Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- 400M/chat_template.jinja +1 -0
- 400M/config.json +170 -0
- 400M/generation_config.json +17 -0
- 400M/latest +1 -0
- 400M/model.safetensors +3 -0
- 400M/processor_config.json +62 -0
- 400M/rng_state_0.pth +3 -0
- 400M/rng_state_1.pth +3 -0
- 400M/rng_state_10.pth +3 -0
- 400M/rng_state_11.pth +3 -0
- 400M/rng_state_2.pth +3 -0
- 400M/rng_state_3.pth +3 -0
- 400M/rng_state_4.pth +3 -0
- 400M/rng_state_5.pth +3 -0
- 400M/rng_state_6.pth +3 -0
- 400M/rng_state_7.pth +3 -0
- 400M/rng_state_8.pth +3 -0
- 400M/rng_state_9.pth +3 -0
- 400M/scheduler.pt +3 -0
- 400M/tokenizer.json +3 -0
- 400M/tokenizer_config.json +56 -0
- 400M/trainer_state.json +2883 -0
- 400M/training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -34,3 +34,4 @@ saved_model/**/* 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 |
200M/tokenizer.json 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 |
200M/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
400M/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
400M/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
400M/config.json
ADDED
|
@@ -0,0 +1,170 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma4UnifiedForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"audio_config": {
|
| 6 |
+
"_name_or_path": "",
|
| 7 |
+
"architectures": null,
|
| 8 |
+
"audio_embed_dim": 640,
|
| 9 |
+
"chunk_size_feed_forward": 0,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"is_encoder_decoder": false,
|
| 17 |
+
"label2id": {
|
| 18 |
+
"LABEL_0": 0,
|
| 19 |
+
"LABEL_1": 1
|
| 20 |
+
},
|
| 21 |
+
"model_type": "gemma4_unified_audio",
|
| 22 |
+
"output_attentions": false,
|
| 23 |
+
"output_hidden_states": false,
|
| 24 |
+
"problem_type": null,
|
| 25 |
+
"return_dict": true,
|
| 26 |
+
"rms_norm_eps": 1e-06
|
| 27 |
+
},
|
| 28 |
+
"audio_token_id": 258881,
|
| 29 |
+
"boa_token_id": 256000,
|
| 30 |
+
"boi_token_id": 255999,
|
| 31 |
+
"bos_token_id": 2,
|
| 32 |
+
"dtype": "bfloat16",
|
| 33 |
+
"eoa_token_index": 258883,
|
| 34 |
+
"eoi_token_id": 258882,
|
| 35 |
+
"eos_token_id": 1,
|
| 36 |
+
"hidden_size": 3840,
|
| 37 |
+
"image_token_id": 258880,
|
| 38 |
+
"initializer_range": 0.02,
|
| 39 |
+
"model_type": "gemma4_unified",
|
| 40 |
+
"pad_token_id": 0,
|
| 41 |
+
"text_config": {
|
| 42 |
+
"attention_bias": false,
|
| 43 |
+
"attention_dropout": 0.0,
|
| 44 |
+
"attention_k_eq_v": true,
|
| 45 |
+
"bos_token_id": 2,
|
| 46 |
+
"dtype": "bfloat16",
|
| 47 |
+
"enable_moe_block": false,
|
| 48 |
+
"eos_token_id": 1,
|
| 49 |
+
"final_logit_softcapping": 30.0,
|
| 50 |
+
"global_head_dim": 512,
|
| 51 |
+
"head_dim": 256,
|
| 52 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 53 |
+
"hidden_size": 3840,
|
| 54 |
+
"hidden_size_per_layer_input": 0,
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 15360,
|
| 57 |
+
"layer_types": [
|
| 58 |
+
"sliding_attention",
|
| 59 |
+
"sliding_attention",
|
| 60 |
+
"sliding_attention",
|
| 61 |
+
"sliding_attention",
|
| 62 |
+
"sliding_attention",
|
| 63 |
+
"full_attention",
|
| 64 |
+
"sliding_attention",
|
| 65 |
+
"sliding_attention",
|
| 66 |
+
"sliding_attention",
|
| 67 |
+
"sliding_attention",
|
| 68 |
+
"sliding_attention",
|
| 69 |
+
"full_attention",
|
| 70 |
+
"sliding_attention",
|
| 71 |
+
"sliding_attention",
|
| 72 |
+
"sliding_attention",
|
| 73 |
+
"sliding_attention",
|
| 74 |
+
"sliding_attention",
|
| 75 |
+
"full_attention",
|
| 76 |
+
"sliding_attention",
|
| 77 |
+
"sliding_attention",
|
| 78 |
+
"sliding_attention",
|
| 79 |
+
"sliding_attention",
|
| 80 |
+
"sliding_attention",
|
| 81 |
+
"full_attention",
|
| 82 |
+
"sliding_attention",
|
| 83 |
+
"sliding_attention",
|
| 84 |
+
"sliding_attention",
|
| 85 |
+
"sliding_attention",
|
| 86 |
+
"sliding_attention",
|
| 87 |
+
"full_attention",
|
| 88 |
+
"sliding_attention",
|
| 89 |
+
"sliding_attention",
|
| 90 |
+
"sliding_attention",
|
| 91 |
+
"sliding_attention",
|
| 92 |
+
"sliding_attention",
|
| 93 |
+
"full_attention",
|
| 94 |
+
"sliding_attention",
|
| 95 |
+
"sliding_attention",
|
| 96 |
+
"sliding_attention",
|
| 97 |
+
"sliding_attention",
|
| 98 |
+
"sliding_attention",
|
| 99 |
+
"full_attention",
|
| 100 |
+
"sliding_attention",
|
| 101 |
+
"sliding_attention",
|
| 102 |
+
"sliding_attention",
|
| 103 |
+
"sliding_attention",
|
| 104 |
+
"sliding_attention",
|
| 105 |
+
"full_attention"
|
| 106 |
+
],
|
| 107 |
+
"max_position_embeddings": 262144,
|
| 108 |
+
"model_type": "gemma4_unified_text",
|
| 109 |
+
"moe_intermediate_size": null,
|
| 110 |
+
"num_attention_heads": 16,
|
| 111 |
+
"num_experts": null,
|
| 112 |
+
"num_global_key_value_heads": 1,
|
| 113 |
+
"num_hidden_layers": 48,
|
| 114 |
+
"num_key_value_heads": 8,
|
| 115 |
+
"num_kv_shared_layers": 0,
|
| 116 |
+
"pad_token_id": 0,
|
| 117 |
+
"rms_norm_eps": 1e-06,
|
| 118 |
+
"rope_parameters": {
|
| 119 |
+
"full_attention": {
|
| 120 |
+
"partial_rotary_factor": 0.25,
|
| 121 |
+
"rope_theta": 1000000.0,
|
| 122 |
+
"rope_type": "proportional"
|
| 123 |
+
},
|
| 124 |
+
"sliding_attention": {
|
| 125 |
+
"rope_theta": 10000.0,
|
| 126 |
+
"rope_type": "default"
|
| 127 |
+
}
|
| 128 |
+
},
|
| 129 |
+
"sliding_window": 1024,
|
| 130 |
+
"tie_word_embeddings": true,
|
| 131 |
+
"top_k_experts": null,
|
| 132 |
+
"use_bidirectional_attention": "vision",
|
| 133 |
+
"use_cache": false,
|
| 134 |
+
"use_double_wide_mlp": false,
|
| 135 |
+
"vocab_size": 262144,
|
| 136 |
+
"vocab_size_per_layer_input": 262144
|
| 137 |
+
},
|
| 138 |
+
"tie_word_embeddings": true,
|
| 139 |
+
"transformers_version": "5.10.0",
|
| 140 |
+
"use_cache": false,
|
| 141 |
+
"video_token_id": 258884,
|
| 142 |
+
"vision_config": {
|
| 143 |
+
"_name_or_path": "",
|
| 144 |
+
"architectures": null,
|
| 145 |
+
"chunk_size_feed_forward": 0,
|
| 146 |
+
"dtype": "bfloat16",
|
| 147 |
+
"id2label": {
|
| 148 |
+
"0": "LABEL_0",
|
| 149 |
+
"1": "LABEL_1"
|
| 150 |
+
},
|
| 151 |
+
"initializer_range": 0.02,
|
| 152 |
+
"is_encoder_decoder": false,
|
| 153 |
+
"label2id": {
|
| 154 |
+
"LABEL_0": 0,
|
| 155 |
+
"LABEL_1": 1
|
| 156 |
+
},
|
| 157 |
+
"mm_embed_dim": 3840,
|
| 158 |
+
"mm_posemb_size": 1120,
|
| 159 |
+
"model_type": "gemma4_unified_vision",
|
| 160 |
+
"num_soft_tokens": 280,
|
| 161 |
+
"output_attentions": false,
|
| 162 |
+
"output_hidden_states": false,
|
| 163 |
+
"output_proj_dims": 3840,
|
| 164 |
+
"patch_size": 16,
|
| 165 |
+
"pooling_kernel_size": 3,
|
| 166 |
+
"problem_type": null,
|
| 167 |
+
"return_dict": true,
|
| 168 |
+
"rms_norm_eps": 1e-06
|
| 169 |
+
}
|
| 170 |
+
}
|
400M/generation_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 2,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
1,
|
| 6 |
+
1
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 0,
|
| 9 |
+
"suppress_tokens": [
|
| 10 |
+
258883,
|
| 11 |
+
258882
|
| 12 |
+
],
|
| 13 |
+
"temperature": 1.0,
|
| 14 |
+
"top_k": 64,
|
| 15 |
+
"top_p": 0.95,
|
| 16 |
+
"transformers_version": "5.10.0"
|
| 17 |
+
}
|
400M/latest
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
global_step4070
|
400M/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa363b88ca525a7cee5bd9819479cb183619fd70a8d44408f798ad78abf8d9a0
|
| 3 |
+
size 23919549408
|
400M/processor_config.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"audio_ms_per_token": 40,
|
| 3 |
+
"audio_seq_length": 750,
|
| 4 |
+
"feature_extractor": {
|
| 5 |
+
"audio_samples_per_token": 640,
|
| 6 |
+
"feature_extractor_type": "Gemma4UnifiedAudioFeatureExtractor",
|
| 7 |
+
"feature_size": 640,
|
| 8 |
+
"padding_side": "right",
|
| 9 |
+
"padding_value": 0.0,
|
| 10 |
+
"return_attention_mask": true,
