 94 |
+
"step": 12
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.003289473684210526,
|
| 98 |
+
"grad_norm": 0.84765625,
|
| 99 |
+
"learning_rate": 0.0027586206896551726,
|
| 100 |
+
"loss": 1.3471019268035889,
|
| 101 |
+
"step": 13
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.00354251012145749,
|
| 105 |
+
"grad_norm": 0.9375,
|
| 106 |
+
"learning_rate": 0.002988505747126437,
|
| 107 |
+
"loss": 1.6374365091323853,
|
| 108 |
+
"step": 14
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0037955465587044533,
|
| 112 |
+
"grad_norm": 0.92578125,
|
| 113 |
+
"learning_rate": 0.003218390804597701,
|
| 114 |
+
"loss": 1.5217013359069824,
|
| 115 |
+
"step": 15
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.004048582995951417,
|
| 119 |
+
"grad_norm": 0.828125,
|
| 120 |
+
"learning_rate": 0.003448275862068966,
|
| 121 |
+
"loss": 1.4502065181732178,
|
| 122 |
+
"step": 16
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.00430161943319838,
|
| 126 |
+
"grad_norm": 0.75,
|
| 127 |
+
"learning_rate": 0.0036781609195402297,
|
| 128 |
+
"loss": 1.5312974452972412,
|
| 129 |
+
"step": 17
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.004554655870445344,
|
| 133 |
+
"grad_norm": 0.77734375,
|
| 134 |
+
"learning_rate": 0.003908045977011495,
|
| 135 |
+
"loss": 1.4265103340148926,
|
| 136 |
+
"step": 18
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.004807692307692308,
|
| 140 |
+
"grad_norm": 0.86328125,
|
| 141 |
+
"learning_rate": 0.004137931034482759,
|
| 142 |
+
"loss": 1.4511473178863525,
|
| 143 |
+
"step": 19
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.005060728744939271,
|
| 147 |
+
"grad_norm": 0.9296875,
|
| 148 |
+
"learning_rate": 0.0043678160919540234,
|
| 149 |
+
"loss": 1.4595309495925903,
|
| 150 |
+
"step": 20
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.005313765182186235,
|
| 154 |
+
"grad_norm": 1.0703125,
|
| 155 |
+
"learning_rate": 0.004597701149425287,
|
| 156 |
+
"loss": 1.4979453086853027,
|
| 157 |
+
"step": 21
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.005566801619433198,
|
| 161 |
+
"grad_norm": 1.171875,
|
| 162 |
+
"learning_rate": 0.004827586206896552,
|
| 163 |
+
"loss": 1.4711323976516724,
|
| 164 |
+
"step": 22
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.005819838056680162,
|
| 168 |
+
"grad_norm": 0.98046875,
|
| 169 |
+
"learning_rate": 0.005057471264367817,
|
| 170 |
+
"loss": 1.4307830333709717,
|
| 171 |
+
"step": 23
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.006072874493927126,
|
| 175 |
+
"grad_norm": 1.1015625,
|
| 176 |
+
"learning_rate": 0.0052873563218390806,
|
| 177 |
+
"loss": 1.4139049053192139,
|
| 178 |
+
"step": 24
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.006325910931174089,
|
| 182 |
+
"grad_norm": 1.2578125,
|
| 183 |
+
"learning_rate": 0.005517241379310345,
|
| 184 |
+
"loss": 1.5090482234954834,
|
| 185 |
+
"step": 25
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.006578947368421052,
|
| 189 |
+
"grad_norm": 0.87109375,
|
| 190 |
+
"learning_rate": 0.005747126436781609,
|
| 191 |
+
"loss": 1.4550621509552002,
|
| 192 |
+
"step": 26
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.0068319838056680165,
|
| 196 |
+
"grad_norm": 0.9609375,
|
| 197 |
+
"learning_rate": 0.005977011494252874,
|
| 198 |
+
"loss": 1.4417916536331177,
|
| 199 |
+
"step": 27
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.00708502024291498,
|
| 203 |
+
"grad_norm": 0.8515625,
|
| 204 |
+
"learning_rate": 0.0062068965517241385,
|
| 205 |
+
"loss": 1.3825995922088623,
|
| 206 |
+
"step": 28
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.007338056680161943,
|
| 210 |
+
"grad_norm": 0.9453125,
|
| 211 |
+
"learning_rate": 0.006436781609195402,
|
| 212 |
+
"loss": 1.369423270225525,
|
| 213 |
+
"step": 29
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0075910931174089065,
|
| 217 |
+
"grad_norm": 0.8203125,
|
| 218 |
+
"learning_rate": 0.006666666666666666,
|
| 219 |
+
"loss": 1.3526304960250854,
|
| 220 |
+
"step": 30
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.00784412955465587,
|
| 224 |
+
"grad_norm": 0.6015625,
|
| 225 |
+
"learning_rate": 0.006896551724137932,
|
| 226 |
+
"loss": 1.3979132175445557,
|
| 227 |
+
"step": 31
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.008097165991902834,
|
| 231 |
+
"grad_norm": 0.640625,
|
| 232 |
+
"learning_rate": 0.007126436781609196,
|
| 233 |
+
"loss": 1.3506033420562744,
|
| 234 |
+
"step": 32
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.008350202429149798,
|
| 238 |
+
"grad_norm": 0.76171875,
|
| 239 |
+
"learning_rate": 0.0073563218390804595,
|
| 240 |
+
"loss": 1.4074552059173584,
|
| 241 |
+
"step": 33
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.00860323886639676,
|
| 245 |
+
"grad_norm": 0.75390625,
|
| 246 |
+
"learning_rate": 0.007586206896551724,
|
| 247 |
+
"loss": 1.4538687467575073,
|
| 248 |
+
"step": 34
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.008856275303643725,
|
| 252 |
+
"grad_norm": 0.80078125,
|
| 253 |
+
"learning_rate": 0.00781609195402299,
|
| 254 |
+
"loss": 1.4242024421691895,
|
| 255 |
+
"step": 35
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.009109311740890687,
|
| 259 |
+
"grad_norm": 0.95703125,
|
| 260 |
+
"learning_rate": 0.008045977011494253,
|
| 261 |
+
"loss": 1.5446974039077759,
|
| 262 |
+
"step": 36
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.009362348178137652,
|
| 266 |
+
"grad_norm": 0.765625,
|
| 267 |
+
"learning_rate": 0.008275862068965517,
|
| 268 |
+
"loss": 1.4819495677947998,
|
| 269 |
+
"step": 37
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.009615384615384616,
|
| 273 |
+
"grad_norm": 0.59765625,
|
| 274 |
+
"learning_rate": 0.008505747126436782,
|
| 275 |
+
"loss": 1.470993161201477,
|
| 276 |
+
"step": 38
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.009868421052631578,
|
| 280 |
+
"grad_norm": 0.609375,
|
| 281 |
+
"learning_rate": 0.008735632183908047,
|
| 282 |
+
"loss": 1.4080774784088135,
|
| 283 |
+
"step": 39
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.010121457489878543,
|
| 287 |
+
"grad_norm": 0.6484375,
|
| 288 |
+
"learning_rate": 0.00896551724137931,
|
| 289 |
+
"loss": 1.4043101072311401,
|
| 290 |
+
"step": 40
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.010374493927125507,
|
| 294 |
+
"grad_norm": 0.8125,
|
| 295 |
+
"learning_rate": 0.009195402298850575,
|
| 296 |
+
"loss": 1.4553110599517822,
|
| 297 |
+
"step": 41
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.01062753036437247,
|
| 301 |
+
"grad_norm": 0.6640625,
|
| 302 |
+
"learning_rate": 0.00942528735632184,
|
| 303 |
+
"loss": 1.4775452613830566,
|
| 304 |
+
"step": 42
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.010880566801619434,
|
| 308 |
+
"grad_norm": 0.6953125,
|
| 309 |
+
"learning_rate": 0.009655172413793104,
|
| 310 |
+
"loss": 1.4560859203338623,
|
| 311 |
+
"step": 43
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.011133603238866396,
|
| 315 |
+
"grad_norm": 0.62890625,
|
| 316 |
+
"learning_rate": 0.009885057471264369,
|
| 317 |
+
"loss": 1.3985958099365234,
|
| 318 |
+
"step": 44
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.01138663967611336,
|
| 322 |
+
"grad_norm": 0.5390625,
|
| 323 |
+
"learning_rate": 0.010114942528735633,
|
| 324 |
+
"loss": 1.4953653812408447,
|
| 325 |
+
"step": 45
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.011639676113360324,
|
| 329 |
+
"grad_norm": 0.578125,
|
| 330 |
+
"learning_rate": 0.010344827586206898,
|
| 331 |
+
"loss": 1.4428997039794922,
|
| 332 |
+
"step": 46
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.011892712550607287,
|
| 336 |
+
"grad_norm": 0.58984375,
|
| 337 |
+
"learning_rate": 0.010574712643678161,
|
| 338 |
+
"loss": 1.4267104864120483,
|
| 339 |
+
"step": 47
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.012145748987854251,
|
| 343 |
+
"grad_norm": 0.875,
|
| 344 |
+
"learning_rate": 0.010804597701149426,
|
| 345 |
+
"loss": 1.3390274047851562,
|
| 346 |
+
"step": 48
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.012398785425101215,
|
| 350 |
+
"grad_norm": 0.59375,
|
| 351 |
+
"learning_rate": 0.01103448275862069,
|
| 352 |
+
"loss": 1.2774426937103271,
|
| 353 |
+
"step": 49
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.012651821862348178,
|
| 357 |
+
"grad_norm": 0.546875,
|
| 358 |
+
"learning_rate": 0.011264367816091954,
|
| 359 |
+
"loss": 1.448158621788025,
|
| 360 |
+
"step": 50
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.012904858299595142,
|
| 364 |
+
"grad_norm": 0.5703125,
|
| 365 |
+
"learning_rate": 0.011494252873563218,
|
| 366 |
+
"loss": 1.552297830581665,
|
| 367 |
+
"step": 51
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.013157894736842105,
|
| 371 |
+
"grad_norm": 0.55859375,
|
| 372 |
+
"learning_rate": 0.011724137931034481,
|
| 373 |
+
"loss": 1.3143055438995361,
|
| 374 |
+
"step": 52
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.013410931174089069,
|
| 378 |
+
"grad_norm": 0.49609375,
|
| 379 |
+
"learning_rate": 0.011954022988505748,
|
| 380 |
+
"loss": 1.5644612312316895,
|
| 381 |
+
"step": 53
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.013663967611336033,
|
| 385 |
+
"grad_norm": 0.56640625,
|
| 386 |
+
"learning_rate": 0.012183908045977012,
|
| 387 |
+
"loss": 1.3711479902267456,
|
| 388 |
+
"step": 54
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.013917004048582995,
|
| 392 |
+
"grad_norm": 0.90625,
|
| 393 |
+
"learning_rate": 0.012413793103448277,
|
| 394 |
+
"loss": 1.5378882884979248,
|
| 395 |
+
"step": 55
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.01417004048582996,
|
| 399 |
+
"grad_norm": 0.6875,
|
| 400 |
+
"learning_rate": 0.01264367816091954,
|
| 401 |
+
"loss": 1.5556209087371826,
|
| 402 |
+
"step": 56
