Add files using upload-large-folder tool
Browse files- .gitattributes +2 -0
- config.json +40 -0
- generation_config.json +12 -0
- model.safetensors +3 -0
- special_tokens_map.json +17 -0
- tokenizer.json +3 -0
- tokenizer_config.json +3 -0
- trainer_state.json +1442 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer_config.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "HKUSTAudio/Llasa-1B",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 128000,
|
| 9 |
+
"eos_token_id": [
|
| 10 |
+
128001,
|
| 11 |
+
128008,
|
| 12 |
+
128009
|
| 13 |
+
],
|
| 14 |
+
"head_dim": 64,
|
| 15 |
+
"hidden_act": "silu",
|
| 16 |
+
"hidden_size": 2048,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 8192,
|
| 19 |
+
"max_position_embeddings": 131072,
|
| 20 |
+
"mlp_bias": false,
|
| 21 |
+
"model_type": "llama",
|
| 22 |
+
"num_attention_heads": 32,
|
| 23 |
+
"num_hidden_layers": 16,
|
| 24 |
+
"num_key_value_heads": 8,
|
| 25 |
+
"pretraining_tp": 1,
|
| 26 |
+
"rms_norm_eps": 1e-05,
|
| 27 |
+
"rope_scaling": {
|
| 28 |
+
"factor": 32.0,
|
| 29 |
+
"high_freq_factor": 4.0,
|
| 30 |
+
"low_freq_factor": 1.0,
|
| 31 |
+
"original_max_position_embeddings": 8192,
|
| 32 |
+
"rope_type": "llama3"
|
| 33 |
+
},
|
| 34 |
+
"rope_theta": 500000.0,
|
| 35 |
+
"tie_word_embeddings": true,
|
| 36 |
+
"torch_dtype": "bfloat16",
|
| 37 |
+
"transformers_version": "4.48.1",
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 193800
|
| 40 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 128000,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
128001,
|
| 6 |
+
128008,
|
| 7 |
+
128009
|
| 8 |
+
],
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "4.48.1"
|
| 12 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08fe2271b99ae3d444c277018a0b5846d3e07a74d83ec4abed62c19fa764bf3c
|
| 3 |
+
size 2740113872
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|begin_of_text|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|eot_id|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<|end_of_text|>"
|
| 17 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:71d92f3dbf3c23d734e6356241cef149b42fe79848176a54145b6f9a886fd73b
|
| 3 |
+
size 29521206
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c51ef4d5bc8bb8449bc16cd24176a66e0edbea7cbd3d3dbe29f102686cb4068
|
| 3 |
+
size 11710463
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1442 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.14467427771475466,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 1998,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0,
|
| 13 |
+
"eval_loss": 7.394782543182373,
|
| 14 |
+
"eval_runtime": 2.5063,
|
| 15 |
+
"eval_samples_per_second": 4.788,
|
| 16 |
+
"eval_steps_per_second": 1.197,
|
| 17 |
+
"step": 0
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.000724095484057831,
|
| 21 |
+
"grad_norm": 6.9375,
|
| 22 |
+
"learning_rate": 1.2048192771084338e-06,
|
| 23 |
+
"loss": 7.2474,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.001448190968115662,
|
| 28 |
+
"grad_norm": 4.28125,
|
| 29 |
+
"learning_rate": 2.4096385542168676e-06,
|
| 30 |
+
"loss": 7.2559,
|
| 31 |
+
"step": 20
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0021722864521734934,
|
| 35 |
+
"grad_norm": 2.890625,
|
| 36 |
+
"learning_rate": 3.614457831325301e-06,
|
| 37 |
+
"loss": 7.2363,
|
| 38 |
+
"step": 30
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.002896381936231324,
|
| 42 |
+
"grad_norm": 1.9921875,
|
| 43 |
+
"learning_rate": 4.819277108433735e-06,
|
| 44 |
+
"loss": 7.2322,
|
| 45 |
+
"step": 40
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0036204774202891555,
|
| 49 |
+
"grad_norm": 1.8125,
|
| 50 |
+
"learning_rate": 6.024096385542169e-06,
|
| 51 |
+
"loss": 7.1863,
|
| 52 |
+
"step": 50
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.004344572904346987,
|
| 56 |
+
"grad_norm": 1.5625,
|
| 57 |
+
"learning_rate": 7.228915662650602e-06,
|
| 58 |
+
"loss": 7.1754,
|
| 59 |
+
"step": 60
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.005068668388404818,
|
| 63 |
+
"grad_norm": 1.7265625,
|
| 64 |
+
"learning_rate": 8.433734939759036e-06,
|
| 65 |
+
"loss": 7.1641,
|
| 66 |
+
"step": 70
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.005792763872462648,
|
| 70 |
+
"grad_norm": 1.6328125,
|
| 71 |
+
"learning_rate": 9.63855421686747e-06,
|
| 72 |
+
"loss": 7.1237,
|
| 73 |
+
"step": 80
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.00651685935652048,
|
| 77 |
+
"grad_norm": 1.6640625,
|
| 78 |
+
"learning_rate": 1.0843373493975904e-05,
|
| 79 |
+
"loss": 7.0891,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.007240954840578311,
|
| 84 |
+
"grad_norm": 1.5625,
|
| 85 |
+
"learning_rate": 1.2048192771084338e-05,
|
| 86 |
+
"loss": 7.0901,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.007965050324636142,
|
| 91 |
+
"grad_norm": 1.4453125,
|
| 92 |
+
"learning_rate": 1.3253012048192772e-05,
|
| 93 |
+
"loss": 7.0337,
|
| 94 |
+
"step": 110
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.008689145808693973,
|
| 98 |
+
"grad_norm": 1.953125,
|
| 99 |
+
"learning_rate": 1.4457831325301205e-05,
|
| 100 |
+
"loss": 7.0387,
|
| 101 |
+
"step": 120
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.009413241292751805,
|
| 105 |
+
"grad_norm": 1.4375,
|
| 106 |
+
"learning_rate": 1.566265060240964e-05,
|
| 107 |
+
"loss": 7.018,
|
| 108 |
+
"step": 130
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.010137336776809636,
|
| 112 |
+