|
| 11 |
+
"sampling_rate": 16000
|
| 12 |
+
},
|
| 13 |
+
"image_processor": {
|
| 14 |
+
"do_convert_rgb": true,
|
| 15 |
+
"do_normalize": false,
|
| 16 |
+
"do_rescale": true,
|
| 17 |
+
"do_resize": true,
|
| 18 |
+
"image_mean": [
|
| 19 |
+
0.0,
|
| 20 |
+
0.0,
|
| 21 |
+
0.0
|
| 22 |
+
],
|
| 23 |
+
"image_processor_type": "Gemma4UnifiedImageProcessor",
|
| 24 |
+
"image_std": [
|
| 25 |
+
1.0,
|
| 26 |
+
1.0,
|
| 27 |
+
1.0
|
| 28 |
+
],
|
| 29 |
+
"max_soft_tokens": 280,
|
| 30 |
+
"patch_size": 16,
|
| 31 |
+
"pooling_kernel_size": 3,
|
| 32 |
+
"resample": 3,
|
| 33 |
+
"rescale_factor": 0.00392156862745098
|
| 34 |
+
},
|
| 35 |
+
"image_seq_length": 280,
|
| 36 |
+
"processor_class": "Gemma4UnifiedProcessor",
|
| 37 |
+
"video_processor": {
|
| 38 |
+
"do_convert_rgb": true,
|
| 39 |
+
"do_normalize": true,
|
| 40 |
+
"do_rescale": true,
|
| 41 |
+
"do_resize": true,
|
| 42 |
+
"do_sample_frames": true,
|
| 43 |
+
"image_mean": [
|
| 44 |
+
0.0,
|
| 45 |
+
0.0,
|
| 46 |
+
0.0
|
| 47 |
+
],
|
| 48 |
+
"image_std": [
|
| 49 |
+
1.0,
|
| 50 |
+
1.0,
|
| 51 |
+
1.0
|
| 52 |
+
],
|
| 53 |
+
"max_soft_tokens": 70,
|
| 54 |
+
"num_frames": 32,
|
| 55 |
+
"patch_size": 16,
|
| 56 |
+
"pooling_kernel_size": 3,
|
| 57 |
+
"resample": 3,
|
| 58 |
+
"rescale_factor": 0.00392156862745098,
|
| 59 |
+
"return_metadata": false,
|
| 60 |
+
"video_processor_type": "Gemma4UnifiedVideoProcessor"
|
| 61 |
+
}
|
| 62 |
+
}
|
400M/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577
|
| 3 |
+
size 14469
|
400M/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4
|
| 3 |
+
size 14469
|
400M/rng_state_10.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3
|
| 3 |
+
size 14476
|
400M/rng_state_11.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88
|
| 3 |
+
size 14476
|
400M/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf
|
| 3 |
+
size 14469
|
400M/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a
|
| 3 |
+
size 14469
|
400M/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f
|
| 3 |
+
size 14469
|
400M/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553
|
| 3 |
+
size 14469
|
400M/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144
|
| 3 |
+
size 14469
|
400M/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7
|
| 3 |
+
size 14469
|
400M/rng_state_8.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11
|
| 3 |
+
size 14469
|
400M/rng_state_9.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f
|
| 3 |
+
size 14469
|
400M/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c373bca4c9a6ef963a59ee82f3b16a9a90b89b812375170f487261fe4db9e20a
|
| 3 |
+
size 1465
|
400M/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6
|
| 3 |
+
size 32170070
|
400M/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"audio_token": "<|audio|>",
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"boa_token": "<|audio>",
|
| 5 |
+
"boi_token": "<|image>",
|
| 6 |
+
"bos_token": "<bos>",
|
| 7 |
+
"eoa_token": "<audio|>",
|
| 8 |
+
"eoc_token": "<channel|>",
|
| 9 |
+
"eoi_token": "<image|>",
|
| 10 |
+
"eos_token": "<eos>",
|
| 11 |
+
"eot_token": "<turn|>",
|
| 12 |
+
"escape_token": "<|\"|>",
|
| 13 |
+
"etc_token": "<tool_call|>",
|
| 14 |
+
"etd_token": "<tool|>",
|
| 15 |
+
"etr_token": "<tool_response|>",
|
| 16 |
+
"extra_special_tokens": [
|
| 17 |
+
"<|video|>"
|
| 18 |
+
],
|
| 19 |
+
"image_token": "<|image|>",
|
| 20 |
+
"is_local": true,
|
| 21 |
+
"local_files_only": true,
|
| 22 |
+
"mask_token": "<mask>",
|
| 23 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 24 |
+
"model_specific_special_tokens": {
|
| 25 |
+
"audio_token": "<|audio|>",
|
| 26 |
+
"boa_token": "<|audio>",
|
| 27 |
+
"boi_token": "<|image>",
|
| 28 |
+
"eoa_token": "<audio|>",
|
| 29 |
+
"eoc_token": "<channel|>",
|
| 30 |
+
"eoi_token": "<image|>",
|
| 31 |
+
"eot_token": "<turn|>",
|
| 32 |
+
"escape_token": "<|\"|>",
|
| 33 |
+
"etc_token": "<tool_call|>",
|
| 34 |
+
"etd_token": "<tool|>",
|
| 35 |
+
"etr_token": "<tool_response|>",
|
| 36 |
+
"image_token": "<|image|>",
|
| 37 |
+
"soc_token": "<|channel>",
|
| 38 |
+
"sot_token": "<|turn>",
|
| 39 |
+
"stc_token": "<|tool_call>",
|
| 40 |
+
"std_token": "<|tool>",
|
| 41 |
+
"str_token": "<|tool_response>",
|
| 42 |
+
"think_token": "<|think|>"
|
| 43 |
+
},
|
| 44 |
+
"pad_token": "<pad>",
|
| 45 |
+
"padding_side": "right",
|
| 46 |
+
"processor_class": "Gemma4UnifiedProcessor",
|
| 47 |
+
"soc_token": "<|channel>",
|
| 48 |
+
"sot_token": "<|turn>",
|
| 49 |
+
"split_special_tokens": false,
|
| 50 |
+
"stc_token": "<|tool_call>",
|
| 51 |
+
"std_token": "<|tool>",
|
| 52 |
+
"str_token": "<|tool_response>",
|
| 53 |
+
"think_token": "<|think|>",
|
| 54 |
+
"tokenizer_class": "GemmaTokenizer",
|
| 55 |
+
"unk_token": "<unk>"
|
| 56 |
+
}
|
400M/trainer_state.json
ADDED
|
@@ -0,0 +1,2883 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.5205935021744692,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 4070,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0012790995139421847,
|
| 14 |
+
"grad_norm": 623.6208359240255,
|
| 15 |
+
"learning_rate": 2.910112359550562e-06,
|
| 16 |
+
"loss": 1.5341346740722657,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0025581990278843694,
|
| 21 |
+
"grad_norm": 463.91385336199573,
|
| 22 |
+
"learning_rate": 3.921348314606742e-06,
|
| 23 |
+
"loss": 0.9242143630981445,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.003837298541826554,
|
| 28 |
+
"grad_norm": 53.84492921608342,
|
| 29 |
+
"learning_rate": 4.932584269662922e-06,
|
| 30 |
+
"loss": 0.7964631080627441,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.005116398055768739,
|
| 35 |
+
"grad_norm": 61.330614407626,
|
| 36 |
+
"learning_rate": 5.943820224719102e-06,
|
| 37 |
+
"loss": 0.7263487815856934,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.006395497569710923,
|
| 42 |
+
"grad_norm": 44.648257436551425,
|
| 43 |
+
"learning_rate": 6.955056179775282e-06,
|
| 44 |
+
"loss": 0.6948539733886718,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.007674597083653108,
|
| 49 |
+
"grad_norm": 56.621196014184925,
|
| 50 |
+
"learning_rate": 7.966292134831462e-06,
|
| 51 |
+
"loss": 0.6822492122650147,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.008953696597595293,
|
| 56 |
+
"grad_norm": 37.47997089126972,
|
| 57 |
+
"learning_rate": 8.977528089887641e-06,
|
| 58 |
+
"loss": 0.6963601112365723,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.010232796111537478,
|
| 63 |
+
"grad_norm": 105.99404311228525,
|
| 64 |
+
"learning_rate": 9.988764044943822e-06,
|
| 65 |
+
"loss": 0.6244803428649902,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.011511895625479662,
|
| 70 |
+
"grad_norm": 24.305519470285297,
|
| 71 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 72 |
+
"loss": 0.6451755046844483,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.012790995139421847,
|
| 77 |
+
"grad_norm": 103.85716120098365,
|
| 78 |
+
"learning_rate": 1.2011235955056182e-05,
|
| 79 |
+
"loss": 0.7046442031860352,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.014070094653364031,
|
| 84 |
+
"grad_norm": 8.153387119451759,
|
| 85 |
+
"learning_rate": 1.3022471910112362e-05,
|
| 86 |
+
"loss": 0.7193686962127686,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.015349194167306216,
|
| 91 |
+
"grad_norm": 11.299752541462171,
|
| 92 |
+
"learning_rate": 1.403370786516854e-05,
|
| 93 |
+
"loss": 0.6899770259857178,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.0166282936812484,
|
| 98 |
+
"grad_norm": 51.069015129101714,
|
| 99 |
+
"learning_rate": 1.504494382022472e-05,
|
| 100 |
+