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.014423076923076924,
|
| 406 |
+
"grad_norm": 0.58203125,
|
| 407 |
+
"learning_rate": 0.012873563218390805,
|
| 408 |
+
"loss": 1.537410020828247,
|
| 409 |
+
"step": 57
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.014676113360323886,
|
| 413 |
+
"grad_norm": 0.60546875,
|
| 414 |
+
"learning_rate": 0.01310344827586207,
|
| 415 |
+
"loss": 1.5693731307983398,
|
| 416 |
+
"step": 58
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.01492914979757085,
|
| 420 |
+
"grad_norm": 0.59765625,
|
| 421 |
+
"learning_rate": 0.013333333333333332,
|
| 422 |
+
"loss": 1.6816911697387695,
|
| 423 |
+
"step": 59
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.015182186234817813,
|
| 427 |
+
"grad_norm": 0.61328125,
|
| 428 |
+
"learning_rate": 0.013563218390804597,
|
| 429 |
+
"loss": 1.6039944887161255,
|
| 430 |
+
"step": 60
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.015435222672064777,
|
| 434 |
+
"grad_norm": 0.66796875,
|
| 435 |
+
"learning_rate": 0.013793103448275864,
|
| 436 |
+
"loss": 1.4081966876983643,
|
| 437 |
+
"step": 61
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.01568825910931174,
|
| 441 |
+
"grad_norm": 0.6328125,
|
| 442 |
+
"learning_rate": 0.014022988505747127,
|
| 443 |
+
"loss": 1.6515611410140991,
|
| 444 |
+
"step": 62
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.015941295546558706,
|
| 448 |
+
"grad_norm": 0.64453125,
|
| 449 |
+
"learning_rate": 0.014252873563218391,
|
| 450 |
+
"loss": 1.519553780555725,
|
| 451 |
+
"step": 63
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.016194331983805668,
|
| 455 |
+
"grad_norm": 0.55078125,
|
| 456 |
+
"learning_rate": 0.014482758620689656,
|
| 457 |
+
"loss": 1.5124565362930298,
|
| 458 |
+
"step": 64
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.01644736842105263,
|
| 462 |
+
"grad_norm": 0.470703125,
|
| 463 |
+
"learning_rate": 0.014712643678160919,
|
| 464 |
+
"loss": 1.6325674057006836,
|
| 465 |
+
"step": 65
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.016700404858299597,
|
| 469 |
+
"grad_norm": 0.59765625,
|
| 470 |
+
"learning_rate": 0.014942528735632184,
|
| 471 |
+
"loss": 1.6184418201446533,
|
| 472 |
+
"step": 66
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.01695344129554656,
|
| 476 |
+
"grad_norm": 0.671875,
|
| 477 |
+
"learning_rate": 0.015172413793103448,
|
| 478 |
+
"loss": 1.662506341934204,
|
| 479 |
+
"step": 67
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.01720647773279352,
|
| 483 |
+
"grad_norm": 0.58984375,
|
| 484 |
+
"learning_rate": 0.015402298850574711,
|
| 485 |
+
"loss": 1.6800591945648193,
|
| 486 |
+
"step": 68
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.017459514170040488,
|
| 490 |
+
"grad_norm": 0.6640625,
|
| 491 |
+
"learning_rate": 0.01563218390804598,
|
| 492 |
+
"loss": 1.5753778219223022,
|
| 493 |
+
"step": 69
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.01771255060728745,
|
| 497 |
+
"grad_norm": 0.58984375,
|
| 498 |
+
"learning_rate": 0.015862068965517243,
|
| 499 |
+
"loss": 1.6762198209762573,
|
| 500 |
+
"step": 70
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.017965587044534412,
|
| 504 |
+
"grad_norm": 0.474609375,
|
| 505 |
+
"learning_rate": 0.016091954022988506,
|
| 506 |
+
"loss": 1.5733668804168701,
|
| 507 |
+
"step": 71
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.018218623481781375,
|
| 511 |
+
"grad_norm": 0.4921875,
|
| 512 |
+
"learning_rate": 0.016321839080459772,
|
| 513 |
+
"loss": 1.7303202152252197,
|
| 514 |
+
"step": 72
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.01847165991902834,
|
| 518 |
+
"grad_norm": 0.796875,
|
| 519 |
+
"learning_rate": 0.016551724137931035,
|
| 520 |
+
"loss": 1.7469077110290527,
|
| 521 |
+
"step": 73
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.018724696356275303,
|
| 525 |
+
"grad_norm": 0.5390625,
|
| 526 |
+
"learning_rate": 0.016781609195402298,
|
| 527 |
+
"loss": 1.6346166133880615,
|
| 528 |
+
"step": 74
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.018977732793522266,
|
| 532 |
+
"grad_norm": 0.546875,
|
| 533 |
+
"learning_rate": 0.017011494252873564,
|
| 534 |
+
"loss": 1.8797675371170044,
|
| 535 |
+
"step": 75
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.019230769230769232,
|
| 539 |
+
"grad_norm": 0.51953125,
|
| 540 |
+
"learning_rate": 0.017241379310344827,
|
| 541 |
+
"loss": 1.6815588474273682,
|
| 542 |
+
"step": 76
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.019483805668016194,
|
| 546 |
+
"grad_norm": 0.59375,
|
| 547 |
+
"learning_rate": 0.017471264367816094,
|
| 548 |
+
"loss": 1.5040360689163208,
|
| 549 |
+
"step": 77
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.019736842105263157,
|
| 553 |
+
"grad_norm": 0.51171875,
|
| 554 |
+
"learning_rate": 0.017701149425287357,
|
| 555 |
+
"loss": 1.80476975440979,
|
| 556 |
+
"step": 78
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.019989878542510123,
|
| 560 |
+
"grad_norm": 2.390625,
|
| 561 |
+
"learning_rate": 0.01793103448275862,
|
| 562 |
+
"loss": 1.929945945739746,
|
| 563 |
+
"step": 79
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.020242914979757085,
|
| 567 |
+
"grad_norm": 0.400390625,
|
| 568 |
+
"learning_rate": 0.018160919540229886,
|
| 569 |
+
"loss": 1.7132759094238281,
|
| 570 |
+
"step": 80
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.020495951417004048,
|
| 574 |
+
"grad_norm": 1.015625,
|
| 575 |
+
"learning_rate": 0.01839080459770115,
|
| 576 |
+
"loss": 1.8109431266784668,
|
| 577 |
+
"step": 81
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.020748987854251014,
|
| 581 |
+
"grad_norm": 0.5625,
|
| 582 |
+
"learning_rate": 0.018620689655172412,
|
| 583 |
+
"loss": 1.807767629623413,
|
| 584 |
+
"step": 82
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.021002024291497976,
|
| 588 |
+
"grad_norm": 0.6328125,
|
| 589 |
+
"learning_rate": 0.01885057471264368,
|
| 590 |
+
"loss": 1.6236469745635986,
|
| 591 |
+
"step": 83
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.02125506072874494,
|
| 595 |
+
"grad_norm": 0.5234375,
|
| 596 |
+
"learning_rate": 0.01908045977011494,
|
| 597 |
+
"loss": 1.6389708518981934,
|
| 598 |
+
"step": 84
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.0215080971659919,
|
| 602 |
+
"grad_norm": 0.52734375,
|
| 603 |
+
"learning_rate": 0.019310344827586208,
|
| 604 |
+
"loss": 1.7953280210494995,
|
| 605 |
+
"step": 85
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.021761133603238867,
|
| 609 |
+
"grad_norm": 0.515625,
|
| 610 |
+
"learning_rate": 0.01954022988505747,
|
| 611 |
+
"loss": 1.8677895069122314,
|
| 612 |
+
"step": 86
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.02201417004048583,
|
| 616 |
+
"grad_norm": 0.50390625,
|
| 617 |
+
"learning_rate": 0.019770114942528737,
|
| 618 |
+
"loss": 1.8045387268066406,
|
| 619 |
+
"step": 87
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.022267206477732792,
|
| 623 |
+
"grad_norm": 0.58984375,
|
| 624 |
+
"learning_rate": 0.02,
|
| 625 |
+
"loss": 1.9091248512268066,
|
| 626 |
+
"step": 88
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.022520242914979758,
|
| 630 |
+
"grad_norm": 0.609375,
|
| 631 |
+
"learning_rate": 0.02,
|
| 632 |
+
"loss": 1.808335304260254,
|
| 633 |
+
"step": 89
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.02277327935222672,
|
| 637 |
+
"grad_norm": 0.5546875,
|
| 638 |
+
"learning_rate": 0.02,
|
| 639 |
+
"loss": 1.8608583211898804,
|
| 640 |
+
"step": 90
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.023026315789473683,
|
| 644 |
+
"grad_norm": 0.54296875,
|
| 645 |
+
"learning_rate": 0.02,
|
| 646 |
+
"loss": 1.9061346054077148,
|
| 647 |
+
"step": 91
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.02327935222672065,
|
| 651 |
+
"grad_norm": 0.421875,
|
| 652 |
+
"learning_rate": 0.02,
|
| 653 |
+
"loss": 1.9825823307037354,
|
| 654 |
+
"step": 92
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.02353238866396761,
|
| 658 |
+
"grad_norm": 0.4609375,
|
| 659 |
+
"learning_rate": 0.02,
|
| 660 |
+
"loss": 1.8379669189453125,
|
| 661 |
+
"step": 93
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.023785425101214574,
|
| 665 |
+
"grad_norm": 0.494140625,
|
| 666 |
+
"learning_rate": 0.02,
|
| 667 |
+
"loss": 1.6752934455871582,
|
| 668 |
+
"step": 94
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.02403846153846154,
|
| 672 |
+
"grad_norm": 0.46484375,
|
| 673 |
+
"learning_rate": 0.02,
|
| 674 |
+
"loss": 1.7464494705200195,
|
| 675 |
+
"step": 95
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.024291497975708502,
|
| 679 |
+
"grad_norm": 0.453125,
|
| 680 |
+
"learning_rate": 0.02,
|
| 681 |
+
"loss": 1.8795886039733887,
|
| 682 |
+
"step": 96
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.024544534412955465,
|
| 686 |
+
"grad_norm": 0.419921875,
|
| 687 |
+
"learning_rate": 0.02,
|
| 688 |
+
"loss": 1.8445587158203125,
|
| 689 |
+
"step": 97
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.02479757085020243,
|
| 693 |
+
"grad_norm": 0.40625,
|
| 694 |
+
"learning_rate": 0.02,
|
| 695 |
+
"loss": 1.842057228088379,
|
| 696 |
+
"step": 98
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.025050607287449393,
|
| 700 |
+
"grad_norm": 0.421875,
|
| 701 |
+
"learning_rate": 0.02,
|
| 702 |
+
"loss": 1.6107308864593506,
|
| 703 |
+
"step": 99
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.025303643724696356,
|
| 707 |
+
"grad_norm": 0.392578125,
|
| 708 |
+
"learning_rate": 0.02,
|
| 709 |
+
"loss": 1.6814723014831543,
|
| 710 |
+
"step": 100
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.025556680161943318,
|
| 714 |
+
"grad_norm": 0.404296875,
|
| 715 |
+
"learning_rate": 0.02,
|
| 716 |
+
"loss": 1.706100583076477,
|
| 717 |
+
"step": 101