"grad_norm": 1.3515625,
|
| 113 |
+
"learning_rate": 1.6867469879518073e-05,
|
| 114 |
+
"loss": 6.9834,
|
| 115 |
+
"step": 140
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.010861432260867467,
|
| 119 |
+
"grad_norm": 1.4609375,
|
| 120 |
+
"learning_rate": 1.8072289156626505e-05,
|
| 121 |
+
"loss": 6.9829,
|
| 122 |
+
"step": 150
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.011585527744925297,
|
| 126 |
+
"grad_norm": 1.3515625,
|
| 127 |
+
"learning_rate": 1.927710843373494e-05,
|
| 128 |
+
"loss": 6.9724,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.012309623228983128,
|
| 133 |
+
"grad_norm": 1.390625,
|
| 134 |
+
"learning_rate": 2.0481927710843373e-05,
|
| 135 |
+
"loss": 6.9671,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.01303371871304096,
|
| 140 |
+
"grad_norm": 1.4921875,
|
| 141 |
+
"learning_rate": 2.168674698795181e-05,
|
| 142 |
+
"loss": 6.9373,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.01375781419709879,
|
| 147 |
+
"grad_norm": 1.71875,
|
| 148 |
+
"learning_rate": 2.289156626506024e-05,
|
| 149 |
+
"loss": 6.9478,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.014481909681156622,
|
| 154 |
+
"grad_norm": 1.40625,
|
| 155 |
+
"learning_rate": 2.4096385542168677e-05,
|
| 156 |
+
"loss": 6.9258,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.015206005165214453,
|
| 161 |
+
"grad_norm": 1.5625,
|
| 162 |
+
"learning_rate": 2.530120481927711e-05,
|
| 163 |
+
"loss": 6.9085,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.015930100649272284,
|
| 168 |
+
"grad_norm": 1.453125,
|
| 169 |
+
"learning_rate": 2.6506024096385545e-05,
|
| 170 |
+
"loss": 6.9023,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.016654196133330114,
|
| 175 |
+
"grad_norm": 2.125,
|
| 176 |
+
"learning_rate": 2.7710843373493977e-05,
|
| 177 |
+
"loss": 6.9096,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.017378291617387947,
|
| 182 |
+
"grad_norm": 1.453125,
|
| 183 |
+
"learning_rate": 2.891566265060241e-05,
|
| 184 |
+
"loss": 6.8828,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.018102387101445776,
|
| 189 |
+
"grad_norm": 1.4609375,
|
| 190 |
+
"learning_rate": 3.012048192771085e-05,
|
| 191 |
+
"loss": 6.8763,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.01882648258550361,
|
| 196 |
+
"grad_norm": 1.4140625,
|
| 197 |
+
"learning_rate": 3.132530120481928e-05,
|
| 198 |
+
"loss": 6.875,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.01955057806956144,
|
| 203 |
+
"grad_norm": 1.6484375,
|
| 204 |
+
"learning_rate": 3.253012048192771e-05,
|
| 205 |
+
"loss": 6.8641,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.020274673553619272,
|
| 210 |
+
"grad_norm": 1.4453125,
|
| 211 |
+
"learning_rate": 3.3734939759036146e-05,
|
| 212 |
+
"loss": 6.8531,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0209987690376771,
|
| 217 |
+
"grad_norm": 1.59375,
|
| 218 |
+
"learning_rate": 3.4939759036144585e-05,
|
| 219 |
+
"loss": 6.8722,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.021722864521734934,
|
| 224 |
+
"grad_norm": 1.5078125,
|
| 225 |
+
"learning_rate": 3.614457831325301e-05,
|
| 226 |
+
"loss": 6.8478,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.022446960005792764,
|
| 231 |
+
"grad_norm": 1.5859375,
|
| 232 |
+
"learning_rate": 3.734939759036144e-05,
|
| 233 |
+
"loss": 6.8374,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.023171055489850594,
|
| 238 |
+
"grad_norm": 1.4609375,
|
| 239 |
+
"learning_rate": 3.855421686746988e-05,
|
| 240 |
+
"loss": 6.8136,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.023895150973908427,
|
| 245 |
+
"grad_norm": 1.453125,
|
| 246 |
+
"learning_rate": 3.9759036144578314e-05,
|
| 247 |
+
"loss": 6.833,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.024619246457966256,
|
| 252 |
+
"grad_norm": 1.4765625,
|
| 253 |
+
"learning_rate": 4.0963855421686746e-05,
|
| 254 |
+
"loss": 6.8183,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02534334194202409,
|
| 259 |
+
"grad_norm": 1.53125,
|
| 260 |
+
"learning_rate": 4.2168674698795186e-05,
|
| 261 |
+
"loss": 6.8071,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.02606743742608192,
|
| 266 |
+
"grad_norm": 1.921875,
|
| 267 |
+
"learning_rate": 4.337349397590362e-05,
|
| 268 |
+
"loss": 6.8177,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02679153291013975,
|
| 273 |
+
"grad_norm": 1.484375,
|
| 274 |
+
"learning_rate": 4.457831325301205e-05,
|
| 275 |
+
"loss": 6.797,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.02751562839419758,
|
| 280 |
+
"grad_norm": 1.515625,
|
| 281 |
+
"learning_rate": 4.578313253012048e-05,
|
| 282 |
+
"loss": 6.7979,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.028239723878255414,
|
| 287 |
+
"grad_norm": 1.6015625,
|
| 288 |
+
"learning_rate": 4.698795180722892e-05,
|
| 289 |
+
"loss": 6.7899,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.028963819362313244,
|
| 294 |
+
"grad_norm": 1.546875,
|
| 295 |
+
"learning_rate": 4.8192771084337354e-05,
|
| 296 |
+
"loss": 6.7614,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.029687914846371073,
|
| 301 |
+
"grad_norm": 1.5234375,
|
| 302 |
+
"learning_rate": 4.9397590361445786e-05,
|
| 303 |
+
"loss": 6.8134,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.030412010330428906,
|
| 308 |
+
"grad_norm": 1.5078125,
|
| 309 |
+
"learning_rate": 4.999998281045896e-05,
|
| 310 |
+
"loss": 6.7845,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.031136105814486736,