"loss": 0.6754970550537109,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.017907393195190587,
|
| 105 |
+
"grad_norm": 54.175973870128736,
|
| 106 |
+
"learning_rate": 1.60561797752809e-05,
|
| 107 |
+
"loss": 0.6890584468841553,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.01918649270913277,
|
| 112 |
+
"grad_norm": 20.136113415923976,
|
| 113 |
+
"learning_rate": 1.706741573033708e-05,
|
| 114 |
+
"loss": 0.7157495498657227,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.020465592223074956,
|
| 119 |
+
"grad_norm": 37.80605267456,
|
| 120 |
+
"learning_rate": 1.8078651685393256e-05,
|
| 121 |
+
"loss": 0.7468127250671387,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.02174469173701714,
|
| 126 |
+
"grad_norm": 10.709328210046946,
|
| 127 |
+
"learning_rate": 1.908988764044944e-05,
|
| 128 |
+
"loss": 0.8013706207275391,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.023023791250959325,
|
| 133 |
+
"grad_norm": 13.849160742797585,
|
| 134 |
+
"learning_rate": 2e-05,
|
| 135 |
+
"loss": 0.7073573112487793,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.02430289076490151,
|
| 140 |
+
"grad_norm": 252.71253502007187,
|
| 141 |
+
"learning_rate": 2e-05,
|
| 142 |
+
"loss": 0.7326308727264405,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.025581990278843694,
|
| 147 |
+
"grad_norm": 4.329744526145032,
|
| 148 |
+
"learning_rate": 2e-05,
|
| 149 |
+
"loss": 0.7149838924407959,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.02686108979278588,
|
| 154 |
+
"grad_norm": 6.604126459594562,
|
| 155 |
+
"learning_rate": 2e-05,
|
| 156 |
+
"loss": 0.6777316093444824,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.028140189306728063,
|
| 161 |
+
"grad_norm": 9.52386806224655,
|
| 162 |
+
"learning_rate": 2e-05,
|
| 163 |
+
"loss": 0.6922791957855224,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.02941928882067025,
|
| 168 |
+
"grad_norm": 17.21807673598888,
|
| 169 |
+
"learning_rate": 2e-05,
|
| 170 |
+
"loss": 0.7441821575164795,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.03069838833461243,
|
| 175 |
+
"grad_norm": 9.18143504861921,
|
| 176 |
+
"learning_rate": 2e-05,
|
| 177 |
+
"loss": 0.7066239833831787,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.031977487848554614,
|
| 182 |
+
"grad_norm": 4.217817099649031,
|
| 183 |
+
"learning_rate": 2e-05,
|
| 184 |
+
"loss": 0.6791770458221436,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.0332565873624968,
|
| 189 |
+
"grad_norm": 17.965064875677463,
|
| 190 |
+
"learning_rate": 2e-05,
|
| 191 |
+
"loss": 0.6521565914154053,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.03453568687643899,
|
| 196 |
+
"grad_norm": 96.25438261050074,
|
| 197 |
+
"learning_rate": 2e-05,
|
| 198 |
+
"loss": 0.6563163757324219,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.03581478639038117,
|
| 203 |
+
"grad_norm": 14.572604906608621,
|
| 204 |
+
"learning_rate": 2e-05,
|
| 205 |
+
"loss": 0.6991176128387451,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.03709388590432336,
|
| 210 |
+
"grad_norm": 10.646801341452742,
|
| 211 |
+
"learning_rate": 2e-05,
|
| 212 |
+
"loss": 0.6827552318572998,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.03837298541826554,
|
| 217 |
+
"grad_norm": 5.100529139893381,
|
| 218 |
+
"learning_rate": 2e-05,
|
| 219 |
+
"loss": 0.6454049587249756,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.039652084932207725,
|
| 224 |
+
"grad_norm": 3.7542492731409425,
|
| 225 |
+
"learning_rate": 2e-05,
|
| 226 |
+
"loss": 0.6457695007324219,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.04093118444614991,
|
| 231 |
+
"grad_norm": 4.927798665797291,
|
| 232 |
+
"learning_rate": 2e-05,
|
| 233 |
+
"loss": 0.6363472938537598,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0422102839600921,
|
| 238 |
+
"grad_norm": 4.6042140802832074,
|
| 239 |
+
"learning_rate": 2e-05,
|
| 240 |
+
"loss": 0.6547730922698974,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.04348938347403428,
|
| 245 |
+
"grad_norm": 3.9790435649800364,
|
| 246 |
+
"learning_rate": 2e-05,
|
| 247 |
+
"loss": 0.6693540573120117,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.04476848298797646,
|
| 252 |
+
"grad_norm": 6.056606530921015,
|
| 253 |
+
"learning_rate": 2e-05,
|
| 254 |
+
"loss": 0.6478125572204589,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.04604758250191865,
|
| 259 |
+
"grad_norm": 5.000014296778777,
|
| 260 |
+
"learning_rate": 2e-05,
|
| 261 |
+
"loss": 0.6575343132019043,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.047326682015860835,
|
| 266 |
+
"grad_norm": 170.7900036194794,
|
| 267 |
+
"learning_rate": 2e-05,
|
| 268 |
+
"loss": 0.6500011444091797,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.04860578152980302,
|
| 273 |
+
"grad_norm": 4.136859247278033,
|
| 274 |
+
"learning_rate": 2e-05,
|
| 275 |
+
"loss": 0.7492189407348633,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.0498848810437452,
|
| 280 |
+
"grad_norm": 4.9635865373650825,
|
| 281 |
+
"learning_rate": 2e-05,
|
| 282 |
+
"loss": 0.7007633209228515,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.05116398055768739,
|
| 287 |
+
"grad_norm": 7.184333434818073,
|
| 288 |
+
"learning_rate": 2e-05,
|
| 289 |
+
"loss": 0.6674163818359375,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.05244308007162957,
|
| 294 |
+
"grad_norm": 10.770350153540436,
|
| 295 |
+
"learning_rate": 2e-05,
|
| 296 |
+
"loss": 0.6679406642913819,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.05372217958557176,
|
| 301 |
+
"grad_norm": 7.497226452066126,
|
| 302 |
+
"learning_rate": 2e-05,
|
| 303 |
+
"loss": 0.7049285888671875,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.05500127909951394,
|
| 308 |
+
"grad_norm": 14.03042351222223,
|
| 309 |
+
"learning_rate": 2e-05,
|
| 310 |
+
"loss": 0.676944637298584,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.056280378613456125,
|
| 315 |
+
"grad_norm": 3.469593844461007,
|
| 316 |
+
"learning_rate": 2e-05,
|
| 317 |
+
"loss": 0.6626708507537842,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.05755947812739831,
|
| 322 |
+
"grad_norm": 3.410706827588382,
|
| 323 |
+
"learning_rate": 2e-05,
|
| 324 |
+
"loss": 0.65382399559021,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.0588385776413405,
|
| 329 |
+
"grad_norm": 1.980498220638673,
|
| 330 |
+
"learning_rate": 2e-05,
|
| 331 |
+
"loss": 0.7313810348510742,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.060117677155282684,
|
| 336 |
+
"grad_norm": 3.739263153638155,
|
| 337 |
+
"learning_rate": 2e-05,
|
| 338 |
+
"loss": 0.6562081336975097,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.06139677666922486,
|
| 343 |
+
"grad_norm": 2.0018607090496596,
|
| 344 |
+
"learning_rate": 2e-05,
|
| 345 |
+
"loss": 0.6454696178436279,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.06267587618316706,
|
| 350 |
+
"grad_norm": 3.4114991222813154,
|
| 351 |
+
"learning_rate": 2e-05,
|
| 352 |
+
"loss": 0.6556525230407715,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.06395497569710923,
|
| 357 |
+
"grad_norm": 3.205444422603881,
|
| 358 |
+
"learning_rate": 2e-05,
|
| 359 |
+
"loss": 0.6443347454071044,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.06523407521105141,
|
| 364 |
+
"grad_norm": 5.095129800632835,
|
| 365 |
+
"learning_rate": 2e-05,
|
| 366 |
+
"loss": 0.6250532150268555,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.0665131747249936,
|
| 371 |
+
"grad_norm": 2.5894252003422684,
|
| 372 |
+
"learning_rate": 2e-05,
|
| 373 |
+
"loss": 0.6742976188659668,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.06779227423893579,
|
| 378 |
+
"grad_norm": 2.8889178959957196,