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.025809716599190284,
|
| 721 |
+
"grad_norm": 0.443359375,
|
| 722 |
+
"learning_rate": 0.02,
|
| 723 |
+
"loss": 1.8581061363220215,
|
| 724 |
+
"step": 102
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.026062753036437247,
|
| 728 |
+
"grad_norm": 0.419921875,
|
| 729 |
+
"learning_rate": 0.02,
|
| 730 |
+
"loss": 1.7351648807525635,
|
| 731 |
+
"step": 103
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.02631578947368421,
|
| 735 |
+
"grad_norm": 0.373046875,
|
| 736 |
+
"learning_rate": 0.02,
|
| 737 |
+
"loss": 1.9900397062301636,
|
| 738 |
+
"step": 104
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.026568825910931175,
|
| 742 |
+
"grad_norm": 0.38671875,
|
| 743 |
+
"learning_rate": 0.02,
|
| 744 |
+
"loss": 1.871990442276001,
|
| 745 |
+
"step": 105
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.026821862348178137,
|
| 749 |
+
"grad_norm": 0.400390625,
|
| 750 |
+
"learning_rate": 0.02,
|
| 751 |
+
"loss": 1.862628698348999,
|
| 752 |
+
"step": 106
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.0270748987854251,
|
| 756 |
+
"grad_norm": 0.421875,
|
| 757 |
+
"learning_rate": 0.02,
|
| 758 |
+
"loss": 1.8539869785308838,
|
| 759 |
+
"step": 107
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.027327935222672066,
|
| 763 |
+
"grad_norm": 0.408203125,
|
| 764 |
+
"learning_rate": 0.02,
|
| 765 |
+
"loss": 1.8793188333511353,
|
| 766 |
+
"step": 108
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.02758097165991903,
|
| 770 |
+
"grad_norm": 0.3984375,
|
| 771 |
+
"learning_rate": 0.02,
|
| 772 |
+
"loss": 1.8705213069915771,
|
| 773 |
+
"step": 109
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.02783400809716599,
|
| 777 |
+
"grad_norm": 0.388671875,
|
| 778 |
+
"learning_rate": 0.02,
|
| 779 |
+
"loss": 1.911798357963562,
|
| 780 |
+
"step": 110
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.028087044534412957,
|
| 784 |
+
"grad_norm": 0.609375,
|
| 785 |
+
"learning_rate": 0.02,
|
| 786 |
+
"loss": 1.9883649349212646,
|
| 787 |
+
"step": 111
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.02834008097165992,
|
| 791 |
+
"grad_norm": 0.44921875,
|
| 792 |
+
"learning_rate": 0.02,
|
| 793 |
+
"loss": 1.8723022937774658,
|
| 794 |
+
"step": 112
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.028593117408906882,
|
| 798 |
+
"grad_norm": 0.392578125,
|
| 799 |
+
"learning_rate": 0.02,
|
| 800 |
+
"loss": 1.8594697713851929,
|
| 801 |
+
"step": 113
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.028846153846153848,
|
| 805 |
+
"grad_norm": 0.34765625,
|
| 806 |
+
"learning_rate": 0.02,
|
| 807 |
+
"loss": 1.5962891578674316,
|
| 808 |
+
"step": 114
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.02909919028340081,
|
| 812 |
+
"grad_norm": 0.384765625,
|
| 813 |
+
"learning_rate": 0.02,
|
| 814 |
+
"loss": 1.8171108961105347,
|
| 815 |
+
"step": 115
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.029352226720647773,
|
| 819 |
+
"grad_norm": 0.404296875,
|
| 820 |
+
"learning_rate": 0.02,
|
| 821 |
+
"loss": 1.7834141254425049,
|
| 822 |
+
"step": 116
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.029605263157894735,
|
| 826 |
+
"grad_norm": 0.53125,
|
| 827 |
+
"learning_rate": 0.02,
|
| 828 |
+
"loss": 1.9656782150268555,
|
| 829 |
+
"step": 117
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.0298582995951417,
|
| 833 |
+
"grad_norm": 0.361328125,
|
| 834 |
+
"learning_rate": 0.02,
|
| 835 |
+
"loss": 1.8432538509368896,
|
| 836 |
+
"step": 118
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.030111336032388664,
|
| 840 |
+
"grad_norm": 0.365234375,
|
| 841 |
+
"learning_rate": 0.02,
|
| 842 |
+
"loss": 1.8127596378326416,
|
| 843 |
+
"step": 119
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.030364372469635626,
|
| 847 |
+
"grad_norm": 0.6953125,
|
| 848 |
+
"learning_rate": 0.02,
|
| 849 |
+
"loss": 1.923449158668518,
|
| 850 |
+
"step": 120
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.030617408906882592,
|
| 854 |
+
"grad_norm": 0.36328125,
|
| 855 |
+
"learning_rate": 0.02,
|
| 856 |
+
"loss": 1.9775664806365967,
|
| 857 |
+
"step": 121
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.030870445344129555,
|
| 861 |
+
"grad_norm": 0.349609375,
|
| 862 |
+
"learning_rate": 0.02,
|
| 863 |
+
"loss": 1.9287586212158203,
|
| 864 |
+
"step": 122
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.031123481781376517,
|
| 868 |
+
"grad_norm": 0.365234375,
|
| 869 |
+
"learning_rate": 0.02,
|
| 870 |
+
"loss": 1.9073728322982788,
|
| 871 |
+
"step": 123
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.03137651821862348,
|
| 875 |
+
"grad_norm": 0.3671875,
|
| 876 |
+
"learning_rate": 0.02,
|
| 877 |
+
"loss": 1.912489891052246,
|
| 878 |
+
"step": 124
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.031629554655870445,
|
| 882 |
+
"grad_norm": 0.337890625,
|
| 883 |
+
"learning_rate": 0.02,
|
| 884 |
+
"loss": 1.906019926071167,
|
| 885 |
+
"step": 125
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.03188259109311741,
|
| 889 |
+
"grad_norm": 0.373046875,
|
| 890 |
+
"learning_rate": 0.02,
|
| 891 |
+
"loss": 1.8980028629302979,
|
| 892 |
+
"step": 126
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.03213562753036437,
|
| 896 |
+
"grad_norm": 0.3515625,
|
| 897 |
+
"learning_rate": 0.02,
|
| 898 |
+
"loss": 1.8099416494369507,
|
| 899 |
+
"step": 127
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.032388663967611336,
|
| 903 |
+
"grad_norm": 0.34765625,
|
| 904 |
+
"learning_rate": 0.02,
|
| 905 |
+
"loss": 1.753847360610962,
|
| 906 |
+
"step": 128
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.0326417004048583,
|
| 910 |
+
"grad_norm": 0.373046875,
|
| 911 |
+
"learning_rate": 0.02,
|
| 912 |
+
"loss": 1.887335181236267,
|
| 913 |
+
"step": 129
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.03289473684210526,
|
| 917 |
+
"grad_norm": 0.35546875,
|
| 918 |
+
"learning_rate": 0.02,
|
| 919 |
+
"loss": 1.8118953704833984,
|
| 920 |
+
"step": 130
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.03314777327935223,
|
| 924 |
+
"grad_norm": 0.3515625,
|
| 925 |
+
"learning_rate": 0.02,
|
| 926 |
+
"loss": 1.905268907546997,
|
| 927 |
+
"step": 131
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.03340080971659919,
|
| 931 |
+
"grad_norm": 0.302734375,
|
| 932 |
+
"learning_rate": 0.02,
|
| 933 |
+
"loss": 1.6896657943725586,
|
| 934 |
+
"step": 132
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.03365384615384615,
|
| 938 |
+
"grad_norm": 0.36328125,
|
| 939 |
+
"learning_rate": 0.02,
|
| 940 |
+
"loss": 1.8001630306243896,
|
| 941 |
+
"step": 133
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.03390688259109312,
|
| 945 |
+
"grad_norm": 0.3828125,
|
| 946 |
+
"learning_rate": 0.02,
|
| 947 |
+
"loss": 1.7924871444702148,
|
| 948 |
+
"step": 134
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.034159919028340084,
|
| 952 |
+
"grad_norm": 0.384765625,
|
| 953 |
+
"learning_rate": 0.02,
|
| 954 |
+
"loss": 1.8576805591583252,
|
| 955 |
+
"step": 135
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.03441295546558704,
|
| 959 |
+
"grad_norm": 0.388671875,
|
| 960 |
+
"learning_rate": 0.02,
|
| 961 |
+
"loss": 1.9136528968811035,
|
| 962 |
+
"step": 136
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.03466599190283401,
|
| 966 |
+
"grad_norm": 0.388671875,
|
| 967 |
+
"learning_rate": 0.02,
|
| 968 |
+
"loss": 1.8513039350509644,
|
| 969 |
+
"step": 137
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.034919028340080975,
|
| 973 |
+
"grad_norm": 0.32421875,
|
| 974 |
+
"learning_rate": 0.02,
|
| 975 |
+
"loss": 1.6870956420898438,
|
| 976 |
+
"step": 138
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.035172064777327934,
|
| 980 |
+
"grad_norm": 0.33203125,
|
| 981 |
+
"learning_rate": 0.02,
|
| 982 |
+
"loss": 1.8061823844909668,
|
| 983 |
+
"step": 139
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.0354251012145749,
|
| 987 |
+
"grad_norm": 0.44921875,
|
| 988 |
+
"learning_rate": 0.02,
|
| 989 |
+
"loss": 1.7925238609313965,
|
| 990 |
+
"step": 140
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.03567813765182186,
|
| 994 |
+
"grad_norm": 0.34765625,
|
| 995 |
+
"learning_rate": 0.02,
|
| 996 |
+
"loss": 1.9564220905303955,
|
| 997 |
+
"step": 141
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.035931174089068825,
|
| 1001 |
+
"grad_norm": 0.353515625,
|
| 1002 |
+
"learning_rate": 0.02,
|
| 1003 |
+
"loss": 1.8622322082519531,
|
| 1004 |
+
"step": 142
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.03618421052631579,
|
| 1008 |
+
"grad_norm": 0.404296875,
|
| 1009 |
+
"learning_rate": 0.02,
|
| 1010 |
+
"loss": 1.839081883430481,
|
| 1011 |
+
"step": 143
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.03643724696356275,
|
| 1015 |
+
"grad_norm": 0.52734375,
|
| 1016 |
+
"learning_rate": 0.02,
|
| 1017 |
+
"loss": 1.8134467601776123,
|
| 1018 |
+
"step": 144
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.036690283400809716,
|
| 1022 |
+
"grad_norm": 0.298828125,
|
| 1023 |
+
"learning_rate": 0.02,
|
| 1024 |
+
"loss": 1.6601486206054688,
|
| 1025 |
+
"step": 145
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.03694331983805668,
|
| 1029 |
+
"grad_norm": 0.310546875,
|
| 1030 |
+
"learning_rate": 0.02,
|
| 1031 |
+
"loss": 1.8179130554199219,
|
| 1032 |
+
"step": 146
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.03719635627530364,
|
| 1036 |
+
"grad_norm": 0.302734375,
|
| 1037 |
+
"learning_rate": 0.02,
|
| 1038 |
+
"loss": 1.8195692300796509,
|
| 1039 |
+
"step": 147
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.03744939271255061,
|
| 1043 |
+
"grad_norm": 0.322265625,
|
| 1044 |
+
"learning_rate": 0.02,
|
| 1045 |
+
"loss": 1.8244118690490723,
|
| 1046 |
+
"step": 148