|
| 315 |
+
"grad_norm": 1.765625,
|
| 316 |
+
"learning_rate": 4.99998452942724e-05,
|
| 317 |
+
"loss": 6.7727,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.03186020129854457,
|
| 322 |
+
"grad_norm": 1.484375,
|
| 323 |
+
"learning_rate": 4.9999570262655714e-05,
|
| 324 |
+
"loss": 6.7537,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.0325842967826024,
|
| 329 |
+
"grad_norm": 1.53125,
|
| 330 |
+
"learning_rate": 4.9999157717121745e-05,
|
| 331 |
+
"loss": 6.7486,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.03330839226666023,
|
| 336 |
+
"grad_norm": 1.359375,
|
| 337 |
+
"learning_rate": 4.9998607659939787e-05,
|
| 338 |
+
"loss": 6.7585,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.034032487750718064,
|
| 343 |
+
"grad_norm": 1.4375,
|
| 344 |
+
"learning_rate": 4.999792009413549e-05,
|
| 345 |
+
"loss": 6.7467,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.034756583234775894,
|
| 350 |
+
"grad_norm": 1.421875,
|
| 351 |
+
"learning_rate": 4.999709502349091e-05,
|
| 352 |
+
"loss": 6.7446,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.03548067871883372,
|
| 357 |
+
"grad_norm": 1.453125,
|
| 358 |
+
"learning_rate": 4.9996132452544484e-05,
|
| 359 |
+
"loss": 6.7382,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.03620477420289155,
|
| 364 |
+
"grad_norm": 1.4296875,
|
| 365 |
+
"learning_rate": 4.9995032386590986e-05,
|
| 366 |
+
"loss": 6.7563,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.03692886968694938,
|
| 371 |
+
"grad_norm": 1.390625,
|
| 372 |
+
"learning_rate": 4.999379483168148e-05,
|
| 373 |
+
"loss": 6.7614,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.03765296517100722,
|
| 378 |
+
"grad_norm": 1.46875,
|
| 379 |
+
"learning_rate": 4.999241979462333e-05,
|
| 380 |
+
"loss": 6.7422,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.03837706065506505,
|
| 385 |
+
"grad_norm": 1.4296875,
|
| 386 |
+
"learning_rate": 4.999090728298014e-05,
|
| 387 |
+
"loss": 6.7343,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.03910115613912288,
|
| 392 |
+
"grad_norm": 1.390625,
|
| 393 |
+
"learning_rate": 4.9989257305071704e-05,
|
| 394 |
+
"loss": 6.7052,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.03982525162318071,
|
| 399 |
+
"grad_norm": 1.4609375,
|
| 400 |
+
"learning_rate": 4.998746986997397e-05,
|
| 401 |
+
"loss": 6.7377,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.040549347107238544,
|
| 406 |
+
"grad_norm": 1.640625,
|
| 407 |
+
"learning_rate": 4.9985544987519e-05,
|
| 408 |
+
"loss": 6.7441,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.04127344259129637,
|
| 413 |
+
"grad_norm": 1.296875,
|
| 414 |
+
"learning_rate": 4.9983482668294905e-05,
|
| 415 |
+
"loss": 6.7001,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.0419975380753542,
|
| 420 |
+
"grad_norm": 1.4375,
|
| 421 |
+
"learning_rate": 4.998128292364578e-05,
|
| 422 |
+
"loss": 6.7151,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.04272163355941203,
|
| 427 |
+
"grad_norm": 1.453125,
|
| 428 |
+
"learning_rate": 4.997894576567166e-05,
|
| 429 |
+
"loss": 6.7119,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.04344572904346987,
|
| 434 |
+
"grad_norm": 1.4609375,
|
| 435 |
+
"learning_rate": 4.9976471207228426e-05,
|
| 436 |
+
"loss": 6.7282,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.0441698245275277,
|
| 441 |
+
"grad_norm": 1.28125,
|
| 442 |
+
"learning_rate": 4.9973859261927755e-05,
|
| 443 |
+
"loss": 6.726,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.04489392001158553,
|
| 448 |
+
"grad_norm": 1.515625,
|
| 449 |
+
"learning_rate": 4.997110994413707e-05,
|
| 450 |
+
"loss": 6.7266,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.04561801549564336,
|
| 455 |
+
"grad_norm": 1.34375,
|
| 456 |
+
"learning_rate": 4.996822326897939e-05,
|
| 457 |
+
"loss": 6.7016,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.04634211097970119,
|
| 462 |
+
"grad_norm": 1.4453125,
|
| 463 |
+
"learning_rate": 4.9965199252333315e-05,
|
| 464 |
+
"loss": 6.6972,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.047066206463759024,
|
| 469 |
+
"grad_norm": 1.3515625,
|
| 470 |
+
"learning_rate": 4.996203791083291e-05,
|
| 471 |
+
"loss": 6.6964,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.04779030194781685,
|
| 476 |
+
"grad_norm": 1.390625,
|
| 477 |
+
"learning_rate": 4.99587392618676e-05,
|
| 478 |
+
"loss": 6.6926,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.04851439743187468,
|
| 483 |
+
"grad_norm": 1.34375,
|
| 484 |
+
"learning_rate": 4.995530332358211e-05,
|
| 485 |
+
"loss": 6.7038,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.04923849291593251,
|
| 490 |
+
"grad_norm": 1.375,
|
| 491 |
+
"learning_rate": 4.995173011487635e-05,
|
| 492 |
+
"loss": 6.7012,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.04996258839999035,
|
| 497 |
+
"grad_norm": 1.375,
|
| 498 |
+
"learning_rate": 4.994801965540527e-05,
|
| 499 |
+
"loss": 6.7019,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.05068668388404818,
|
| 504 |
+
"grad_norm": 1.4921875,
|
| 505 |
+
"learning_rate": 4.9944171965578836e-05,
|
| 506 |
+
"loss": 6.6656,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.05141077936810601,
|
| 511 |
+
"grad_norm": 1.390625,
|
| 512 |
+
"learning_rate": 4.9940187066561825e-05,
|
| 513 |
+
"loss": 6.6811,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.05213487485216384,
|
| 518 |
+
"grad_norm": 1.3515625,
|
| 519 |
+
"learning_rate": 4.99360649802738e-05,