|
| 379 |
+
"learning_rate": 2e-05,
|
| 380 |
+
"loss": 0.6449506282806396,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.06907137375287797,
|
| 385 |
+
"grad_norm": 3.5369834303098253,
|
| 386 |
+
"learning_rate": 2e-05,
|
| 387 |
+
"loss": 0.6863080024719238,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.07035047326682016,
|
| 392 |
+
"grad_norm": 1.6919645181272793,
|
| 393 |
+
"learning_rate": 2e-05,
|
| 394 |
+
"loss": 0.5828201293945312,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.07162957278076235,
|
| 399 |
+
"grad_norm": 2.7688275845052464,
|
| 400 |
+
"learning_rate": 2e-05,
|
| 401 |
+
"loss": 0.6269129753112793,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.07290867229470453,
|
| 406 |
+
"grad_norm": 9.01577429604241,
|
| 407 |
+
"learning_rate": 2e-05,
|
| 408 |
+
"loss": 0.674321174621582,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.07418777180864672,
|
| 413 |
+
"grad_norm": 1.3605319314128796,
|
| 414 |
+
"learning_rate": 2e-05,
|
| 415 |
+
"loss": 0.6332732200622558,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.07546687132258889,
|
| 420 |
+
"grad_norm": 2.3592211294500034,
|
| 421 |
+
"learning_rate": 2e-05,
|
| 422 |
+
"loss": 0.6280890941619873,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.07674597083653108,
|
| 427 |
+
"grad_norm": 2.1105007442103205,
|
| 428 |
+
"learning_rate": 2e-05,
|
| 429 |
+
"loss": 0.6547659873962403,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.07802507035047326,
|
| 434 |
+
"grad_norm": 1.9882775544536382,
|
| 435 |
+
"learning_rate": 2e-05,
|
| 436 |
+
"loss": 0.6447579860687256,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.07930416986441545,
|
| 441 |
+
"grad_norm": 2.1390953156204024,
|
| 442 |
+
"learning_rate": 2e-05,
|
| 443 |
+
"loss": 0.6662328243255615,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.08058326937835764,
|
| 448 |
+
"grad_norm": 1.822512642342021,
|
| 449 |
+
"learning_rate": 2e-05,
|
| 450 |
+
"loss": 0.6324103832244873,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.08186236889229982,
|
| 455 |
+
"grad_norm": 4.970675015253946,
|
| 456 |
+
"learning_rate": 2e-05,
|
| 457 |
+
"loss": 0.7083280563354493,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.08314146840624201,
|
| 462 |
+
"grad_norm": 1.618436134118923,
|
| 463 |
+
"learning_rate": 2e-05,
|
| 464 |
+
"loss": 0.6183808803558349,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.0844205679201842,
|
| 469 |
+
"grad_norm": 3.040019611591677,
|
| 470 |
+
"learning_rate": 2e-05,
|
| 471 |
+
"loss": 0.6266217231750488,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.08569966743412638,
|
| 476 |
+
"grad_norm": 2.9441294848550443,
|
| 477 |
+
"learning_rate": 2e-05,
|
| 478 |
+
"loss": 0.6195056438446045,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.08697876694806855,
|
| 483 |
+
"grad_norm": 2.2931610909553206,
|
| 484 |
+
"learning_rate": 2e-05,
|
| 485 |
+
"loss": 0.6402240753173828,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.08825786646201074,
|
| 490 |
+
"grad_norm": 1.5306455027972703,
|
| 491 |
+
"learning_rate": 2e-05,
|
| 492 |
+
"loss": 0.6292661666870117,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.08953696597595293,
|
| 497 |
+
"grad_norm": 3.1898947482255084,
|
| 498 |
+
"learning_rate": 2e-05,
|
| 499 |
+
"loss": 0.6502187728881836,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.09081606548989511,
|
| 504 |
+
"grad_norm": 2.4950893350851175,
|
| 505 |
+
"learning_rate": 2e-05,
|
| 506 |
+
"loss": 0.6811255931854248,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.0920951650038373,
|
| 511 |
+
"grad_norm": 11.041031364083185,
|
| 512 |
+
"learning_rate": 2e-05,
|
| 513 |
+
"loss": 0.6565009117126465,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.09337426451777948,
|
| 518 |
+
"grad_norm": 6.2379473566814365,
|
| 519 |
+
"learning_rate": 2e-05,
|
| 520 |
+
"loss": 0.6482836723327636,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.09465336403172167,
|
| 525 |
+
"grad_norm": 2.778467542831759,
|
| 526 |
+
"learning_rate": 2e-05,
|
| 527 |
+
"loss": 0.6700276851654052,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.09593246354566386,
|
| 532 |
+
"grad_norm": 3.156097269104696,
|
| 533 |
+
"learning_rate": 2e-05,
|
| 534 |
+
"loss": 0.6324517726898193,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.09721156305960604,
|
| 539 |
+
"grad_norm": 5.284389107775213,
|
| 540 |
+
"learning_rate": 2e-05,
|
| 541 |
+
"loss": 0.6471449851989746,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.09849066257354822,
|
| 546 |
+
"grad_norm": 2.412749474777184,
|
| 547 |
+
"learning_rate": 2e-05,
|
| 548 |
+
"loss": 0.6330071449279785,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.0997697620874904,
|
| 553 |
+
"grad_norm": 5.1191263952404995,
|
| 554 |
+
"learning_rate": 2e-05,
|
| 555 |
+
"loss": 0.6730413436889648,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.10104886160143259,
|
| 560 |
+
"grad_norm": 2.6642492725506637,
|
| 561 |
+
"learning_rate": 2e-05,
|
| 562 |
+
"loss": 0.6396522521972656,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.10232796111537477,
|
| 567 |
+
"grad_norm": 3.1687128884117226,
|
| 568 |
+
"learning_rate": 2e-05,
|
| 569 |
+
"loss": 0.6374218940734864,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.10360706062931696,
|
| 574 |
+
"grad_norm": 1.3308499280343475,
|
| 575 |
+
"learning_rate": 2e-05,
|
| 576 |
+
"loss": 0.6415046691894531,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.10488616014325915,
|
| 581 |
+
"grad_norm": 3.3291664977384006,
|
| 582 |
+
"learning_rate": 2e-05,
|
| 583 |
+
"loss": 0.6248232841491699,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.10616525965720133,
|
| 588 |
+
"grad_norm": 4.859932361527834,
|
| 589 |
+
"learning_rate": 2e-05,
|
| 590 |
+
"loss": 0.6399627685546875,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.10744435917114352,
|
| 595 |
+
"grad_norm": 1.3955977789644918,
|
| 596 |
+
"learning_rate": 2e-05,
|
| 597 |
+
"loss": 0.6317929744720459,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.1087234586850857,
|
| 602 |
+
"grad_norm": 1.556340444654958,
|
| 603 |
+
"learning_rate": 2e-05,
|
| 604 |
+
"loss": 0.6560098648071289,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.11000255819902788,
|
| 609 |
+
"grad_norm": 2.5492831528808524,
|
| 610 |
+
"learning_rate": 2e-05,
|
| 611 |
+
"loss": 0.6061997890472413,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.11128165771297006,
|
| 616 |
+
"grad_norm": 1.4147821750495504,
|
| 617 |
+
"learning_rate": 2e-05,
|
| 618 |
+
"loss": 0.6266984462738037,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.11256075722691225,
|
| 623 |
+
"grad_norm": 2.8987948576009255,
|
| 624 |
+
"learning_rate": 2e-05,
|
| 625 |
+
"loss": 0.6040752410888672,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.11383985674085444,
|
| 630 |
+
"grad_norm": 5.819805029378425,
|
| 631 |
+
"learning_rate": 2e-05,
|
| 632 |
+
"loss": 0.6051603317260742,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.11511895625479662,
|
| 637 |
+
"grad_norm": 2.6668997063491373,
|
| 638 |
+
"learning_rate": 2e-05,
|
| 639 |
+
"loss": 0.6632390022277832,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.11639805576873881,
|
| 644 |
+
"grad_norm": 10.472386585595824,
|
| 645 |
+
"learning_rate": 2e-05,
|
| 646 |
+
"loss": 0.5980897903442383,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.117677155282681,
|
| 651 |
+
"grad_norm": 1.6504750774509547,
|
| 652 |
+
"learning_rate": 2e-05,
|
| 653 |
+
"loss": 0.6363026618957519,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.11895625479662318,
|
| 658 |
+