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.03770242914979757,
|
| 1050 |
+
"grad_norm": 0.36328125,
|
| 1051 |
+
"learning_rate": 0.02,
|
| 1052 |
+
"loss": 1.9877636432647705,
|
| 1053 |
+
"step": 149
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.03795546558704453,
|
| 1057 |
+
"grad_norm": 0.326171875,
|
| 1058 |
+
"learning_rate": 0.02,
|
| 1059 |
+
"loss": 1.8807326555252075,
|
| 1060 |
+
"step": 150
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.0382085020242915,
|
| 1064 |
+
"grad_norm": 0.341796875,
|
| 1065 |
+
"learning_rate": 0.02,
|
| 1066 |
+
"loss": 1.799276351928711,
|
| 1067 |
+
"step": 151
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.038461538461538464,
|
| 1071 |
+
"grad_norm": 0.35546875,
|
| 1072 |
+
"learning_rate": 0.02,
|
| 1073 |
+
"loss": 1.8706992864608765,
|
| 1074 |
+
"step": 152
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.03871457489878542,
|
| 1078 |
+
"grad_norm": 0.380859375,
|
| 1079 |
+
"learning_rate": 0.02,
|
| 1080 |
+
"loss": 1.806701898574829,
|
| 1081 |
+
"step": 153
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.03896761133603239,
|
| 1085 |
+
"grad_norm": 0.3828125,
|
| 1086 |
+
"learning_rate": 0.02,
|
| 1087 |
+
"loss": 1.9343551397323608,
|
| 1088 |
+
"step": 154
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.039220647773279355,
|
| 1092 |
+
"grad_norm": 0.3515625,
|
| 1093 |
+
"learning_rate": 0.02,
|
| 1094 |
+
"loss": 1.8837814331054688,
|
| 1095 |
+
"step": 155
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.039473684210526314,
|
| 1099 |
+
"grad_norm": 0.275390625,
|
| 1100 |
+
"learning_rate": 0.02,
|
| 1101 |
+
"loss": 1.7536430358886719,
|
| 1102 |
+
"step": 156
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.03972672064777328,
|
| 1106 |
+
"grad_norm": 0.3203125,
|
| 1107 |
+
"learning_rate": 0.02,
|
| 1108 |
+
"loss": 2.000185966491699,
|
| 1109 |
+
"step": 157
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.039979757085020245,
|
| 1113 |
+
"grad_norm": 0.3125,
|
| 1114 |
+
"learning_rate": 0.02,
|
| 1115 |
+
"loss": 1.8336604833602905,
|
| 1116 |
+
"step": 158
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.040232793522267205,
|
| 1120 |
+
"grad_norm": 0.298828125,
|
| 1121 |
+
"learning_rate": 0.02,
|
| 1122 |
+
"loss": 1.8002318143844604,
|
| 1123 |
+
"step": 159
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.04048582995951417,
|
| 1127 |
+
"grad_norm": 0.291015625,
|
| 1128 |
+
"learning_rate": 0.02,
|
| 1129 |
+
"loss": 1.9328553676605225,
|
| 1130 |
+
"step": 160
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.040738866396761136,
|
| 1134 |
+
"grad_norm": 0.271484375,
|
| 1135 |
+
"learning_rate": 0.02,
|
| 1136 |
+
"loss": 1.7373687028884888,
|
| 1137 |
+
"step": 161
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.040991902834008095,
|
| 1141 |
+
"grad_norm": 0.283203125,
|
| 1142 |
+
"learning_rate": 0.02,
|
| 1143 |
+
"loss": 1.6631128787994385,
|
| 1144 |
+
"step": 162
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.04124493927125506,
|
| 1148 |
+
"grad_norm": 0.275390625,
|
| 1149 |
+
"learning_rate": 0.02,
|
| 1150 |
+
"loss": 1.5757675170898438,
|
| 1151 |
+
"step": 163
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.04149797570850203,
|
| 1155 |
+
"grad_norm": 0.287109375,
|
| 1156 |
+
"learning_rate": 0.02,
|
| 1157 |
+
"loss": 1.7366318702697754,
|
| 1158 |
+
"step": 164
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.041751012145748986,
|
| 1162 |
+
"grad_norm": 0.29296875,
|
| 1163 |
+
"learning_rate": 0.02,
|
| 1164 |
+
"loss": 1.8514606952667236,
|
| 1165 |
+
"step": 165
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.04200404858299595,
|
| 1169 |
+
"grad_norm": 0.322265625,
|
| 1170 |
+
"learning_rate": 0.02,
|
| 1171 |
+
"loss": 1.810927391052246,
|
| 1172 |
+
"step": 166
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.04225708502024292,
|
| 1176 |
+
"grad_norm": 0.3125,
|
| 1177 |
+
"learning_rate": 0.02,
|
| 1178 |
+
"loss": 1.7555034160614014,
|
| 1179 |
+
"step": 167
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.04251012145748988,
|
| 1183 |
+
"grad_norm": 0.310546875,
|
| 1184 |
+
"learning_rate": 0.02,
|
| 1185 |
+
"loss": 1.804673433303833,
|
| 1186 |
+
"step": 168
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.04276315789473684,
|
| 1190 |
+
"grad_norm": 0.283203125,
|
| 1191 |
+
"learning_rate": 0.02,
|
| 1192 |
+
"loss": 1.8926290273666382,
|
| 1193 |
+
"step": 169
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.0430161943319838,
|
| 1197 |
+
"grad_norm": 0.302734375,
|
| 1198 |
+
"learning_rate": 0.02,
|
| 1199 |
+
"loss": 1.928168773651123,
|
| 1200 |
+
"step": 170
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.04326923076923077,
|
| 1204 |
+
"grad_norm": 0.28125,
|
| 1205 |
+
"learning_rate": 0.02,
|
| 1206 |
+
"loss": 1.9081146717071533,
|
| 1207 |
+
"step": 171
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.043522267206477734,
|
| 1211 |
+
"grad_norm": 0.296875,
|
| 1212 |
+
"learning_rate": 0.02,
|
| 1213 |
+
"loss": 1.8865680694580078,
|
| 1214 |
+
"step": 172
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.04377530364372469,
|
| 1218 |
+
"grad_norm": 0.29296875,
|
| 1219 |
+
"learning_rate": 0.02,
|
| 1220 |
+
"loss": 1.677335500717163,
|
| 1221 |
+
"step": 173
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.04402834008097166,
|
| 1225 |
+
"grad_norm": 0.265625,
|
| 1226 |
+
"learning_rate": 0.02,
|
| 1227 |
+
"loss": 1.5733604431152344,
|
| 1228 |
+
"step": 174
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.044281376518218625,
|
| 1232 |
+
"grad_norm": 0.265625,
|
| 1233 |
+
"learning_rate": 0.02,
|
| 1234 |
+
"loss": 1.7141188383102417,
|
| 1235 |
+
"step": 175
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.044534412955465584,
|
| 1239 |
+
"grad_norm": 0.26171875,
|
| 1240 |
+
"learning_rate": 0.02,
|
| 1241 |
+
"loss": 1.7283015251159668,
|
| 1242 |
+
"step": 176
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.04478744939271255,
|
| 1246 |
+
"grad_norm": 0.26953125,
|
| 1247 |
+
"learning_rate": 0.02,
|
| 1248 |
+
"loss": 1.6554383039474487,
|
| 1249 |
+
"step": 177
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.045040485829959516,
|
| 1253 |
+
"grad_norm": 0.326171875,
|
| 1254 |
+
"learning_rate": 0.02,
|
| 1255 |
+
"loss": 1.8230128288269043,
|
| 1256 |
+
"step": 178
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.045293522267206475,
|
| 1260 |
+
"grad_norm": 0.357421875,
|
| 1261 |
+
"learning_rate": 0.02,
|
| 1262 |
+
"loss": 1.8195432424545288,
|
| 1263 |
+
"step": 179
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.04554655870445344,
|
| 1267 |
+
"grad_norm": 0.37109375,
|
| 1268 |
+
"learning_rate": 0.02,
|
| 1269 |
+
"loss": 1.7151989936828613,
|
| 1270 |
+
"step": 180
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.04579959514170041,
|
| 1274 |
+
"grad_norm": 0.314453125,
|
| 1275 |
+
"learning_rate": 0.02,
|
| 1276 |
+
"loss": 1.8021490573883057,
|
| 1277 |
+
"step": 181
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.046052631578947366,
|
| 1281 |
+
"grad_norm": 0.50390625,
|
| 1282 |
+
"learning_rate": 0.02,
|
| 1283 |
+
"loss": 1.7367699146270752,
|
| 1284 |
+
"step": 182
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.04630566801619433,
|
| 1288 |
+
"grad_norm": 0.275390625,
|
| 1289 |
+
"learning_rate": 0.02,
|
| 1290 |
+
"loss": 1.862687349319458,
|
| 1291 |
+
"step": 183
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.0465587044534413,
|
| 1295 |
+
"grad_norm": 0.25390625,
|
| 1296 |
+
"learning_rate": 0.02,
|
| 1297 |
+
"loss": 1.7368934154510498,
|
| 1298 |
+
"step": 184
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.04681174089068826,
|
| 1302 |
+
"grad_norm": 0.28515625,
|
| 1303 |
+
"learning_rate": 0.02,
|
| 1304 |
+
"loss": 1.767867922782898,
|
| 1305 |
+
"step": 185
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.04706477732793522,
|
| 1309 |
+
"grad_norm": 0.314453125,
|
| 1310 |
+
"learning_rate": 0.02,
|
| 1311 |
+
"loss": 1.7828075885772705,
|
| 1312 |
+
"step": 186
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.04731781376518219,
|
| 1316 |
+
"grad_norm": 0.310546875,
|
| 1317 |
+
"learning_rate": 0.02,
|
| 1318 |
+
"loss": 1.8469303846359253,
|
| 1319 |
+
"step": 187
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.04757085020242915,
|
| 1323 |
+
"grad_norm": 0.4765625,
|
| 1324 |
+
"learning_rate": 0.02,
|
| 1325 |
+
"loss": 1.7417302131652832,
|
| 1326 |
+
"step": 188
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.047823886639676114,
|
| 1330 |
+
"grad_norm": 0.255859375,
|
| 1331 |
+
"learning_rate": 0.02,
|
| 1332 |
+
"loss": 1.767709732055664,
|
| 1333 |
+
"step": 189
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.04807692307692308,
|
| 1337 |
+
"grad_norm": 0.255859375,
|
| 1338 |
+
"learning_rate": 0.02,
|
| 1339 |
+
"loss": 1.6383213996887207,
|
| 1340 |
+
"step": 190
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.04832995951417004,
|
| 1344 |
+
"grad_norm": 0.263671875,
|
| 1345 |
+
"learning_rate": 0.02,
|
| 1346 |
+
"loss": 1.7237427234649658,
|
| 1347 |
+
"step": 191
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.048582995951417005,
|
| 1351 |
+
"grad_norm": 0.275390625,
|
| 1352 |
+
"learning_rate": 0.02,
|
| 1353 |
+
"loss": 1.851733684539795,
|
| 1354 |
+
"step": 192
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.04883603238866397,
|
| 1358 |
+
"grad_norm": 0.310546875,
|
| 1359 |
+
"learning_rate": 0.02,
|
| 1360 |
+
"loss": 1.6895689964294434,
|
| 1361 |
+
"step": 193
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.04908906882591093,
|
| 1365 |
+
"grad_norm": 0.390625,
|
| 1366 |
+
"learning_rate": 0.02,
|
| 1367 |
+
"loss": 1.9249845743179321,
|
| 1368 |
+
"step": 194
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.049342105263157895,