|
| 520 |
+
"loss": 6.6885,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.05285897033622167,
|
| 525 |
+
"grad_norm": 1.40625,
|
| 526 |
+
"learning_rate": 4.9931805729388916e-05,
|
| 527 |
+
"loss": 6.69,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.0535830658202795,
|
| 532 |
+
"grad_norm": 1.3203125,
|
| 533 |
+
"learning_rate": 4.992740933733581e-05,
|
| 534 |
+
"loss": 6.675,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.05430716130433733,
|
| 539 |
+
"grad_norm": 1.3671875,
|
| 540 |
+
"learning_rate": 4.992287582829751e-05,
|
| 541 |
+
"loss": 6.6997,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.05503125678839516,
|
| 546 |
+
"grad_norm": 1.390625,
|
| 547 |
+
"learning_rate": 4.991820522721128e-05,
|
| 548 |
+
"loss": 6.6923,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.05575535227245299,
|
| 553 |
+
"grad_norm": 1.40625,
|
| 554 |
+
"learning_rate": 4.991339755976845e-05,
|
| 555 |
+
"loss": 6.6936,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.05647944775651083,
|
| 560 |
+
"grad_norm": 1.515625,
|
| 561 |
+
"learning_rate": 4.990845285241434e-05,
|
| 562 |
+
"loss": 6.6684,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.05720354324056866,
|
| 567 |
+
"grad_norm": 1.328125,
|
| 568 |
+
"learning_rate": 4.9903371132348046e-05,
|
| 569 |
+
"loss": 6.6656,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.05792763872462649,
|
| 574 |
+
"grad_norm": 1.34375,
|
| 575 |
+
"learning_rate": 4.9898152427522334e-05,
|
| 576 |
+
"loss": 6.6807,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.05865173420868432,
|
| 581 |
+
"grad_norm": 1.25,
|
| 582 |
+
"learning_rate": 4.9892796766643503e-05,
|
| 583 |
+
"loss": 6.6901,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.059375829692742146,
|
| 588 |
+
"grad_norm": 1.4140625,
|
| 589 |
+
"learning_rate": 4.988730417917115e-05,
|
| 590 |
+
"loss": 6.6851,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.06009992517679998,
|
| 595 |
+
"grad_norm": 1.5859375,
|
| 596 |
+
"learning_rate": 4.98816746953181e-05,
|
| 597 |
+
"loss": 6.6608,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.06082402066085781,
|
| 602 |
+
"grad_norm": 1.3359375,
|
| 603 |
+
"learning_rate": 4.987590834605017e-05,
|
| 604 |
+
"loss": 6.6617,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.06154811614491564,
|
| 609 |
+
"grad_norm": 1.28125,
|
| 610 |
+
"learning_rate": 4.9870005163086045e-05,
|
| 611 |
+
"loss": 6.6574,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.06227221162897347,
|
| 616 |
+
"grad_norm": 1.3359375,
|
| 617 |
+
"learning_rate": 4.986396517889706e-05,
|
| 618 |
+
"loss": 6.6633,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.06299630711303131,
|
| 623 |
+
"grad_norm": 1.3203125,
|
| 624 |
+
"learning_rate": 4.985778842670707e-05,
|
| 625 |
+
"loss": 6.6807,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.06372040259708914,
|
| 630 |
+
"grad_norm": 1.3515625,
|
| 631 |
+
"learning_rate": 4.9851474940492256e-05,
|
| 632 |
+
"loss": 6.6425,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.06444449808114697,
|
| 637 |
+
"grad_norm": 1.3984375,
|
| 638 |
+
"learning_rate": 4.9845024754980876e-05,
|
| 639 |
+
"loss": 6.658,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.0651685935652048,
|
| 644 |
+
"grad_norm": 1.4453125,
|
| 645 |
+
"learning_rate": 4.983843790565316e-05,
|
| 646 |
+
"loss": 6.6633,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.06589268904926263,
|
| 651 |
+
"grad_norm": 1.3671875,
|
| 652 |
+
"learning_rate": 4.983171442874106e-05,
|
| 653 |
+
"loss": 6.6828,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.06661678453332046,
|
| 658 |
+
"grad_norm": 1.3515625,
|
| 659 |
+
"learning_rate": 4.982485436122809e-05,
|
| 660 |
+
"loss": 6.6251,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.06734088001737829,
|
| 665 |
+
"grad_norm": 1.2578125,
|
| 666 |
+
"learning_rate": 4.981785774084909e-05,
|
| 667 |
+
"loss": 6.6573,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.06806497550143613,
|
| 672 |
+
"grad_norm": 1.3828125,
|
| 673 |
+
"learning_rate": 4.981072460609002e-05,
|
| 674 |
+
"loss": 6.6677,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.06878907098549396,
|
| 679 |
+
"grad_norm": 1.3125,
|
| 680 |
+
"learning_rate": 4.980345499618778e-05,
|
| 681 |
+
"loss": 6.6577,
|
| 682 |
+
"step": 950
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.06951316646955179,
|
| 686 |
+
"grad_norm": 1.2890625,
|
| 687 |
+
"learning_rate": 4.9796048951129936e-05,
|
| 688 |
+
"loss": 6.6628,
|
| 689 |
+
"step": 960
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.07023726195360962,
|
| 693 |
+
"grad_norm": 1.3984375,
|
| 694 |
+
"learning_rate": 4.978850651165458e-05,
|
| 695 |
+
"loss": 6.6443,
|
| 696 |
+
"step": 970
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.07096135743766745,
|
| 700 |
+
"grad_norm": 1.421875,
|
| 701 |
+
"learning_rate": 4.9780827719250035e-05,
|
| 702 |
+
"loss": 6.6617,
|
| 703 |
+
"step": 980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.07168545292172528,
|
| 707 |
+
"grad_norm": 1.3515625,
|
| 708 |
+
"learning_rate": 4.977301261615465e-05,
|
| 709 |
+
"loss": 6.6679,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.0724095484057831,
|
| 714 |
+
"grad_norm": 1.34375,
|
| 715 |
+
"learning_rate": 4.9765061245356594e-05,
|
| 716 |
+
"loss": 6.6385,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.0724095484057831,
|
| 721 |
+
"eval_loss": 6.789700984954834,