"grad_norm": 2.2303029332193622,
|
| 659 |
+
"learning_rate": 2e-05,
|
| 660 |
+
"loss": 0.6277225494384766,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.12023535431056537,
|
| 665 |
+
"grad_norm": 6.614007656325677,
|
| 666 |
+
"learning_rate": 2e-05,
|
| 667 |
+
"loss": 0.628856086730957,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.12151445382450754,
|
| 672 |
+
"grad_norm": 2.8583085848948158,
|
| 673 |
+
"learning_rate": 2e-05,
|
| 674 |
+
"loss": 0.645742130279541,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.12279355333844973,
|
| 679 |
+
"grad_norm": 4.562616521022261,
|
| 680 |
+
"learning_rate": 2e-05,
|
| 681 |
+
"loss": 0.6232656478881836,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.12407265285239191,
|
| 686 |
+
"grad_norm": 1.6307222970631774,
|
| 687 |
+
"learning_rate": 2e-05,
|
| 688 |
+
"loss": 0.6029533386230469,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.1253517523663341,
|
| 693 |
+
"grad_norm": 2.300518163213002,
|
| 694 |
+
"learning_rate": 2e-05,
|
| 695 |
+
"loss": 0.6161417961120605,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.12663085188027629,
|
| 700 |
+
"grad_norm": 6.2028343698878,
|
| 701 |
+
"learning_rate": 2e-05,
|
| 702 |
+
"loss": 0.6521723747253418,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.12790995139421846,
|
| 707 |
+
"grad_norm": 1.3776781323051128,
|
| 708 |
+
"learning_rate": 2e-05,
|
| 709 |
+
"loss": 0.6107587814331055,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.12918905090816066,
|
| 714 |
+
"grad_norm": 20.073643216938986,
|
| 715 |
+
"learning_rate": 2e-05,
|
| 716 |
+
"loss": 0.6437624931335449,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.13046815042210283,
|
| 721 |
+
"grad_norm": 1.9033248345662102,
|
| 722 |
+
"learning_rate": 2e-05,
|
| 723 |
+
"loss": 0.6639025688171387,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.13174724993604503,
|
| 728 |
+
"grad_norm": 2.6068994484059096,
|
| 729 |
+
"learning_rate": 2e-05,
|
| 730 |
+
"loss": 0.6434185981750489,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.1330263494499872,
|
| 735 |
+
"grad_norm": 3.3603545258908336,
|
| 736 |
+
"learning_rate": 2e-05,
|
| 737 |
+
"loss": 0.6232691764831543,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.1343054489639294,
|
| 742 |
+
"grad_norm": 6.927639779770455,
|
| 743 |
+
"learning_rate": 2e-05,
|
| 744 |
+
"loss": 0.6380214691162109,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.13558454847787157,
|
| 749 |
+
"grad_norm": 4.819538292193655,
|
| 750 |
+
"learning_rate": 2e-05,
|
| 751 |
+
"loss": 0.613338041305542,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.13686364799181377,
|
| 756 |
+
"grad_norm": 2.5506324496776234,
|
| 757 |
+
"learning_rate": 2e-05,
|
| 758 |
+
"loss": 0.6329474449157715,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.13814274750575595,
|
| 763 |
+
"grad_norm": 2.3190465926542494,
|
| 764 |
+
"learning_rate": 2e-05,
|
| 765 |
+
"loss": 0.6157898902893066,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.13942184701969812,
|
| 770 |
+
"grad_norm": 1.9441146816676962,
|
| 771 |
+
"learning_rate": 2e-05,
|
| 772 |
+
"loss": 0.6196088314056396,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.14070094653364032,
|
| 777 |
+
"grad_norm": 1.3698539182342104,
|
| 778 |
+
"learning_rate": 2e-05,
|
| 779 |
+
"loss": 0.6251972198486329,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.1419800460475825,
|
| 784 |
+
"grad_norm": 1.6861606365575037,
|
| 785 |
+
"learning_rate": 2e-05,
|
| 786 |
+
"loss": 0.6314759731292725,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.1432591455615247,
|
| 791 |
+
"grad_norm": 2.462889582220508,
|
| 792 |
+
"learning_rate": 2e-05,
|
| 793 |
+
"loss": 0.65068359375,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.14453824507546686,
|
| 798 |
+
"grad_norm": 1.7742424329216033,
|
| 799 |
+
"learning_rate": 2e-05,
|
| 800 |
+
"loss": 0.6642440795898438,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.14581734458940906,
|
| 805 |
+
"grad_norm": 2.320157848873566,
|
| 806 |
+
"learning_rate": 2e-05,
|
| 807 |
+
"loss": 0.6293258666992188,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.14709644410335124,
|
| 812 |
+
"grad_norm": 2.1787448877032576,
|
| 813 |
+
"learning_rate": 2e-05,
|
| 814 |
+
"loss": 0.6295669555664063,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.14837554361729344,
|
| 819 |
+
"grad_norm": 1.9357128893077755,
|
| 820 |
+
"learning_rate": 2e-05,
|
| 821 |
+
"loss": 0.6496104240417481,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.1496546431312356,
|
| 826 |
+
"grad_norm": 4.110272654702471,
|
| 827 |
+
"learning_rate": 2e-05,
|
| 828 |
+
"loss": 0.6259574890136719,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.15093374264517778,
|
| 833 |
+
"grad_norm": 2.12522137703711,
|
| 834 |
+
"learning_rate": 2e-05,
|
| 835 |
+
"loss": 0.6465348720550537,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.15221284215911998,
|
| 840 |
+
"grad_norm": 7.509928739655393,
|
| 841 |
+
"learning_rate": 2e-05,
|
| 842 |
+
"loss": 0.6745663642883301,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.15349194167306215,
|
| 847 |
+
"grad_norm": 2.1546043605789813,
|
| 848 |
+
"learning_rate": 2e-05,
|
| 849 |
+
"loss": 0.6271842956542969,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.15477104118700435,
|
| 854 |
+
"grad_norm": 4.245976725772466,
|
| 855 |
+
"learning_rate": 2e-05,
|
| 856 |
+
"loss": 0.6347070217132569,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.15605014070094653,
|
| 861 |
+
"grad_norm": 6.284489606731905,
|
| 862 |
+
"learning_rate": 2e-05,
|
| 863 |
+
"loss": 0.6175366878509522,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.15732924021488873,
|
| 868 |
+
"grad_norm": 2.3968896694270403,
|
| 869 |
+
"learning_rate": 2e-05,
|
| 870 |
+
"loss": 0.6528811931610108,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.1586083397288309,
|
| 875 |
+
"grad_norm": 4.44938644957285,
|
| 876 |
+
"learning_rate": 2e-05,
|
| 877 |
+
"loss": 0.6882292747497558,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.1598874392427731,
|
| 882 |
+
"grad_norm": 1.8152041373141536,
|
| 883 |
+
"learning_rate": 2e-05,
|
| 884 |
+
"loss": 0.639340877532959,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.16116653875671527,
|
| 889 |
+
"grad_norm": 1.788688964184629,
|
| 890 |
+
"learning_rate": 2e-05,
|
| 891 |
+
"loss": 0.6168077468872071,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.16244563827065744,
|
| 896 |
+
"grad_norm": 3.011270442070237,
|
| 897 |
+
"learning_rate": 2e-05,
|
| 898 |
+
"loss": 0.6271864891052246,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.16372473778459964,
|
| 903 |
+
"grad_norm": 3.5372237578631034,
|
| 904 |
+
"learning_rate": 2e-05,
|
| 905 |
+
"loss": 0.6144254684448243,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.16500383729854182,
|
| 910 |
+
"grad_norm": 3.578563468270323,
|
| 911 |
+
"learning_rate": 2e-05,
|
| 912 |
+
"loss": 0.6395900249481201,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.16628293681248402,
|
| 917 |
+
"grad_norm": 2.6762637970824787,
|
| 918 |
+
"learning_rate": 2e-05,
|
| 919 |
+
"loss": 0.63500075340271,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.1675620363264262,
|
| 924 |
+
"grad_norm": 3.1705154085918235,
|
| 925 |
+
"learning_rate": 2e-05,
|
| 926 |
+
"loss": 0.62928786277771,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.1688411358403684,
|
| 931 |
+
"grad_norm": 2.9443177248500025,
|
| 932 |
+
"learning_rate": 2e-05,
|
| 933 |
+
"loss": 0.6346927165985108,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.17012023535431056,
|
| 938 |
+