|
| 1372 |
+
"grad_norm": 0.296875,
|
| 1373 |
+
"learning_rate": 0.02,
|
| 1374 |
+
"loss": 1.8771311044692993,
|
| 1375 |
+
"step": 195
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.04959514170040486,
|
| 1379 |
+
"grad_norm": 0.2734375,
|
| 1380 |
+
"learning_rate": 0.02,
|
| 1381 |
+
"loss": 1.9580897092819214,
|
| 1382 |
+
"step": 196
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.04984817813765182,
|
| 1386 |
+
"grad_norm": 0.30859375,
|
| 1387 |
+
"learning_rate": 0.02,
|
| 1388 |
+
"loss": 1.920396327972412,
|
| 1389 |
+
"step": 197
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.050101214574898786,
|
| 1393 |
+
"grad_norm": 0.310546875,
|
| 1394 |
+
"learning_rate": 0.02,
|
| 1395 |
+
"loss": 1.84848952293396,
|
| 1396 |
+
"step": 198
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.05035425101214575,
|
| 1400 |
+
"grad_norm": 0.298828125,
|
| 1401 |
+
"learning_rate": 0.02,
|
| 1402 |
+
"loss": 1.8590221405029297,
|
| 1403 |
+
"step": 199
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.05060728744939271,
|
| 1407 |
+
"grad_norm": 0.54296875,
|
| 1408 |
+
"learning_rate": 0.02,
|
| 1409 |
+
"loss": 1.7710858583450317,
|
| 1410 |
+
"step": 200
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.05086032388663968,
|
| 1414 |
+
"grad_norm": 0.298828125,
|
| 1415 |
+
"learning_rate": 0.02,
|
| 1416 |
+
"loss": 1.8579610586166382,
|
| 1417 |
+
"step": 201
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.051113360323886636,
|
| 1421 |
+
"grad_norm": 0.27734375,
|
| 1422 |
+
"learning_rate": 0.02,
|
| 1423 |
+
"loss": 1.7782834768295288,
|
| 1424 |
+
"step": 202
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.0513663967611336,
|
| 1428 |
+
"grad_norm": 0.28125,
|
| 1429 |
+
"learning_rate": 0.02,
|
| 1430 |
+
"loss": 1.8796536922454834,
|
| 1431 |
+
"step": 203
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.05161943319838057,
|
| 1435 |
+
"grad_norm": 0.375,
|
| 1436 |
+
"learning_rate": 0.02,
|
| 1437 |
+
"loss": 1.708918571472168,
|
| 1438 |
+
"step": 204
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.05187246963562753,
|
| 1442 |
+
"grad_norm": 0.294921875,
|
| 1443 |
+
"learning_rate": 0.02,
|
| 1444 |
+
"loss": 1.7594828605651855,
|
| 1445 |
+
"step": 205
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.05212550607287449,
|
| 1449 |
+
"grad_norm": 0.25390625,
|
| 1450 |
+
"learning_rate": 0.02,
|
| 1451 |
+
"loss": 1.5761895179748535,
|
| 1452 |
+
"step": 206
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.05237854251012146,
|
| 1456 |
+
"grad_norm": 0.283203125,
|
| 1457 |
+
"learning_rate": 0.02,
|
| 1458 |
+
"loss": 1.765464425086975,
|
| 1459 |
+
"step": 207
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.05263157894736842,
|
| 1463 |
+
"grad_norm": 0.287109375,
|
| 1464 |
+
"learning_rate": 0.02,
|
| 1465 |
+
"loss": 1.6899892091751099,
|
| 1466 |
+
"step": 208
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.052884615384615384,
|
| 1470 |
+
"grad_norm": 0.255859375,
|
| 1471 |
+
"learning_rate": 0.02,
|
| 1472 |
+
"loss": 1.9187116622924805,
|
| 1473 |
+
"step": 209
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.05313765182186235,
|
| 1477 |
+
"grad_norm": 0.265625,
|
| 1478 |
+
"learning_rate": 0.02,
|
| 1479 |
+
"loss": 1.7201926708221436,
|
| 1480 |
+
"step": 210
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.05339068825910931,
|
| 1484 |
+
"grad_norm": 0.28125,
|
| 1485 |
+
"learning_rate": 0.02,
|
| 1486 |
+
"loss": 1.8093514442443848,
|
| 1487 |
+
"step": 211
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.053643724696356275,
|
| 1491 |
+
"grad_norm": 0.267578125,
|
| 1492 |
+
"learning_rate": 0.02,
|
| 1493 |
+
"loss": 1.644389271736145,
|
| 1494 |
+
"step": 212
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.05389676113360324,
|
| 1498 |
+
"grad_norm": 0.31640625,
|
| 1499 |
+
"learning_rate": 0.02,
|
| 1500 |
+
"loss": 1.7600808143615723,
|
| 1501 |
+
"step": 213
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.0541497975708502,
|
| 1505 |
+
"grad_norm": 0.28125,
|
| 1506 |
+
"learning_rate": 0.02,
|
| 1507 |
+
"loss": 1.8550102710723877,
|
| 1508 |
+
"step": 214
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.054402834008097166,
|
| 1512 |
+
"grad_norm": 0.26171875,
|
| 1513 |
+
"learning_rate": 0.02,
|
| 1514 |
+
"loss": 1.7401726245880127,
|
| 1515 |
+
"step": 215
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.05465587044534413,
|
| 1519 |
+
"grad_norm": 0.283203125,
|
| 1520 |
+
"learning_rate": 0.02,
|
| 1521 |
+
"loss": 1.9065459966659546,
|
| 1522 |
+
"step": 216
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.05490890688259109,
|
| 1526 |
+
"grad_norm": 0.275390625,
|
| 1527 |
+
"learning_rate": 0.02,
|
| 1528 |
+
"loss": 1.569657802581787,
|
| 1529 |
+
"step": 217
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.05516194331983806,
|
| 1533 |
+
"grad_norm": 0.2470703125,
|
| 1534 |
+
"learning_rate": 0.02,
|
| 1535 |
+
"loss": 1.883676528930664,
|
| 1536 |
+
"step": 218
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.05541497975708502,
|
| 1540 |
+
"grad_norm": 0.2578125,
|
| 1541 |
+
"learning_rate": 0.02,
|
| 1542 |
+
"loss": 1.9213581085205078,
|
| 1543 |
+
"step": 219
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.05566801619433198,
|
| 1547 |
+
"grad_norm": 0.302734375,
|
| 1548 |
+
"learning_rate": 0.02,
|
| 1549 |
+
"loss": 1.7225048542022705,
|
| 1550 |
+
"step": 220
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.05592105263157895,
|
| 1554 |
+
"grad_norm": 0.27734375,
|
| 1555 |
+
"learning_rate": 0.02,
|
| 1556 |
+
"loss": 1.687207579612732,
|
| 1557 |
+
"step": 221
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.056174089068825914,
|
| 1561 |
+
"grad_norm": 0.2890625,
|
| 1562 |
+
"learning_rate": 0.02,
|
| 1563 |
+
"loss": 1.7699179649353027,
|
| 1564 |
+
"step": 222
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.05642712550607287,
|
| 1568 |
+
"grad_norm": 0.28125,
|
| 1569 |
+
"learning_rate": 0.02,
|
| 1570 |
+
"loss": 1.8854470252990723,
|
| 1571 |
+
"step": 223
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.05668016194331984,
|
| 1575 |
+
"grad_norm": 0.294921875,
|
| 1576 |
+
"learning_rate": 0.02,
|
| 1577 |
+
"loss": 1.8999043703079224,
|
| 1578 |
+
"step": 224
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.056933198380566805,
|
| 1582 |
+
"grad_norm": 0.244140625,
|
| 1583 |
+
"learning_rate": 0.02,
|
| 1584 |
+
"loss": 1.863234043121338,
|
| 1585 |
+
"step": 225
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.057186234817813764,
|
| 1589 |
+
"grad_norm": 0.2265625,
|
| 1590 |
+
"learning_rate": 0.02,
|
| 1591 |
+
"loss": 1.7562830448150635,
|
| 1592 |
+
"step": 226
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.05743927125506073,
|
| 1596 |
+
"grad_norm": 0.244140625,
|
| 1597 |
+
"learning_rate": 0.02,
|
| 1598 |
+
"loss": 1.8581035137176514,
|
| 1599 |
+
"step": 227
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.057692307692307696,
|
| 1603 |
+
"grad_norm": 0.28515625,
|
| 1604 |
+
"learning_rate": 0.02,
|
| 1605 |
+
"loss": 1.910494089126587,
|
| 1606 |
+
"step": 228
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.057945344129554655,
|
| 1610 |
+
"grad_norm": 0.287109375,
|
| 1611 |
+
"learning_rate": 0.02,
|
| 1612 |
+
"loss": 1.5930771827697754,
|
| 1613 |
+
"step": 229
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.05819838056680162,
|
| 1617 |
+
"grad_norm": 0.322265625,
|
| 1618 |
+
"learning_rate": 0.02,
|
| 1619 |
+
"loss": 1.9999498128890991,
|
| 1620 |
+
"step": 230
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.058451417004048586,
|
| 1624 |
+
"grad_norm": 0.263671875,
|
| 1625 |
+
"learning_rate": 0.02,
|
| 1626 |
+
"loss": 1.8463115692138672,
|
| 1627 |
+
"step": 231
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.058704453441295545,
|
| 1631 |
+
"grad_norm": 0.255859375,
|
| 1632 |
+
"learning_rate": 0.02,
|
| 1633 |
+
"loss": 1.6839101314544678,
|
| 1634 |
+
"step": 232
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.05895748987854251,
|
| 1638 |
+
"grad_norm": 0.25390625,
|
| 1639 |
+
"learning_rate": 0.02,
|
| 1640 |
+
"loss": 1.6041113138198853,
|
| 1641 |
+
"step": 233
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.05921052631578947,
|
| 1645 |
+
"grad_norm": 0.255859375,
|
| 1646 |
+
"learning_rate": 0.02,
|
| 1647 |
+
"loss": 1.7488685846328735,
|
| 1648 |
+
"step": 234
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.059463562753036436,
|
| 1652 |
+
"grad_norm": 0.2412109375,
|
| 1653 |
+
"learning_rate": 0.02,
|
| 1654 |
+
"loss": 1.7088991403579712,
|
| 1655 |
+
"step": 235
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.0597165991902834,
|
| 1659 |
+
"grad_norm": 0.244140625,
|
| 1660 |
+
"learning_rate": 0.02,
|
| 1661 |
+
"loss": 1.789707899093628,
|
| 1662 |
+
"step": 236
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.05996963562753036,
|
| 1666 |
+
"grad_norm": 0.2470703125,
|
| 1667 |
+
"learning_rate": 0.02,
|
| 1668 |
+
"loss": 1.6540621519088745,
|
| 1669 |
+
"step": 237
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.06022267206477733,
|
| 1673 |
+
"grad_norm": 0.337890625,
|
| 1674 |
+
"learning_rate": 0.02,
|
| 1675 |
+
"loss": 1.831263780593872,
|
| 1676 |
+
"step": 238
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.06047570850202429,
|
| 1680 |
+
"grad_norm": 0.25,
|
| 1681 |
+
"learning_rate": 0.02,
|
| 1682 |
+
"loss": 1.7118489742279053,
|
| 1683 |
+
"step": 239
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.06072874493927125,
|
| 1687 |
+
"grad_norm": 0.2314453125,
|
| 1688 |
+
"learning_rate": 0.02,
|
| 1689 |
+
"loss": 1.7510230541229248,
|
| 1690 |
+
"step": 240
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.06098178137651822,
|
| 1694 |
+
"grad_norm": 0.2216796875,