|
| 722 |
+
"eval_runtime": 1.6435,
|
| 723 |
+
"eval_samples_per_second": 7.302,
|
| 724 |
+
"eval_steps_per_second": 1.825,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.07313364388984094,
|
| 729 |
+
"grad_norm": 1.296875,
|
| 730 |
+
"learning_rate": 4.9756973650593583e-05,
|
| 731 |
+
"loss": 6.6351,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.07385773937389876,
|
| 736 |
+
"grad_norm": 1.3828125,
|
| 737 |
+
"learning_rate": 4.9748749876352655e-05,
|
| 738 |
+
"loss": 6.6602,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.07458183485795661,
|
| 743 |
+
"grad_norm": 1.3203125,
|
| 744 |
+
"learning_rate": 4.974038996786991e-05,
|
| 745 |
+
"loss": 6.653,
|
| 746 |
+
"step": 1030
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.07530593034201444,
|
| 750 |
+
"grad_norm": 1.4453125,
|
| 751 |
+
"learning_rate": 4.9731893971130306e-05,
|
| 752 |
+
"loss": 6.6532,
|
| 753 |
+
"step": 1040
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.07603002582607227,
|
| 757 |
+
"grad_norm": 1.421875,
|
| 758 |
+
"learning_rate": 4.972326193286736e-05,
|
| 759 |
+
"loss": 6.6297,
|
| 760 |
+
"step": 1050
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.0767541213101301,
|
| 764 |
+
"grad_norm": 1.3828125,
|
| 765 |
+
"learning_rate": 4.971449390056289e-05,
|
| 766 |
+
"loss": 6.6379,
|
| 767 |
+
"step": 1060
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.07747821679418793,
|
| 771 |
+
"grad_norm": 1.2890625,
|
| 772 |
+
"learning_rate": 4.970558992244679e-05,
|
| 773 |
+
"loss": 6.6478,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.07820231227824576,
|
| 778 |
+
"grad_norm": 1.4609375,
|
| 779 |
+
"learning_rate": 4.969655004749674e-05,
|
| 780 |
+
"loss": 6.6616,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.07892640776230359,
|
| 785 |
+
"grad_norm": 1.4296875,
|
| 786 |
+
"learning_rate": 4.968737432543794e-05,
|
| 787 |
+
"loss": 6.6314,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.07965050324636141,
|
| 792 |
+
"grad_norm": 1.2265625,
|
| 793 |
+
"learning_rate": 4.967806280674283e-05,
|
| 794 |
+
"loss": 6.6465,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.08037459873041924,
|
| 799 |
+
"grad_norm": 1.5390625,
|
| 800 |
+
"learning_rate": 4.9668615542630836e-05,
|
| 801 |
+
"loss": 6.6517,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.08109869421447709,
|
| 806 |
+
"grad_norm": 1.2734375,
|
| 807 |
+
"learning_rate": 4.965903258506806e-05,
|
| 808 |
+
"loss": 6.6278,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.08182278969853492,
|
| 813 |
+
"grad_norm": 1.3125,
|
| 814 |
+
"learning_rate": 4.964931398676701e-05,
|
| 815 |
+
"loss": 6.6246,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.08254688518259275,
|
| 820 |
+
"grad_norm": 1.296875,
|
| 821 |
+
"learning_rate": 4.9639459801186306e-05,
|
| 822 |
+
"loss": 6.616,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.08327098066665058,
|
| 827 |
+
"grad_norm": 1.3515625,
|
| 828 |
+
"learning_rate": 4.962947008253038e-05,
|
| 829 |
+
"loss": 6.6492,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.0839950761507084,
|
| 834 |
+
"grad_norm": 1.390625,
|
| 835 |
+
"learning_rate": 4.96193448857492e-05,
|
| 836 |
+
"loss": 6.6096,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.08471917163476624,
|
| 841 |
+
"grad_norm": 1.421875,
|
| 842 |
+
"learning_rate": 4.960908426653793e-05,
|
| 843 |
+
"loss": 6.6282,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.08544326711882406,
|
| 848 |
+
"grad_norm": 1.296875,
|
| 849 |
+
"learning_rate": 4.959868828133667e-05,
|
| 850 |
+
"loss": 6.6053,
|
| 851 |
+
"step": 1180
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.0861673626028819,
|
| 855 |
+
"grad_norm": 1.3359375,
|
| 856 |
+
"learning_rate": 4.95881569873301e-05,
|
| 857 |
+
"loss": 6.6031,
|
| 858 |
+
"step": 1190
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.08689145808693974,
|
| 862 |
+
"grad_norm": 1.3984375,
|
| 863 |
+
"learning_rate": 4.95774904424472e-05,
|
| 864 |
+
"loss": 6.609,
|
| 865 |
+
"step": 1200
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.08761555357099757,
|
| 869 |
+
"grad_norm": 1.34375,
|
| 870 |
+
"learning_rate": 4.9566688705360915e-05,
|
| 871 |
+
"loss": 6.6291,
|
| 872 |
+
"step": 1210
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.0883396490550554,
|
| 876 |
+
"grad_norm": 1.296875,
|
| 877 |
+
"learning_rate": 4.955575183548783e-05,
|
| 878 |
+
"loss": 6.6252,
|
| 879 |
+
"step": 1220
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.08906374453911323,
|
| 883 |
+
"grad_norm": 1.296875,
|
| 884 |
+
"learning_rate": 4.9544679892987855e-05,
|
| 885 |
+
"loss": 6.6246,
|
| 886 |
+
"step": 1230
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.08978784002317106,
|
| 890 |
+
"grad_norm": 1.28125,
|
| 891 |
+
"learning_rate": 4.9533472938763884e-05,
|
| 892 |
+
"loss": 6.6155,
|
| 893 |
+
"step": 1240
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 0.09051193550722889,
|
| 897 |
+
"grad_norm": 1.3203125,
|
| 898 |
+
"learning_rate": 4.9522131034461464e-05,
|
| 899 |
+
"loss": 6.6263,
|
| 900 |
+
"step": 1250
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 0.09123603099128672,
|
| 904 |
+
"grad_norm": 1.328125,
|
| 905 |
+
"learning_rate": 4.951065424246845e-05,
|
| 906 |
+
"loss": 6.6219,
|
| 907 |
+
"step": 1260
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 0.09196012647534454,
|
| 911 |
+
"grad_norm": 1.375,
|
| 912 |
+