"grad_norm": 2.7168177942420817,
|
| 939 |
+
"learning_rate": 2e-05,
|
| 940 |
+
"loss": 0.636958646774292,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.17139933486825276,
|
| 945 |
+
"grad_norm": 5.3238376411308055,
|
| 946 |
+
"learning_rate": 2e-05,
|
| 947 |
+
"loss": 0.6074342250823974,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.17267843438219493,
|
| 952 |
+
"grad_norm": 1.981751065904129,
|
| 953 |
+
"learning_rate": 2e-05,
|
| 954 |
+
"loss": 0.626393985748291,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.1739575338961371,
|
| 959 |
+
"grad_norm": 6.208058419015704,
|
| 960 |
+
"learning_rate": 2e-05,
|
| 961 |
+
"loss": 0.6492095947265625,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.1752366334100793,
|
| 966 |
+
"grad_norm": 3.2819184556188,
|
| 967 |
+
"learning_rate": 2e-05,
|
| 968 |
+
"loss": 0.6482612609863281,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.17651573292402148,
|
| 973 |
+
"grad_norm": 2.59120863821285,
|
| 974 |
+
"learning_rate": 2e-05,
|
| 975 |
+
"loss": 0.6150693416595459,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.17779483243796368,
|
| 980 |
+
"grad_norm": 1.8538499529211816,
|
| 981 |
+
"learning_rate": 2e-05,
|
| 982 |
+
"loss": 0.6163236618041992,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.17907393195190585,
|
| 987 |
+
"grad_norm": 2.7084822537552924,
|
| 988 |
+
"learning_rate": 2e-05,
|
| 989 |
+
"loss": 0.6104673385620117,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.18035303146584805,
|
| 994 |
+
"grad_norm": 4.439571622311464,
|
| 995 |
+
"learning_rate": 2e-05,
|
| 996 |
+
"loss": 0.6378762722015381,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.18163213097979022,
|
| 1001 |
+
"grad_norm": 1.3621008857551957,
|
| 1002 |
+
"learning_rate": 2e-05,
|
| 1003 |
+
"loss": 0.6388207912445069,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.18291123049373242,
|
| 1008 |
+
"grad_norm": 1.2226259864885263,
|
| 1009 |
+
"learning_rate": 2e-05,
|
| 1010 |
+
"loss": 0.6120697021484375,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.1841903300076746,
|
| 1015 |
+
"grad_norm": 2.1688106933527322,
|
| 1016 |
+
"learning_rate": 2e-05,
|
| 1017 |
+
"loss": 0.5997806549072265,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.18546942952161677,
|
| 1022 |
+
"grad_norm": 1.8126305762978137,
|
| 1023 |
+
"learning_rate": 2e-05,
|
| 1024 |
+
"loss": 0.6054566383361817,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.18674852903555897,
|
| 1029 |
+
"grad_norm": 1.5781503725015227,
|
| 1030 |
+
"learning_rate": 2e-05,
|
| 1031 |
+
"loss": 0.6335046291351318,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.18802762854950114,
|
| 1036 |
+
"grad_norm": 3.525019895765238,
|
| 1037 |
+
"learning_rate": 2e-05,
|
| 1038 |
+
"loss": 0.631711196899414,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.18930672806344334,
|
| 1043 |
+
"grad_norm": 2.713602868229821,
|
| 1044 |
+
"learning_rate": 2e-05,
|
| 1045 |
+
"loss": 0.6091978073120117,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.1905858275773855,
|
| 1050 |
+
"grad_norm": 2.16321468664903,
|
| 1051 |
+
"learning_rate": 2e-05,
|
| 1052 |
+
"loss": 0.6511111259460449,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.1918649270913277,
|
| 1057 |
+
"grad_norm": 1.3249486358544464,
|
| 1058 |
+
"learning_rate": 2e-05,
|
| 1059 |
+
"loss": 0.6071283340454101,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.1931440266052699,
|
| 1064 |
+
"grad_norm": 1.4458246810171957,
|
| 1065 |
+
"learning_rate": 2e-05,
|
| 1066 |
+
"loss": 0.6332869529724121,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.1944231261192121,
|
| 1071 |
+
"grad_norm": 2.10165435585039,
|
| 1072 |
+
"learning_rate": 2e-05,
|
| 1073 |
+
"loss": 0.6224959850311279,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.19570222563315426,
|
| 1078 |
+
"grad_norm": 1.4509883881939123,
|
| 1079 |
+
"learning_rate": 2e-05,
|
| 1080 |
+
"loss": 0.5771361351013183,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.19698132514709643,
|
| 1085 |
+
"grad_norm": 4.326909238866829,
|
| 1086 |
+
"learning_rate": 2e-05,
|
| 1087 |
+
"loss": 0.637632942199707,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.19826042466103863,
|
| 1092 |
+
"grad_norm": 1.6950057566100827,
|
| 1093 |
+
"learning_rate": 2e-05,
|
| 1094 |
+
"loss": 0.6042330741882325,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.1995395241749808,
|
| 1099 |
+
"grad_norm": 1.417431774916637,
|
| 1100 |
+
"learning_rate": 2e-05,
|
| 1101 |
+
"loss": 0.6295817375183106,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.200818623688923,
|
| 1106 |
+
"grad_norm": 1.2405770238129155,
|
| 1107 |
+
"learning_rate": 2e-05,
|
| 1108 |
+
"loss": 0.5979316711425782,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.20209772320286518,
|
| 1113 |
+
"grad_norm": 1.1186847226524728,
|
| 1114 |
+
"learning_rate": 2e-05,
|
| 1115 |
+
"loss": 0.6549808502197265,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.20337682271680738,
|
| 1120 |
+
"grad_norm": 0.9806293817291409,
|
| 1121 |
+
"learning_rate": 2e-05,
|
| 1122 |
+
"loss": 0.6115755081176758,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.20465592223074955,
|
| 1127 |
+
"grad_norm": 0.9738594430445625,
|
| 1128 |
+
"learning_rate": 2e-05,
|
| 1129 |
+
"loss": 0.6076329231262207,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.20593502174469175,
|
| 1134 |
+
"grad_norm": 0.9886089005609494,
|
| 1135 |
+
"learning_rate": 2e-05,
|
| 1136 |
+
"loss": 0.5939586162567139,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.20721412125863392,
|
| 1141 |
+
"grad_norm": 1.3917061095403442,
|
| 1142 |
+
"learning_rate": 2e-05,
|
| 1143 |
+
"loss": 0.6180770874023438,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.2084932207725761,
|
| 1148 |
+
"grad_norm": 1.269503803416184,
|
| 1149 |
+
"learning_rate": 2e-05,
|
| 1150 |
+
"loss": 0.5889194488525391,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.2097723202865183,
|
| 1155 |
+
"grad_norm": 0.949283862884103,
|
| 1156 |
+
"learning_rate": 2e-05,
|
| 1157 |
+
"loss": 0.598993968963623,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.21105141980046047,
|
| 1162 |
+
"grad_norm": 1.3178112537921784,
|
| 1163 |
+
"learning_rate": 2e-05,
|
| 1164 |
+
"loss": 0.62374267578125,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.21233051931440267,
|
| 1169 |
+
"grad_norm": 3.3463082639228907,
|
| 1170 |
+
"learning_rate": 2e-05,
|
| 1171 |
+
"loss": 0.6416094779968262,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.21360961882834484,
|
| 1176 |
+
"grad_norm": 3.5250533260833783,
|
| 1177 |
+
"learning_rate": 2e-05,
|
| 1178 |
+
"loss": 0.5813175201416015,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.21488871834228704,
|
| 1183 |
+
"grad_norm": 3.1862615057182113,
|
| 1184 |
+
"learning_rate": 2e-05,
|
| 1185 |
+
"loss": 0.6030552864074707,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.2161678178562292,
|
| 1190 |
+
"grad_norm": 1.561881602369914,
|
| 1191 |
+
"learning_rate": 2e-05,
|
| 1192 |
+
"loss": 0.6324304580688477,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.2174469173701714,
|
| 1197 |
+
"grad_norm": 2.2515726293089977,
|
| 1198 |
+
"learning_rate": 2e-05,
|
| 1199 |
+
"loss": 0.6021174907684326,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.21872601688411358,
|
| 1204 |
+
"grad_norm": 2.882171305521348,
|
| 1205 |
+
"learning_rate": 2e-05,
|
| 1206 |
+
"loss": 0.6039133071899414,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.22000511639805576,
|
| 1211 |
+
"grad_norm": 1.7822903956791751,
|
| 1212 |
+
"learning_rate": 2e-05,
|
| 1213 |
+