|
| 1695 |
+
"learning_rate": 0.02,
|
| 1696 |
+
"loss": 1.8188490867614746,
|
| 1697 |
+
"step": 241
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.061234817813765184,
|
| 1701 |
+
"grad_norm": 0.2333984375,
|
| 1702 |
+
"learning_rate": 0.02,
|
| 1703 |
+
"loss": 1.7565772533416748,
|
| 1704 |
+
"step": 242
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.06148785425101214,
|
| 1708 |
+
"grad_norm": 0.2431640625,
|
| 1709 |
+
"learning_rate": 0.02,
|
| 1710 |
+
"loss": 1.8934447765350342,
|
| 1711 |
+
"step": 243
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.06174089068825911,
|
| 1715 |
+
"grad_norm": 0.224609375,
|
| 1716 |
+
"learning_rate": 0.02,
|
| 1717 |
+
"loss": 1.5972774028778076,
|
| 1718 |
+
"step": 244
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.061993927125506075,
|
| 1722 |
+
"grad_norm": 0.24609375,
|
| 1723 |
+
"learning_rate": 0.02,
|
| 1724 |
+
"loss": 1.6532171964645386,
|
| 1725 |
+
"step": 245
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.062246963562753034,
|
| 1729 |
+
"grad_norm": 0.236328125,
|
| 1730 |
+
"learning_rate": 0.02,
|
| 1731 |
+
"loss": 1.8103532791137695,
|
| 1732 |
+
"step": 246
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.0625,
|
| 1736 |
+
"grad_norm": 1.1015625,
|
| 1737 |
+
"learning_rate": 0.02,
|
| 1738 |
+
"loss": 1.623213768005371,
|
| 1739 |
+
"step": 247
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.06275303643724696,
|
| 1743 |
+
"grad_norm": 0.228515625,
|
| 1744 |
+
"learning_rate": 0.02,
|
| 1745 |
+
"loss": 1.5337111949920654,
|
| 1746 |
+
"step": 248
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.06300607287449393,
|
| 1750 |
+
"grad_norm": 0.26171875,
|
| 1751 |
+
"learning_rate": 0.02,
|
| 1752 |
+
"loss": 1.6231749057769775,
|
| 1753 |
+
"step": 249
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.06325910931174089,
|
| 1757 |
+
"grad_norm": 0.255859375,
|
| 1758 |
+
"learning_rate": 0.02,
|
| 1759 |
+
"loss": 1.5969362258911133,
|
| 1760 |
+
"step": 250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.06351214574898785,
|
| 1764 |
+
"grad_norm": 0.27734375,
|
| 1765 |
+
"learning_rate": 0.02,
|
| 1766 |
+
"loss": 1.8024649620056152,
|
| 1767 |
+
"step": 251
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.06376518218623482,
|
| 1771 |
+
"grad_norm": 0.287109375,
|
| 1772 |
+
"learning_rate": 0.02,
|
| 1773 |
+
"loss": 1.7619102001190186,
|
| 1774 |
+
"step": 252
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.06401821862348178,
|
| 1778 |
+
"grad_norm": 0.25,
|
| 1779 |
+
"learning_rate": 0.02,
|
| 1780 |
+
"loss": 1.812684416770935,
|
| 1781 |
+
"step": 253
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.06427125506072874,
|
| 1785 |
+
"grad_norm": 0.388671875,
|
| 1786 |
+
"learning_rate": 0.02,
|
| 1787 |
+
"loss": 1.6886664628982544,
|
| 1788 |
+
"step": 254
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.06452429149797571,
|
| 1792 |
+
"grad_norm": 0.283203125,
|
| 1793 |
+
"learning_rate": 0.02,
|
| 1794 |
+
"loss": 1.746760606765747,
|
| 1795 |
+
"step": 255
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.06477732793522267,
|
| 1799 |
+
"grad_norm": 0.2470703125,
|
| 1800 |
+
"learning_rate": 0.02,
|
| 1801 |
+
"loss": 1.7671563625335693,
|
| 1802 |
+
"step": 256
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.06503036437246963,
|
| 1806 |
+
"grad_norm": 0.240234375,
|
| 1807 |
+
"learning_rate": 0.02,
|
| 1808 |
+
"loss": 1.9502842426300049,
|
| 1809 |
+
"step": 257
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.0652834008097166,
|
| 1813 |
+
"grad_norm": 2.53125,
|
| 1814 |
+
"learning_rate": 0.02,
|
| 1815 |
+
"loss": 1.60953688621521,
|
| 1816 |
+
"step": 258
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.06553643724696356,
|
| 1820 |
+
"grad_norm": 0.3203125,
|
| 1821 |
+
"learning_rate": 0.02,
|
| 1822 |
+
"loss": 1.987889289855957,
|
| 1823 |
+
"step": 259
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.06578947368421052,
|
| 1827 |
+
"grad_norm": 0.251953125,
|
| 1828 |
+
"learning_rate": 0.02,
|
| 1829 |
+
"loss": 1.867415189743042,
|
| 1830 |
+
"step": 260
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.0660425101214575,
|
| 1834 |
+
"grad_norm": 0.287109375,
|
| 1835 |
+
"learning_rate": 0.02,
|
| 1836 |
+
"loss": 1.8441483974456787,
|
| 1837 |
+
"step": 261
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.06629554655870445,
|
| 1841 |
+
"grad_norm": 0.259765625,
|
| 1842 |
+
"learning_rate": 0.02,
|
| 1843 |
+
"loss": 1.7997292280197144,
|
| 1844 |
+
"step": 262
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.06654858299595141,
|
| 1848 |
+
"grad_norm": 0.2490234375,
|
| 1849 |
+
"learning_rate": 0.02,
|
| 1850 |
+
"loss": 1.7468944787979126,
|
| 1851 |
+
"step": 263
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.06680161943319839,
|
| 1855 |
+
"grad_norm": 0.2197265625,
|
| 1856 |
+
"learning_rate": 0.02,
|
| 1857 |
+
"loss": 1.8009114265441895,
|
| 1858 |
+
"step": 264
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.06705465587044535,
|
| 1862 |
+
"grad_norm": 0.232421875,
|
| 1863 |
+
"learning_rate": 0.02,
|
| 1864 |
+
"loss": 1.5228102207183838,
|
| 1865 |
+
"step": 265
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.0673076923076923,
|
| 1869 |
+
"grad_norm": 0.232421875,
|
| 1870 |
+
"learning_rate": 0.02,
|
| 1871 |
+
"loss": 1.817479133605957,
|
| 1872 |
+
"step": 266
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.06756072874493928,
|
| 1876 |
+
"grad_norm": 0.2373046875,
|
| 1877 |
+
"learning_rate": 0.02,
|
| 1878 |
+
"loss": 1.6446462869644165,
|
| 1879 |
+
"step": 267
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.06781376518218624,
|
| 1883 |
+
"grad_norm": 0.25,
|
| 1884 |
+
"learning_rate": 0.02,
|
| 1885 |
+
"loss": 1.8258877992630005,
|
| 1886 |
+
"step": 268
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.0680668016194332,
|
| 1890 |
+
"grad_norm": 0.21484375,
|
| 1891 |
+
"learning_rate": 0.02,
|
| 1892 |
+
"loss": 1.6597766876220703,
|
| 1893 |
+
"step": 269
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.06831983805668017,
|
| 1897 |
+
"grad_norm": 0.2265625,
|
| 1898 |
+
"learning_rate": 0.02,
|
| 1899 |
+
"loss": 1.6289044618606567,
|
| 1900 |
+
"step": 270
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.06857287449392713,
|
| 1904 |
+
"grad_norm": 0.232421875,
|
| 1905 |
+
"learning_rate": 0.02,
|
| 1906 |
+
"loss": 1.7609938383102417,
|
| 1907 |
+
"step": 271
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.06882591093117409,
|
| 1911 |
+
"grad_norm": 0.23046875,
|
| 1912 |
+
"learning_rate": 0.02,
|
| 1913 |
+
"loss": 1.7204265594482422,
|
| 1914 |
+
"step": 272
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.06907894736842106,
|
| 1918 |
+
"grad_norm": 0.2275390625,
|
| 1919 |
+
"learning_rate": 0.02,
|
| 1920 |
+
"loss": 1.6945526599884033,
|
| 1921 |
+
"step": 273
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.06933198380566802,
|
| 1925 |
+
"grad_norm": 0.23046875,
|
| 1926 |
+
"learning_rate": 0.02,
|
| 1927 |
+
"loss": 1.5438995361328125,
|
| 1928 |
+
"step": 274
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.06958502024291498,
|
| 1932 |
+
"grad_norm": 0.28515625,
|
| 1933 |
+
"learning_rate": 0.02,
|
| 1934 |
+
"loss": 1.7701501846313477,
|
| 1935 |
+
"step": 275
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.06983805668016195,
|
| 1939 |
+
"grad_norm": 0.259765625,
|
| 1940 |
+
"learning_rate": 0.02,
|
| 1941 |
+
"loss": 1.6865530014038086,
|
| 1942 |
+
"step": 276
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.07009109311740891,
|
| 1946 |
+
"grad_norm": 0.21875,
|
| 1947 |
+
"learning_rate": 0.02,
|
| 1948 |
+
"loss": 1.6084356307983398,
|
| 1949 |
+
"step": 277
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.07034412955465587,
|
| 1953 |
+
"grad_norm": 0.224609375,
|
| 1954 |
+
"learning_rate": 0.02,
|
| 1955 |
+
"loss": 1.6889640092849731,
|
| 1956 |
+
"step": 278
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.07059716599190283,
|
| 1960 |
+
"grad_norm": 0.2333984375,
|
| 1961 |
+
"learning_rate": 0.02,
|
| 1962 |
+
"loss": 1.555455207824707,
|
| 1963 |
+
"step": 279
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.0708502024291498,
|
| 1967 |
+
"grad_norm": 0.2373046875,
|
| 1968 |
+
"learning_rate": 0.02,
|
| 1969 |
+
"loss": 1.728455662727356,
|
| 1970 |
+
"step": 280
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.07110323886639676,
|
| 1974 |
+
"grad_norm": 0.2216796875,
|
| 1975 |
+
"learning_rate": 0.02,
|
| 1976 |
+
"loss": 1.5139386653900146,
|
| 1977 |
+
"step": 281
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.07135627530364372,
|
| 1981 |
+
"grad_norm": 0.2373046875,
|
| 1982 |
+
"learning_rate": 0.02,
|
| 1983 |
+
"loss": 1.8783767223358154,
|
| 1984 |
+
"step": 282
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.07160931174089069,
|
| 1988 |
+
"grad_norm": 0.2275390625,
|
| 1989 |
+
"learning_rate": 0.02,
|
| 1990 |
+
"loss": 1.6216429471969604,
|
| 1991 |
+
"step": 283
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.07186234817813765,
|
| 1995 |
+
"grad_norm": 0.224609375,
|
| 1996 |
+
"learning_rate": 0.02,
|
| 1997 |
+
"loss": 1.5281383991241455,
|
| 1998 |
+
"step": 284
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.07211538461538461,
|
| 2002 |
+
"grad_norm": 0.2314453125,
|
| 2003 |
+
"learning_rate": 0.02,
|
| 2004 |
+
"loss": 1.6406452655792236,
|
| 2005 |
+
"step": 285
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.07236842105263158,
|
| 2009 |
+
"grad_norm": 0.25,
|
| 2010 |
+
"learning_rate": 0.02,
|
| 2011 |
+
"loss": 1.6633228063583374,
|
| 2012 |
+
"step": 286
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.07262145748987854,
|
| 2016 |
+
"grad_norm": 0.232421875,
|
| 2017 |
+
"learning_rate": 0.02,
|
| 2018 |
+