"learning_rate": 4.9499042625914674e-05,
|
| 913 |
+
"loss": 6.6144,
|
| 914 |
+
"step": 1270
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.09268422195940237,
|
| 918 |
+
"grad_norm": 1.3984375,
|
| 919 |
+
"learning_rate": 4.9487296248671586e-05,
|
| 920 |
+
"loss": 6.6153,
|
| 921 |
+
"step": 1280
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.09340831744346022,
|
| 925 |
+
"grad_norm": 1.2421875,
|
| 926 |
+
"learning_rate": 4.9475415175351926e-05,
|
| 927 |
+
"loss": 6.6136,
|
| 928 |
+
"step": 1290
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 0.09413241292751805,
|
| 932 |
+
"grad_norm": 1.34375,
|
| 933 |
+
"learning_rate": 4.946339947130933e-05,
|
| 934 |
+
"loss": 6.586,
|
| 935 |
+
"step": 1300
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 0.09485650841157588,
|
| 939 |
+
"grad_norm": 1.359375,
|
| 940 |
+
"learning_rate": 4.945124920263797e-05,
|
| 941 |
+
"loss": 6.6185,
|
| 942 |
+
"step": 1310
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 0.0955806038956337,
|
| 946 |
+
"grad_norm": 1.3359375,
|
| 947 |
+
"learning_rate": 4.9438964436172264e-05,
|
| 948 |
+
"loss": 6.6248,
|
| 949 |
+
"step": 1320
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 0.09630469937969154,
|
| 953 |
+
"grad_norm": 1.2890625,
|
| 954 |
+
"learning_rate": 4.9426545239486424e-05,
|
| 955 |
+
"loss": 6.605,
|
| 956 |
+
"step": 1330
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 0.09702879486374937,
|
| 960 |
+
"grad_norm": 1.3125,
|
| 961 |
+
"learning_rate": 4.94139916808941e-05,
|
| 962 |
+
"loss": 6.6525,
|
| 963 |
+
"step": 1340
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 0.0977528903478072,
|
| 967 |
+
"grad_norm": 1.2421875,
|
| 968 |
+
"learning_rate": 4.9401303829448075e-05,
|
| 969 |
+
"loss": 6.6178,
|
| 970 |
+
"step": 1350
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.09847698583186502,
|
| 974 |
+
"grad_norm": 1.28125,
|
| 975 |
+
"learning_rate": 4.938848175493976e-05,
|
| 976 |
+
"loss": 6.5978,
|
| 977 |
+
"step": 1360
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.09920108131592285,
|
| 981 |
+
"grad_norm": 1.296875,
|
| 982 |
+
"learning_rate": 4.937552552789894e-05,
|
| 983 |
+
"loss": 6.6059,
|
| 984 |
+
"step": 1370
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.0999251767999807,
|
| 988 |
+
"grad_norm": 1.3203125,
|
| 989 |
+
"learning_rate": 4.936243521959329e-05,
|
| 990 |
+
"loss": 6.6078,
|
| 991 |
+
"step": 1380
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.10064927228403853,
|
| 995 |
+
"grad_norm": 1.2890625,
|
| 996 |
+
"learning_rate": 4.934921090202803e-05,
|
| 997 |
+
"loss": 6.6135,
|
| 998 |
+
"step": 1390
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.10137336776809636,
|
| 1002 |
+
"grad_norm": 1.375,
|
| 1003 |
+
"learning_rate": 4.933585264794552e-05,
|
| 1004 |
+
"loss": 6.6068,
|
| 1005 |
+
"step": 1400
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 0.10209746325215419,
|
| 1009 |
+
"grad_norm": 1.34375,
|
| 1010 |
+
"learning_rate": 4.932236053082486e-05,
|
| 1011 |
+
"loss": 6.6026,
|
| 1012 |
+
"step": 1410
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 0.10282155873621202,
|
| 1016 |
+
"grad_norm": 1.2578125,
|
| 1017 |
+
"learning_rate": 4.930873462488148e-05,
|
| 1018 |
+
"loss": 6.6064,
|
| 1019 |
+
"step": 1420
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 0.10354565422026984,
|
| 1023 |
+
"grad_norm": 1.375,
|
| 1024 |
+
"learning_rate": 4.9294975005066734e-05,
|
| 1025 |
+
"loss": 6.6075,
|
| 1026 |
+
"step": 1430
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 0.10426974970432767,
|
| 1030 |
+
"grad_norm": 1.2421875,
|
| 1031 |
+
"learning_rate": 4.9281081747067494e-05,
|
| 1032 |
+
"loss": 6.5982,
|
| 1033 |
+
"step": 1440
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 0.1049938451883855,
|
| 1037 |
+
"grad_norm": 1.484375,
|
| 1038 |
+
"learning_rate": 4.926705492730572e-05,
|
| 1039 |
+
"loss": 6.6109,
|
| 1040 |
+
"step": 1450
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.10571794067244333,
|
| 1044 |
+
"grad_norm": 1.3828125,
|
| 1045 |
+
"learning_rate": 4.925289462293807e-05,
|
| 1046 |
+
"loss": 6.596,
|
| 1047 |
+
"step": 1460
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 0.10644203615650118,
|
| 1051 |
+
"grad_norm": 1.3125,
|
| 1052 |
+
"learning_rate": 4.923860091185542e-05,
|
| 1053 |
+
"loss": 6.606,
|
| 1054 |
+
"step": 1470
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 0.107166131640559,
|
| 1058 |
+
"grad_norm": 1.2578125,
|
| 1059 |
+
"learning_rate": 4.922417387268251e-05,
|
| 1060 |
+
"loss": 6.6,
|
| 1061 |
+
"step": 1480
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 0.10789022712461684,
|
| 1065 |
+
"grad_norm": 1.328125,
|
| 1066 |
+
"learning_rate": 4.9209613584777445e-05,
|
| 1067 |
+
"loss": 6.5953,
|
| 1068 |
+
"step": 1490
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 0.10861432260867467,
|
| 1072 |
+
"grad_norm": 1.2890625,
|
| 1073 |
+
"learning_rate": 4.919492012823128e-05,
|
| 1074 |
+
"loss": 6.5907,
|
| 1075 |
+
"step": 1500
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 0.1093384180927325,
|
| 1079 |
+
"grad_norm": 1.28125,
|
| 1080 |
+
"learning_rate": 4.918009358386761e-05,
|
| 1081 |
+
"loss": 6.6144,
|
| 1082 |
+
"step": 1510
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.11006251357679032,
|
| 1086 |
+
"grad_norm": 1.3359375,
|
| 1087 |
+
"learning_rate": 4.916513403324207e-05,
|
| 1088 |
+
"loss": 6.5911,
|
| 1089 |
+
"step": 1520