"loss": 0.6312612056732178,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.22128421591199796,
|
| 1218 |
+
"grad_norm": 1.020240331434232,
|
| 1219 |
+
"learning_rate": 2e-05,
|
| 1220 |
+
"loss": 0.6174317836761475,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.22256331542594013,
|
| 1225 |
+
"grad_norm": 1.2001544084835765,
|
| 1226 |
+
"learning_rate": 2e-05,
|
| 1227 |
+
"loss": 0.6398989677429199,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.22384241493988233,
|
| 1232 |
+
"grad_norm": 1.47402791893427,
|
| 1233 |
+
"learning_rate": 2e-05,
|
| 1234 |
+
"loss": 0.6119761943817139,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.2251215144538245,
|
| 1239 |
+
"grad_norm": 2.055369256527985,
|
| 1240 |
+
"learning_rate": 2e-05,
|
| 1241 |
+
"loss": 0.6196990013122559,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.2264006139677667,
|
| 1246 |
+
"grad_norm": 1.5190061918217799,
|
| 1247 |
+
"learning_rate": 2e-05,
|
| 1248 |
+
"loss": 0.60753173828125,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.22767971348170887,
|
| 1253 |
+
"grad_norm": 1.0211065145486882,
|
| 1254 |
+
"learning_rate": 2e-05,
|
| 1255 |
+
"loss": 0.6249045372009278,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.22895881299565107,
|
| 1260 |
+
"grad_norm": 3.530948994051942,
|
| 1261 |
+
"learning_rate": 2e-05,
|
| 1262 |
+
"loss": 0.5920787811279297,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.23023791250959325,
|
| 1267 |
+
"grad_norm": 5.0504619669596185,
|
| 1268 |
+
"learning_rate": 2e-05,
|
| 1269 |
+
"loss": 0.6117641448974609,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.23151701202353542,
|
| 1274 |
+
"grad_norm": 4.544881002014028,
|
| 1275 |
+
"learning_rate": 2e-05,
|
| 1276 |
+
"loss": 0.6159785747528076,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.23279611153747762,
|
| 1281 |
+
"grad_norm": 5.447221215607949,
|
| 1282 |
+
"learning_rate": 2e-05,
|
| 1283 |
+
"loss": 0.5855494499206543,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.2340752110514198,
|
| 1288 |
+
"grad_norm": 1.354746137837009,
|
| 1289 |
+
"learning_rate": 2e-05,
|
| 1290 |
+
"loss": 0.6252618312835694,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.235354310565362,
|
| 1295 |
+
"grad_norm": 1.2175526202028102,
|
| 1296 |
+
"learning_rate": 2e-05,
|
| 1297 |
+
"loss": 0.6352687358856202,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.23663341007930416,
|
| 1302 |
+
"grad_norm": 1.5626912067449465,
|
| 1303 |
+
"learning_rate": 2e-05,
|
| 1304 |
+
"loss": 0.6236725807189941,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.23791250959324636,
|
| 1309 |
+
"grad_norm": 2.418620586616736,
|
| 1310 |
+
"learning_rate": 2e-05,
|
| 1311 |
+
"loss": 0.6202504634857178,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.23919160910718854,
|
| 1316 |
+
"grad_norm": 6.501326474729567,
|
| 1317 |
+
"learning_rate": 2e-05,
|
| 1318 |
+
"loss": 0.6046298980712891,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.24047070862113074,
|
| 1323 |
+
"grad_norm": 13.332115309703186,
|
| 1324 |
+
"learning_rate": 2e-05,
|
| 1325 |
+
"loss": 0.6176700592041016,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.2417498081350729,
|
| 1330 |
+
"grad_norm": 1.5564620734124854,
|
| 1331 |
+
"learning_rate": 2e-05,
|
| 1332 |
+
"loss": 0.6322036743164062,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.24302890764901508,
|
| 1337 |
+
"grad_norm": 1.3252946439407358,
|
| 1338 |
+
"learning_rate": 2e-05,
|
| 1339 |
+
"loss": 0.6052451610565186,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.24430800716295728,
|
| 1344 |
+
"grad_norm": 3.7754178720563902,
|
| 1345 |
+
"learning_rate": 2e-05,
|
| 1346 |
+
"loss": 0.621720838546753,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.24558710667689945,
|
| 1351 |
+
"grad_norm": 1.7709721126063882,
|
| 1352 |
+
"learning_rate": 2e-05,
|
| 1353 |
+
"loss": 0.6091246604919434,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.24686620619084165,
|
| 1358 |
+
"grad_norm": 1.3070411526030667,
|
| 1359 |
+
"learning_rate": 2e-05,
|
| 1360 |
+
"loss": 0.6220854759216309,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.24814530570478383,
|
| 1365 |
+
"grad_norm": 1.8401872322058064,
|
| 1366 |
+
"learning_rate": 2e-05,
|
| 1367 |
+
"loss": 0.6449881553649902,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.24942440521872603,
|
| 1372 |
+
"grad_norm": 2.4285418670456496,
|
| 1373 |
+
"learning_rate": 2e-05,
|
| 1374 |
+
"loss": 0.594057559967041,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.2507035047326682,
|
| 1379 |
+
"grad_norm": 1.7235239750037874,
|
| 1380 |
+
"learning_rate": 2e-05,
|
| 1381 |
+
"loss": 0.6182984828948974,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.2519826042466104,
|
| 1386 |
+
"grad_norm": 1.6285019660644895,
|
| 1387 |
+
"learning_rate": 2e-05,
|
| 1388 |
+
"loss": 0.5909505367279053,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.25326170376055257,
|
| 1393 |
+
"grad_norm": 1.2819655053500427,
|
| 1394 |
+
"learning_rate": 2e-05,
|
| 1395 |
+
"loss": 0.6325978755950927,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.25454080327449474,
|
| 1400 |
+
"grad_norm": 1.2159380347170259,
|
| 1401 |
+
"learning_rate": 2e-05,
|
| 1402 |
+
"loss": 0.5887606620788575,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.2558199027884369,
|
| 1407 |
+
"grad_norm": 0.9505599796559441,
|
| 1408 |
+
"learning_rate": 2e-05,
|
| 1409 |
+
"loss": 0.574717903137207,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.25709900230237914,
|
| 1414 |
+
"grad_norm": 1.2581472202385364,
|
| 1415 |
+
"learning_rate": 2e-05,
|
| 1416 |
+
"loss": 0.5856706142425537,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.2583781018163213,
|
| 1421 |
+
"grad_norm": 1.1363388383660025,
|
| 1422 |
+
"learning_rate": 2e-05,
|
| 1423 |
+
"loss": 0.5883913040161133,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.2596572013302635,
|
| 1428 |
+
"grad_norm": 1.560965561232806,
|
| 1429 |
+
"learning_rate": 2e-05,
|
| 1430 |
+
"loss": 0.5746139049530029,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.26093630084420566,
|
| 1435 |
+
"grad_norm": 2.1211040191156343,
|
| 1436 |
+
"learning_rate": 2e-05,
|
| 1437 |
+
"loss": 0.6083070755004882,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.2622154003581479,
|
| 1442 |
+
"grad_norm": 1.713441497273608,
|
| 1443 |
+
"learning_rate": 2e-05,
|
| 1444 |
+
"loss": 0.6275928020477295,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.26349449987209006,
|
| 1449 |
+
"grad_norm": 2.038098915097047,
|
| 1450 |
+
"learning_rate": 2e-05,
|
| 1451 |
+
"loss": 0.6330203533172607,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.26477359938603223,
|
| 1456 |
+
"grad_norm": 1.2264196348153609,
|
| 1457 |
+
"learning_rate": 2e-05,
|
| 1458 |
+
"loss": 0.6101226806640625,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.2660526988999744,
|
| 1463 |
+
"grad_norm": 1.8210121510445536,
|
| 1464 |
+
"learning_rate": 2e-05,
|
| 1465 |
+
"loss": 0.6420550346374512,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.2673317984139166,
|
| 1470 |
+
"grad_norm": 0.8963201860659803,
|
| 1471 |
+
"learning_rate": 2e-05,
|
| 1472 |
+
"loss": 0.5967807769775391,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.2686108979278588,
|
| 1477 |
+
"grad_norm": 2.0260697664647003,
|
| 1478 |
+
"learning_rate": 2e-05,
|
| 1479 |
+
"loss": 0.6149022579193115,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.269889997441801,
|
| 1484 |
+
"grad_norm": 4.914335014084526,
|
| 1485 |
+
"learning_rate": 2e-05,
|
| 1486 |
+
"loss": 0.6114978790283203,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.27116909695574315,
|