"loss": 1.5398434400558472,
|
| 2019 |
+
"step": 287
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.0728744939271255,
|
| 2023 |
+
"grad_norm": 0.24609375,
|
| 2024 |
+
"learning_rate": 0.02,
|
| 2025 |
+
"loss": 1.7598520517349243,
|
| 2026 |
+
"step": 288
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.07312753036437247,
|
| 2030 |
+
"grad_norm": 0.26171875,
|
| 2031 |
+
"learning_rate": 0.02,
|
| 2032 |
+
"loss": 1.6095614433288574,
|
| 2033 |
+
"step": 289
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.07338056680161943,
|
| 2037 |
+
"grad_norm": 0.57421875,
|
| 2038 |
+
"learning_rate": 0.02,
|
| 2039 |
+
"loss": 1.6392581462860107,
|
| 2040 |
+
"step": 290
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.07363360323886639,
|
| 2044 |
+
"grad_norm": 0.287109375,
|
| 2045 |
+
"learning_rate": 0.02,
|
| 2046 |
+
"loss": 1.8370375633239746,
|
| 2047 |
+
"step": 291
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.07388663967611336,
|
| 2051 |
+
"grad_norm": 0.216796875,
|
| 2052 |
+
"learning_rate": 0.02,
|
| 2053 |
+
"loss": 1.629447102546692,
|
| 2054 |
+
"step": 292
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.07413967611336032,
|
| 2058 |
+
"grad_norm": 0.2138671875,
|
| 2059 |
+
"learning_rate": 0.02,
|
| 2060 |
+
"loss": 1.8073177337646484,
|
| 2061 |
+
"step": 293
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.07439271255060728,
|
| 2065 |
+
"grad_norm": 0.2177734375,
|
| 2066 |
+
"learning_rate": 0.02,
|
| 2067 |
+
"loss": 1.7319437265396118,
|
| 2068 |
+
"step": 294
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.07464574898785425,
|
| 2072 |
+
"grad_norm": 0.2265625,
|
| 2073 |
+
"learning_rate": 0.02,
|
| 2074 |
+
"loss": 1.8226912021636963,
|
| 2075 |
+
"step": 295
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.07489878542510121,
|
| 2079 |
+
"grad_norm": 0.2119140625,
|
| 2080 |
+
"learning_rate": 0.02,
|
| 2081 |
+
"loss": 1.4903874397277832,
|
| 2082 |
+
"step": 296
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.07515182186234817,
|
| 2086 |
+
"grad_norm": 0.2421875,
|
| 2087 |
+
"learning_rate": 0.02,
|
| 2088 |
+
"loss": 1.5817227363586426,
|
| 2089 |
+
"step": 297
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.07540485829959515,
|
| 2093 |
+
"grad_norm": 0.2421875,
|
| 2094 |
+
"learning_rate": 0.02,
|
| 2095 |
+
"loss": 1.5707061290740967,
|
| 2096 |
+
"step": 298
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.0756578947368421,
|
| 2100 |
+
"grad_norm": 0.2275390625,
|
| 2101 |
+
"learning_rate": 0.02,
|
| 2102 |
+
"loss": 1.8374524116516113,
|
| 2103 |
+
"step": 299
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.07591093117408906,
|
| 2107 |
+
"grad_norm": 0.234375,
|
| 2108 |
+
"learning_rate": 0.02,
|
| 2109 |
+
"loss": 1.8860702514648438,
|
| 2110 |
+
"step": 300
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.07616396761133604,
|
| 2114 |
+
"grad_norm": 0.275390625,
|
| 2115 |
+
"learning_rate": 0.02,
|
| 2116 |
+
"loss": 1.7141436338424683,
|
| 2117 |
+
"step": 301
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.076417004048583,
|
| 2121 |
+
"grad_norm": 0.2412109375,
|
| 2122 |
+
"learning_rate": 0.02,
|
| 2123 |
+
"loss": 1.7473678588867188,
|
| 2124 |
+
"step": 302
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.07667004048582995,
|
| 2128 |
+
"grad_norm": 0.2177734375,
|
| 2129 |
+
"learning_rate": 0.02,
|
| 2130 |
+
"loss": 1.5174469947814941,
|
| 2131 |
+
"step": 303
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.07692307692307693,
|
| 2135 |
+
"grad_norm": 0.2158203125,
|
| 2136 |
+
"learning_rate": 0.02,
|
| 2137 |
+
"loss": 1.7078137397766113,
|
| 2138 |
+
"step": 304
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.07717611336032389,
|
| 2142 |
+
"grad_norm": 0.2236328125,
|
| 2143 |
+
"learning_rate": 0.02,
|
| 2144 |
+
"loss": 1.7914185523986816,
|
| 2145 |
+
"step": 305
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.07742914979757085,
|
| 2149 |
+
"grad_norm": 0.2197265625,
|
| 2150 |
+
"learning_rate": 0.02,
|
| 2151 |
+
"loss": 1.6762712001800537,
|
| 2152 |
+
"step": 306
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.07768218623481782,
|
| 2156 |
+
"grad_norm": 0.248046875,
|
| 2157 |
+
"learning_rate": 0.02,
|
| 2158 |
+
"loss": 1.8586490154266357,
|
| 2159 |
+
"step": 307
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.07793522267206478,
|
| 2163 |
+
"grad_norm": 0.2333984375,
|
| 2164 |
+
"learning_rate": 0.02,
|
| 2165 |
+
"loss": 1.640291690826416,
|
| 2166 |
+
"step": 308
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.07818825910931174,
|
| 2170 |
+
"grad_norm": 0.2451171875,
|
| 2171 |
+
"learning_rate": 0.02,
|
| 2172 |
+
"loss": 1.5084575414657593,
|
| 2173 |
+
"step": 309
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.07844129554655871,
|
| 2177 |
+
"grad_norm": 0.240234375,
|
| 2178 |
+
"learning_rate": 0.02,
|
| 2179 |
+
"loss": 1.6774191856384277,
|
| 2180 |
+
"step": 310
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.07869433198380567,
|
| 2184 |
+
"grad_norm": 0.2255859375,
|
| 2185 |
+
"learning_rate": 0.02,
|
| 2186 |
+
"loss": 1.682057499885559,
|
| 2187 |
+
"step": 311
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.07894736842105263,
|
| 2191 |
+
"grad_norm": 0.2265625,
|
| 2192 |
+
"learning_rate": 0.02,
|
| 2193 |
+
"loss": 1.8212378025054932,
|
| 2194 |
+
"step": 312
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.0792004048582996,
|
| 2198 |
+
"grad_norm": 0.349609375,
|
| 2199 |
+
"learning_rate": 0.02,
|
| 2200 |
+
"loss": 1.6554744243621826,
|
| 2201 |
+
"step": 313
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.07945344129554656,
|
| 2205 |
+
"grad_norm": 0.5078125,
|
| 2206 |
+
"learning_rate": 0.02,
|
| 2207 |
+
"loss": 1.537276268005371,
|
| 2208 |
+
"step": 314
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.07970647773279352,
|
| 2212 |
+
"grad_norm": 0.353515625,
|
| 2213 |
+
"learning_rate": 0.02,
|
| 2214 |
+
"loss": 1.8137785196304321,
|
| 2215 |
+
"step": 315
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.07995951417004049,
|
| 2219 |
+
"grad_norm": 0.2041015625,
|
| 2220 |
+
"learning_rate": 0.02,
|
| 2221 |
+
"loss": 1.6575486660003662,
|
| 2222 |
+
"step": 316
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.08021255060728745,
|
| 2226 |
+
"grad_norm": 0.2119140625,
|
| 2227 |
+
"learning_rate": 0.02,
|
| 2228 |
+
"loss": 1.6336121559143066,
|
| 2229 |
+
"step": 317
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.08046558704453441,
|
| 2233 |
+
"grad_norm": 0.2138671875,
|
| 2234 |
+
"learning_rate": 0.02,
|
| 2235 |
+
"loss": 1.642393946647644,
|
| 2236 |
+
"step": 318
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.08071862348178138,
|
| 2240 |
+
"grad_norm": 0.2431640625,
|
| 2241 |
+
"learning_rate": 0.02,
|
| 2242 |
+
"loss": 1.5971171855926514,
|
| 2243 |
+
"step": 319
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.08097165991902834,
|
| 2247 |
+
"grad_norm": 0.236328125,
|
| 2248 |
+
"learning_rate": 0.02,
|
| 2249 |
+
"loss": 1.6344141960144043,
|
| 2250 |
+
"step": 320
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.0812246963562753,
|
| 2254 |
+
"grad_norm": 0.224609375,
|
| 2255 |
+
"learning_rate": 0.02,
|
| 2256 |
+
"loss": 1.7557191848754883,
|
| 2257 |
+
"step": 321
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.08147773279352227,
|
| 2261 |
+
"grad_norm": 0.2314453125,
|
| 2262 |
+
"learning_rate": 0.02,
|
| 2263 |
+
"loss": 1.662795901298523,
|
| 2264 |
+
"step": 322
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.08173076923076923,
|
| 2268 |
+
"grad_norm": 0.4296875,
|
| 2269 |
+
"learning_rate": 0.02,
|
| 2270 |
+
"loss": 1.8351266384124756,
|
| 2271 |
+
"step": 323
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.08198380566801619,
|
| 2275 |
+
"grad_norm": 0.2255859375,
|
| 2276 |
+
"learning_rate": 0.02,
|
| 2277 |
+
"loss": 1.5431427955627441,
|
| 2278 |
+
"step": 324
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.08223684210526316,
|
| 2282 |
+
"grad_norm": 0.2236328125,
|
| 2283 |
+
"learning_rate": 0.02,
|
| 2284 |
+
"loss": 1.8382834196090698,
|
| 2285 |
+
"step": 325
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.08248987854251012,
|
| 2289 |
+
"grad_norm": 0.2177734375,
|
| 2290 |
+
"learning_rate": 0.02,
|
| 2291 |
+
"loss": 1.6902275085449219,
|
| 2292 |
+
"step": 326
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.08274291497975708,
|
| 2296 |
+
"grad_norm": 0.232421875,
|
| 2297 |
+
"learning_rate": 0.02,
|
| 2298 |
+
"loss": 1.6618757247924805,
|
| 2299 |
+
"step": 327
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.08299595141700405,
|
| 2303 |
+
"grad_norm": 0.2080078125,
|
| 2304 |
+
"learning_rate": 0.02,
|
| 2305 |
+
"loss": 1.5447195768356323,
|
| 2306 |
+
"step": 328
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.08324898785425101,
|
| 2310 |
+
"grad_norm": 0.205078125,
|
| 2311 |
+
"learning_rate": 0.02,
|
| 2312 |
+
"loss": 1.6230026483535767,
|
| 2313 |
+
"step": 329
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.08350202429149797,
|
| 2317 |
+
"grad_norm": 0.203125,
|
| 2318 |
+
"learning_rate": 0.02,
|
| 2319 |
+
"loss": 1.5954384803771973,
|
| 2320 |
+
"step": 330
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.08375506072874495,
|
| 2324 |
+
"grad_norm": 0.2197265625,
|
| 2325 |
+
"learning_rate": 0.02,
|
| 2326 |
+
"loss": 1.5904014110565186,
|
| 2327 |
+
"step": 331
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.0840080971659919,
|
| 2331 |
+
"grad_norm": 0.2392578125,
|
| 2332 |
+
"learning_rate": 0.02,
|
| 2333 |
+
"loss": 1.835031509399414,
|
| 2334 |
+
"step": 332
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.08426113360323886,
|
| 2338 |
+
"grad_norm": 0.21484375,
|
| 2339 |
+
"learning_rate": 0.02,
|
| 2340 |
+
"loss": 1.7414573431015015,
|
| 2341 |