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.11078660906084815,
|
| 1093 |
+
"grad_norm": 1.2578125,
|
| 1094 |
+
"learning_rate": 4.9150041558641947e-05,
|
| 1095 |
+
"loss": 6.6071,
|
| 1096 |
+
"step": 1530
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.11151070454490598,
|
| 1100 |
+
"grad_norm": 1.2421875,
|
| 1101 |
+
"learning_rate": 4.913481624308566e-05,
|
| 1102 |
+
"loss": 6.6028,
|
| 1103 |
+
"step": 1540
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 0.11223480002896381,
|
| 1107 |
+
"grad_norm": 1.3203125,
|
| 1108 |
+
"learning_rate": 4.911945817032238e-05,
|
| 1109 |
+
"loss": 6.6121,
|
| 1110 |
+
"step": 1550
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.11295889551302166,
|
| 1114 |
+
"grad_norm": 1.4140625,
|
| 1115 |
+
"learning_rate": 4.91039674248315e-05,
|
| 1116 |
+
"loss": 6.6003,
|
| 1117 |
+
"step": 1560
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 0.11368299099707949,
|
| 1121 |
+
"grad_norm": 1.421875,
|
| 1122 |
+
"learning_rate": 4.908834409182221e-05,
|
| 1123 |
+
"loss": 6.632,
|
| 1124 |
+
"step": 1570
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.11440708648113732,
|
| 1128 |
+
"grad_norm": 1.2578125,
|
| 1129 |
+
"learning_rate": 4.907258825723301e-05,
|
| 1130 |
+
"loss": 6.5953,
|
| 1131 |
+
"step": 1580
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 0.11513118196519515,
|
| 1135 |
+
"grad_norm": 1.34375,
|
| 1136 |
+
"learning_rate": 4.905670000773126e-05,
|
| 1137 |
+
"loss": 6.6028,
|
| 1138 |
+
"step": 1590
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 0.11585527744925297,
|
| 1142 |
+
"grad_norm": 1.3046875,
|
| 1143 |
+
"learning_rate": 4.904067943071268e-05,
|
| 1144 |
+
"loss": 6.592,
|
| 1145 |
+
"step": 1600
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 0.1165793729333108,
|
| 1149 |
+
"grad_norm": 1.359375,
|
| 1150 |
+
"learning_rate": 4.902452661430087e-05,
|
| 1151 |
+
"loss": 6.6015,
|
| 1152 |
+
"step": 1610
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 0.11730346841736863,
|
| 1156 |
+
"grad_norm": 1.28125,
|
| 1157 |
+
"learning_rate": 4.9008241647346836e-05,
|
| 1158 |
+
"loss": 6.5767,
|
| 1159 |
+
"step": 1620
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 0.11802756390142646,
|
| 1163 |
+
"grad_norm": 1.2734375,
|
| 1164 |
+
"learning_rate": 4.899182461942851e-05,
|
| 1165 |
+
"loss": 6.5894,
|
| 1166 |
+
"step": 1630
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 0.11875165938548429,
|
| 1170 |
+
"grad_norm": 1.28125,
|
| 1171 |
+
"learning_rate": 4.897527562085024e-05,
|
| 1172 |
+
"loss": 6.5856,
|
| 1173 |
+
"step": 1640
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 0.11947575486954214,
|
| 1177 |
+
"grad_norm": 1.3046875,
|
| 1178 |
+
"learning_rate": 4.895859474264229e-05,
|
| 1179 |
+
"loss": 6.5761,
|
| 1180 |
+
"step": 1650
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 0.12019985035359997,
|
| 1184 |
+
"grad_norm": 1.4140625,
|
| 1185 |
+
"learning_rate": 4.894178207656035e-05,
|
| 1186 |
+
"loss": 6.5964,
|
| 1187 |
+
"step": 1660
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 0.1209239458376578,
|
| 1191 |
+
"grad_norm": 1.2109375,
|
| 1192 |
+
"learning_rate": 4.892483771508504e-05,
|
| 1193 |
+
"loss": 6.5657,
|
| 1194 |
+
"step": 1670
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.12164804132171562,
|
| 1198 |
+
"grad_norm": 1.265625,
|
| 1199 |
+
"learning_rate": 4.890776175142138e-05,
|
| 1200 |
+
"loss": 6.5946,
|
| 1201 |
+
"step": 1680
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 0.12237213680577345,
|
| 1205 |
+
"grad_norm": 1.359375,
|
| 1206 |
+
"learning_rate": 4.889055427949829e-05,
|
| 1207 |
+
"loss": 6.5916,
|
| 1208 |
+
"step": 1690
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 0.12309623228983128,
|
| 1212 |
+
"grad_norm": 1.2890625,
|
| 1213 |
+
"learning_rate": 4.8873215393968074e-05,
|
| 1214 |
+
"loss": 6.594,
|
| 1215 |
+
"step": 1700
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 0.12382032777388911,
|
| 1219 |
+
"grad_norm": 1.359375,
|
| 1220 |
+
"learning_rate": 4.885574519020589e-05,
|
| 1221 |
+
"loss": 6.566,
|
| 1222 |
+
"step": 1710
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 0.12454442325794694,
|
| 1226 |
+
"grad_norm": 1.3125,
|
| 1227 |
+
"learning_rate": 4.883814376430924e-05,
|
| 1228 |
+
"loss": 6.5752,
|
| 1229 |
+
"step": 1720
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 0.1252685187420048,
|
| 1233 |
+
"grad_norm": 1.3515625,
|
| 1234 |
+
"learning_rate": 4.882041121309743e-05,
|
| 1235 |
+
"loss": 6.6015,
|
| 1236 |
+
"step": 1730
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 0.12599261422606262,
|
| 1240 |
+
"grad_norm": 1.328125,
|
| 1241 |
+
"learning_rate": 4.880254763411103e-05,
|
| 1242 |
+
"loss": 6.5655,
|
| 1243 |
+
"step": 1740
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 0.12671670971012045,
|
| 1247 |
+
"grad_norm": 1.3125,
|
| 1248 |
+
"learning_rate": 4.878455312561138e-05,
|
| 1249 |
+
"loss": 6.5883,
|
| 1250 |
+
"step": 1750
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.12744080519417827,
|
| 1254 |
+
"grad_norm": 1.3515625,
|
| 1255 |
+
"learning_rate": 4.8766427786579964e-05,
|
| 1256 |
+
"loss": 6.6061,
|
| 1257 |
+
"step": 1760
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 0.1281649006782361,
|
| 1261 |
+
"grad_norm": 1.296875,
|
| 1262 |
+
"learning_rate": 4.874817171671798e-05,
|
| 1263 |
+
"loss": 6.5878,
|
| 1264 |
+
"step": 1770
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 0.12888899616229393,