+
"step": 333
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.08451417004048584,
|
| 2345 |
+
"grad_norm": 0.2314453125,
|
| 2346 |
+
"learning_rate": 0.02,
|
| 2347 |
+
"loss": 1.690129280090332,
|
| 2348 |
+
"step": 334
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.0847672064777328,
|
| 2352 |
+
"grad_norm": 0.2470703125,
|
| 2353 |
+
"learning_rate": 0.02,
|
| 2354 |
+
"loss": 1.8188143968582153,
|
| 2355 |
+
"step": 335
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.08502024291497975,
|
| 2359 |
+
"grad_norm": 0.2451171875,
|
| 2360 |
+
"learning_rate": 0.02,
|
| 2361 |
+
"loss": 1.6343677043914795,
|
| 2362 |
+
"step": 336
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.08527327935222673,
|
| 2366 |
+
"grad_norm": 0.2294921875,
|
| 2367 |
+
"learning_rate": 0.02,
|
| 2368 |
+
"loss": 1.6848071813583374,
|
| 2369 |
+
"step": 337
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.08552631578947369,
|
| 2373 |
+
"grad_norm": 0.19921875,
|
| 2374 |
+
"learning_rate": 0.02,
|
| 2375 |
+
"loss": 1.5910496711730957,
|
| 2376 |
+
"step": 338
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.08577935222672065,
|
| 2380 |
+
"grad_norm": 0.2099609375,
|
| 2381 |
+
"learning_rate": 0.02,
|
| 2382 |
+
"loss": 1.6693202257156372,
|
| 2383 |
+
"step": 339
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.0860323886639676,
|
| 2387 |
+
"grad_norm": 0.2265625,
|
| 2388 |
+
"learning_rate": 0.02,
|
| 2389 |
+
"loss": 1.6638545989990234,
|
| 2390 |
+
"step": 340
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.08628542510121458,
|
| 2394 |
+
"grad_norm": 0.2197265625,
|
| 2395 |
+
"learning_rate": 0.02,
|
| 2396 |
+
"loss": 1.6470742225646973,
|
| 2397 |
+
"step": 341
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.08653846153846154,
|
| 2401 |
+
"grad_norm": 0.216796875,
|
| 2402 |
+
"learning_rate": 0.02,
|
| 2403 |
+
"loss": 1.6216919422149658,
|
| 2404 |
+
"step": 342
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.0867914979757085,
|
| 2408 |
+
"grad_norm": 0.2255859375,
|
| 2409 |
+
"learning_rate": 0.02,
|
| 2410 |
+
"loss": 1.6436307430267334,
|
| 2411 |
+
"step": 343
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.08704453441295547,
|
| 2415 |
+
"grad_norm": 0.2265625,
|
| 2416 |
+
"learning_rate": 0.02,
|
| 2417 |
+
"loss": 1.5941998958587646,
|
| 2418 |
+
"step": 344
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.08729757085020243,
|
| 2422 |
+
"grad_norm": 0.2060546875,
|
| 2423 |
+
"learning_rate": 0.02,
|
| 2424 |
+
"loss": 1.6785557270050049,
|
| 2425 |
+
"step": 345
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.08755060728744939,
|
| 2429 |
+
"grad_norm": 0.2060546875,
|
| 2430 |
+
"learning_rate": 0.02,
|
| 2431 |
+
"loss": 1.6645257472991943,
|
| 2432 |
+
"step": 346
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.08780364372469636,
|
| 2436 |
+
"grad_norm": 0.2001953125,
|
| 2437 |
+
"learning_rate": 0.02,
|
| 2438 |
+
"loss": 1.625213861465454,
|
| 2439 |
+
"step": 347
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.08805668016194332,
|
| 2443 |
+
"grad_norm": 0.2119140625,
|
| 2444 |
+
"learning_rate": 0.02,
|
| 2445 |
+
"loss": 1.570495367050171,
|
| 2446 |
+
"step": 348
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.08830971659919028,
|
| 2450 |
+
"grad_norm": 2.484375,
|
| 2451 |
+
"learning_rate": 0.02,
|
| 2452 |
+
"loss": 1.7367191314697266,
|
| 2453 |
+
"step": 349
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.08856275303643725,
|
| 2457 |
+
"grad_norm": 0.224609375,
|
| 2458 |
+
"learning_rate": 0.02,
|
| 2459 |
+
"loss": 1.5887837409973145,
|
| 2460 |
+
"step": 350
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.08881578947368421,
|
| 2464 |
+
"grad_norm": 0.224609375,
|
| 2465 |
+
"learning_rate": 0.02,
|
| 2466 |
+
"loss": 1.63081693649292,
|
| 2467 |
+
"step": 351
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.08906882591093117,
|
| 2471 |
+
"grad_norm": 0.2421875,
|
| 2472 |
+
"learning_rate": 0.02,
|
| 2473 |
+
"loss": 1.5376256704330444,
|
| 2474 |
+
"step": 352
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.08932186234817814,
|
| 2478 |
+
"grad_norm": 0.2255859375,
|
| 2479 |
+
"learning_rate": 0.02,
|
| 2480 |
+
"loss": 1.5460972785949707,
|
| 2481 |
+
"step": 353
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.0895748987854251,
|
| 2485 |
+
"grad_norm": 0.2236328125,
|
| 2486 |
+
"learning_rate": 0.02,
|
| 2487 |
+
"loss": 1.5327677726745605,
|
| 2488 |
+
"step": 354
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.08982793522267206,
|
| 2492 |
+
"grad_norm": 0.2333984375,
|
| 2493 |
+
"learning_rate": 0.02,
|
| 2494 |
+
"loss": 1.70501708984375,
|
| 2495 |
+
"step": 355
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.09008097165991903,
|
| 2499 |
+
"grad_norm": 0.26171875,
|
| 2500 |
+
"learning_rate": 0.02,
|
| 2501 |
+
"loss": 1.8134093284606934,
|
| 2502 |
+
"step": 356
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.09033400809716599,
|
| 2506 |
+
"grad_norm": 0.234375,
|
| 2507 |
+
"learning_rate": 0.02,
|
| 2508 |
+
"loss": 1.509171485900879,
|
| 2509 |
+
"step": 357
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.09058704453441295,
|
| 2513 |
+
"grad_norm": 0.2138671875,
|
| 2514 |
+
"learning_rate": 0.02,
|
| 2515 |
+
"loss": 1.6045968532562256,
|
| 2516 |
+
"step": 358
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.09084008097165992,
|
| 2520 |
+
"grad_norm": 0.2197265625,
|
| 2521 |
+
"learning_rate": 0.02,
|
| 2522 |
+
"loss": 1.6716241836547852,
|
| 2523 |
+
"step": 359
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.09109311740890688,
|
| 2527 |
+
"grad_norm": 0.2412109375,
|
| 2528 |
+
"learning_rate": 0.02,
|
| 2529 |
+
"loss": 1.8341147899627686,
|
| 2530 |
+
"step": 360
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.09134615384615384,
|
| 2534 |
+
"grad_norm": 0.22265625,
|
| 2535 |
+
"learning_rate": 0.02,
|
| 2536 |
+
"loss": 1.7935357093811035,
|
| 2537 |
+
"step": 361
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.09159919028340081,
|
| 2541 |
+
"grad_norm": 0.2080078125,
|
| 2542 |
+
"learning_rate": 0.02,
|
| 2543 |
+
"loss": 1.6881043910980225,
|
| 2544 |
+
"step": 362
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.09185222672064777,
|
| 2548 |
+
"grad_norm": 0.21484375,
|
| 2549 |
+
"learning_rate": 0.02,
|
| 2550 |
+
"loss": 1.6678187847137451,
|
| 2551 |
+
"step": 363
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.09210526315789473,
|
| 2555 |
+
"grad_norm": 0.2265625,
|
| 2556 |
+
"learning_rate": 0.02,
|
| 2557 |
+
"loss": 1.6596988439559937,
|
| 2558 |
+
"step": 364
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.0923582995951417,
|
| 2562 |
+
"grad_norm": 0.20703125,
|
| 2563 |
+
"learning_rate": 0.02,
|
| 2564 |
+
"loss": 1.6384483575820923,
|
| 2565 |
+
"step": 365
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.09261133603238866,
|
| 2569 |
+
"grad_norm": 0.451171875,
|
| 2570 |
+
"learning_rate": 0.02,
|
| 2571 |
+
"loss": 1.7804160118103027,
|
| 2572 |
+
"step": 366
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.09286437246963562,
|
| 2576 |
+
"grad_norm": 0.2197265625,
|
| 2577 |
+
"learning_rate": 0.02,
|
| 2578 |
+
"loss": 1.7853267192840576,
|
| 2579 |
+
"step": 367
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.0931174089068826,
|
| 2583 |
+
"grad_norm": 0.251953125,
|
| 2584 |
+
"learning_rate": 0.02,
|
| 2585 |
+
"loss": 1.6451454162597656,
|
| 2586 |
+
"step": 368
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.09337044534412955,
|
| 2590 |
+
"grad_norm": 0.234375,
|
| 2591 |
+
"learning_rate": 0.02,
|
| 2592 |
+
"loss": 1.8762222528457642,
|
| 2593 |
+
"step": 369
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.09362348178137651,
|
| 2597 |
+
"grad_norm": 0.22265625,
|
| 2598 |
+
"learning_rate": 0.02,
|
| 2599 |
+
"loss": 1.587763786315918,
|
| 2600 |
+
"step": 370
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.09387651821862349,
|
| 2604 |
+
"grad_norm": 0.23828125,
|
| 2605 |
+
"learning_rate": 0.02,
|
| 2606 |
+
"loss": 1.9008150100708008,
|
| 2607 |
+
"step": 371
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.09412955465587045,
|
| 2611 |
+
"grad_norm": 0.2099609375,
|
| 2612 |
+
"learning_rate": 0.02,
|
| 2613 |
+
"loss": 1.3649919033050537,
|
| 2614 |
+
"step": 372
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.0943825910931174,
|
| 2618 |
+
"grad_norm": 0.2255859375,
|
| 2619 |
+
"learning_rate": 0.02,
|
| 2620 |
+
"loss": 1.6461126804351807,
|
| 2621 |
+
"step": 373
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.09463562753036438,
|
| 2625 |
+
"grad_norm": 0.240234375,
|
| 2626 |
+
"learning_rate": 0.02,
|
| 2627 |
+
"loss": 1.7606909275054932,
|
| 2628 |
+
"step": 374
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.09488866396761134,
|
| 2632 |
+
"grad_norm": 0.22265625,
|
| 2633 |
+
"learning_rate": 0.02,
|
| 2634 |
+
"loss": 1.5905332565307617,
|
| 2635 |
+
"step": 375
|
| 2636 |
+
}
|
| 2637 |
+
],
|
| 2638 |
+
"logging_steps": 1,
|
| 2639 |
+
"max_steps": 4348,
|
| 2640 |
+
"num_input_tokens_seen": 0,
|
| 2641 |
+
"num_train_epochs": 2,
|
| 2642 |
+
"save_steps": 870,
|
| 2643 |
+
"stateful_callbacks": {
|
| 2644 |
+
"TrainerControl": {
|
| 2645 |
+
"args": {
|
| 2646 |
+
"should_epoch_stop": false,
|
| 2647 |
+
"should_evaluate": false,
|
| 2648 |
+
"should_log": false,
|
| 2649 |
+
"should_save": false,
|
| 2650 |
+
"should_training_stop": false
|
| 2651 |
+
},
|
| 2652 |
+
"attributes": {}
|
| 2653 |
+
}
|
| 2654 |
+
},
|
| 2655 |
+
"total_flos": 2.598754036421998e+18,
|
| 2656 |
+
"train_batch_size": 14,
|
| 2657 |
+
"trial_name": null,
|
| 2658 |
+
"trial_params": null
|
| 2659 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff0e4d19f148cbb295eb1d6818e5f6a52d84db174c8eb93358b808d726037330
|
| 3 |
+
size 5713
|