|
| 1268 |
+
"grad_norm": 1.2265625,
|
| 1269 |
+
"learning_rate": 4.872978501644569e-05,
|
| 1270 |
+
"loss": 6.5837,
|
| 1271 |
+
"step": 1780
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 0.12961309164635176,
|
| 1275 |
+
"grad_norm": 1.3203125,
|
| 1276 |
+
"learning_rate": 4.8711267786901914e-05,
|
| 1277 |
+
"loss": 6.5884,
|
| 1278 |
+
"step": 1790
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 0.1303371871304096,
|
| 1282 |
+
"grad_norm": 1.1953125,
|
| 1283 |
+
"learning_rate": 4.869262012994348e-05,
|
| 1284 |
+
"loss": 6.5864,
|
| 1285 |
+
"step": 1800
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 0.13106128261446742,
|
| 1289 |
+
"grad_norm": 1.3125,
|
| 1290 |
+
"learning_rate": 4.867384214814465e-05,
|
| 1291 |
+
"loss": 6.5655,
|
| 1292 |
+
"step": 1810
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.13178537809852525,
|
| 1296 |
+
"grad_norm": 1.1953125,
|
| 1297 |
+
"learning_rate": 4.8654933944796546e-05,
|
| 1298 |
+
"loss": 6.5747,
|
| 1299 |
+
"step": 1820
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.13250947358258308,
|
| 1303 |
+
"grad_norm": 1.2578125,
|
| 1304 |
+
"learning_rate": 4.86358956239066e-05,
|
| 1305 |
+
"loss": 6.5689,
|
| 1306 |
+
"step": 1830
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.1332335690666409,
|
| 1310 |
+
"grad_norm": 1.234375,
|
| 1311 |
+
"learning_rate": 4.861672729019797e-05,
|
| 1312 |
+
"loss": 6.6059,
|
| 1313 |
+
"step": 1840
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 0.13395766455069874,
|
| 1317 |
+
"grad_norm": 1.2890625,
|
| 1318 |
+
"learning_rate": 4.859742904910899e-05,
|
| 1319 |
+
"loss": 6.5985,
|
| 1320 |
+
"step": 1850
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 0.13468176003475657,
|
| 1324 |
+
"grad_norm": 1.25,
|
| 1325 |
+
"learning_rate": 4.8578001006792544e-05,
|
| 1326 |
+
"loss": 6.6006,
|
| 1327 |
+
"step": 1860
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 0.13540585551881443,
|
| 1331 |
+
"grad_norm": 1.28125,
|
| 1332 |
+
"learning_rate": 4.8558443270115524e-05,
|
| 1333 |
+
"loss": 6.5779,
|
| 1334 |
+
"step": 1870
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 0.13612995100287226,
|
| 1338 |
+
"grad_norm": 1.296875,
|
| 1339 |
+
"learning_rate": 4.853875594665821e-05,
|
| 1340 |
+
"loss": 6.566,
|
| 1341 |
+
"step": 1880
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 0.1368540464869301,
|
| 1345 |
+
"grad_norm": 1.328125,
|
| 1346 |
+
"learning_rate": 4.85189391447137e-05,
|
| 1347 |
+
"loss": 6.6019,
|
| 1348 |
+
"step": 1890
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 0.13757814197098792,
|
| 1352 |
+
"grad_norm": 1.265625,
|
| 1353 |
+
"learning_rate": 4.8498992973287324e-05,
|
| 1354 |
+
"loss": 6.5758,
|
| 1355 |
+
"step": 1900
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 0.13830223745504575,
|
| 1359 |
+
"grad_norm": 1.296875,
|
| 1360 |
+
"learning_rate": 4.8478917542096005e-05,
|
| 1361 |
+
"loss": 6.5754,
|
| 1362 |
+
"step": 1910
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 0.13902633293910358,
|
| 1366 |
+
"grad_norm": 1.21875,
|
| 1367 |
+
"learning_rate": 4.845871296156769e-05,
|
| 1368 |
+
"loss": 6.593,
|
| 1369 |
+
"step": 1920
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 0.1397504284231614,
|
| 1373 |
+
"grad_norm": 1.296875,
|
| 1374 |
+
"learning_rate": 4.843837934284072e-05,
|
| 1375 |
+
"loss": 6.5754,
|
| 1376 |
+
"step": 1930
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 0.14047452390721923,
|
| 1380 |
+
"grad_norm": 1.421875,
|
| 1381 |
+
"learning_rate": 4.841791679776326e-05,
|
| 1382 |
+
"loss": 6.5659,
|
| 1383 |
+
"step": 1940
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 0.14119861939127706,
|
| 1387 |
+
"grad_norm": 1.359375,
|
| 1388 |
+
"learning_rate": 4.839732543889263e-05,
|
| 1389 |
+
"loss": 6.5821,
|
| 1390 |
+
"step": 1950
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 0.1419227148753349,
|
| 1394 |
+
"grad_norm": 1.34375,
|
| 1395 |
+
"learning_rate": 4.83766053794947e-05,
|
| 1396 |
+
"loss": 6.58,
|
| 1397 |
+
"step": 1960
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 0.14264681035939272,
|
| 1401 |
+
"grad_norm": 1.2734375,
|
| 1402 |
+
"learning_rate": 4.83557567335433e-05,
|
| 1403 |
+
"loss": 6.5662,
|
| 1404 |
+
"step": 1970
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.14337090584345055,
|
| 1408 |
+
"grad_norm": 1.2265625,
|
| 1409 |
+
"learning_rate": 4.833477961571956e-05,
|
| 1410 |
+
"loss": 6.5675,
|
| 1411 |
+
"step": 1980
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.14409500132750838,
|
| 1415 |
+
"grad_norm": 1.1875,
|
| 1416 |
+
"learning_rate": 4.831367414141129e-05,
|
| 1417 |
+
"loss": 6.5784,
|
| 1418 |
+
"step": 1990
|
| 1419 |
+
}
|
| 1420 |
+
],
|
| 1421 |
+
"logging_steps": 10,
|
| 1422 |
+
"max_steps": 13810,
|
| 1423 |
+
"num_input_tokens_seen": 0,
|
| 1424 |
+
"num_train_epochs": 1,
|
| 1425 |
+
"save_steps": 999,
|
| 1426 |
+
"stateful_callbacks": {
|
| 1427 |
+
"TrainerControl": {
|
| 1428 |
+
"args": {
|
| 1429 |
+
"should_epoch_stop": false,
|
| 1430 |
+
"should_evaluate": false,
|
| 1431 |
+
"should_log": false,
|
| 1432 |
+
"should_save": true,
|
| 1433 |
+
"should_training_stop": false
|
| 1434 |
+
},
|
| 1435 |
+
"attributes": {}
|
| 1436 |
+
}
|
| 1437 |
+
},
|
| 1438 |
+
"total_flos": 4.300582065138565e+18,
|
| 1439 |
+
"train_batch_size": 30,
|
| 1440 |
+
"trial_name": null,
|
| 1441 |
+
"trial_params": null
|
| 1442 |
+
}
|