| 1491 |
+
"grad_norm": 2.106317779910681,
|
| 1492 |
+
"learning_rate": 2e-05,
|
| 1493 |
+
"loss": 0.6006545066833496,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.2724481964696853,
|
| 1498 |
+
"grad_norm": 2.7288315662600158,
|
| 1499 |
+
"learning_rate": 2e-05,
|
| 1500 |
+
"loss": 0.5773192882537842,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.27372729598362755,
|
| 1505 |
+
"grad_norm": 1.2064614446581332,
|
| 1506 |
+
"learning_rate": 2e-05,
|
| 1507 |
+
"loss": 0.5988784313201905,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.2750063954975697,
|
| 1512 |
+
"grad_norm": 1.4196592877324987,
|
| 1513 |
+
"learning_rate": 2e-05,
|
| 1514 |
+
"loss": 0.6252477169036865,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.2762854950115119,
|
| 1519 |
+
"grad_norm": 5.085511259042394,
|
| 1520 |
+
"learning_rate": 2e-05,
|
| 1521 |
+
"loss": 0.6232048034667969,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.27756459452545407,
|
| 1526 |
+
"grad_norm": 1.6573265068305214,
|
| 1527 |
+
"learning_rate": 2e-05,
|
| 1528 |
+
"loss": 0.5724109172821045,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.27884369403939624,
|
| 1533 |
+
"grad_norm": 0.8570570183137078,
|
| 1534 |
+
"learning_rate": 2e-05,
|
| 1535 |
+
"loss": 0.6122853755950928,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.28012279355333847,
|
| 1540 |
+
"grad_norm": 1.0409675006077839,
|
| 1541 |
+
"learning_rate": 2e-05,
|
| 1542 |
+
"loss": 0.5958579540252685,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.28140189306728064,
|
| 1547 |
+
"grad_norm": 2.850717441256514,
|
| 1548 |
+
"learning_rate": 2e-05,
|
| 1549 |
+
"loss": 0.61070556640625,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.2826809925812228,
|
| 1554 |
+
"grad_norm": 1.674482950677382,
|
| 1555 |
+
"learning_rate": 2e-05,
|
| 1556 |
+
"loss": 0.6583208084106446,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.283960092095165,
|
| 1561 |
+
"grad_norm": 1.031060356132696,
|
| 1562 |
+
"learning_rate": 2e-05,
|
| 1563 |
+
"loss": 0.5889376640319824,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.2852391916091072,
|
| 1568 |
+
"grad_norm": 1.2666316468928664,
|
| 1569 |
+
"learning_rate": 2e-05,
|
| 1570 |
+
"loss": 0.6129489421844483,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.2865182911230494,
|
| 1575 |
+
"grad_norm": 1.6411785567635628,
|
| 1576 |
+
"learning_rate": 2e-05,
|
| 1577 |
+
"loss": 0.5777743816375732,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.28779739063699156,
|
| 1582 |
+
"grad_norm": 1.8797050463236573,
|
| 1583 |
+
"learning_rate": 2e-05,
|
| 1584 |
+
"loss": 0.6185701370239258,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.28907649015093373,
|
| 1589 |
+
"grad_norm": 1.1158819337514871,
|
| 1590 |
+
"learning_rate": 2e-05,
|
| 1591 |
+
"loss": 0.6073112487792969,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.2903555896648759,
|
| 1596 |
+
"grad_norm": 1.260587701397972,
|
| 1597 |
+
"learning_rate": 2e-05,
|
| 1598 |
+
"loss": 0.604709243774414,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.29163468917881813,
|
| 1603 |
+
"grad_norm": 2.394044170108493,
|
| 1604 |
+
"learning_rate": 2e-05,
|
| 1605 |
+
"loss": 0.595456886291504,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.2929137886927603,
|
| 1610 |
+
"grad_norm": 1.857401244405304,
|
| 1611 |
+
"learning_rate": 2e-05,
|
| 1612 |
+
"loss": 0.6224223136901855,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.2941928882067025,
|
| 1617 |
+
"grad_norm": 2.2784297913171336,
|
| 1618 |
+
"learning_rate": 2e-05,
|
| 1619 |
+
"loss": 0.6300773620605469,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.29547198772064465,
|
| 1624 |
+
"grad_norm": 0.9471614831952674,
|
| 1625 |
+
"learning_rate": 2e-05,
|
| 1626 |
+
"loss": 0.6309947013854981,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.2967510872345869,
|
| 1631 |
+
"grad_norm": 1.1542869310513502,
|
| 1632 |
+
"learning_rate": 2e-05,
|
| 1633 |
+
"loss": 0.6161383628845215,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.29803018674852905,
|
| 1638 |
+
"grad_norm": 0.9695713668697032,
|
| 1639 |
+
"learning_rate": 2e-05,
|
| 1640 |
+
"loss": 0.6323893547058106,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.2993092862624712,
|
| 1645 |
+
"grad_norm": 0.9319436233814953,
|
| 1646 |
+
"learning_rate": 2e-05,
|
| 1647 |
+
"loss": 0.5904321193695068,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.3005883857764134,
|
| 1652 |
+
"grad_norm": 1.227810386425401,
|
| 1653 |
+
"learning_rate": 2e-05,
|
| 1654 |
+
"loss": 0.6253926277160644,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.30186748529035556,
|
| 1659 |
+
"grad_norm": 1.7096380269139588,
|
| 1660 |
+
"learning_rate": 2e-05,
|
| 1661 |
+
"loss": 0.5987714290618896,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.3031465848042978,
|
| 1666 |
+
"grad_norm": 3.3222742211615315,
|
| 1667 |
+
"learning_rate": 2e-05,
|
| 1668 |
+
"loss": 0.609607458114624,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.30442568431823996,
|
| 1673 |
+
"grad_norm": 1.0979241510256794,
|
| 1674 |
+
"learning_rate": 2e-05,
|
| 1675 |
+
"loss": 0.5992640495300293,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.30570478383218214,
|
| 1680 |
+
"grad_norm": 3.532618720860108,
|
| 1681 |
+
"learning_rate": 2e-05,
|
| 1682 |
+
"loss": 0.6203838348388672,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.3069838833461243,
|
| 1687 |
+
"grad_norm": 1.22304004334949,
|
| 1688 |
+
"learning_rate": 2e-05,
|
| 1689 |
+
"loss": 0.5954967021942139,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.30826298286006654,
|
| 1694 |
+
"grad_norm": 2.410293308883467,
|
| 1695 |
+
"learning_rate": 2e-05,
|
| 1696 |
+
"loss": 0.6060263633728027,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.3095420823740087,
|
| 1701 |
+
"grad_norm": 1.2560451534079777,
|
| 1702 |
+
"learning_rate": 2e-05,
|
| 1703 |
+
"loss": 0.6317928314208985,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.3108211818879509,
|
| 1708 |
+
"grad_norm": 1.0455056153829623,
|
| 1709 |
+
"learning_rate": 2e-05,
|
| 1710 |
+
"loss": 0.5984413146972656,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.31210028140189305,
|
| 1715 |
+
"grad_norm": 1.650132689260639,
|
| 1716 |
+
"learning_rate": 2e-05,
|
| 1717 |
+
"loss": 0.6033206939697265,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.3133793809158352,
|
| 1722 |
+
"grad_norm": 2.5592592325487105,
|
| 1723 |
+
"learning_rate": 2e-05,
|
| 1724 |
+
"loss": 0.5678141593933106,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.31465848042977745,
|
| 1729 |
+
"grad_norm": 1.562608097897649,
|
| 1730 |
+
"learning_rate": 2e-05,
|
| 1731 |
+
"loss": 0.5921920776367188,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.3159375799437196,
|
| 1736 |
+
"grad_norm": 2.192241892940132,
|
| 1737 |
+
"learning_rate": 2e-05,
|
| 1738 |
+
"loss": 0.598881721496582,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.3172166794576618,
|
| 1743 |
+
"grad_norm": 0.9967509265789414,
|
| 1744 |
+
"learning_rate": 2e-05,
|
| 1745 |
+
"loss": 0.575558090209961,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.31849577897160397,
|
| 1750 |
+
"grad_norm": 4.787238755740847,
|
| 1751 |
+
"learning_rate": 2e-05,
|
| 1752 |
+
"loss": 0.5959879875183105,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.3197748784855462,
|
| 1757 |
+
"grad_norm": 1.1428325653163887,
|
| 1758 |
+
"learning_rate": 2e-05,
|
| 1759 |
+
"loss": 0.5783063411712647,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.32105397799948837,
|
| 1764 |
+
"grad_norm": 2.9202327745384973,
|
| 1765 |
+
"learning_rate": 2e-05,
|
| 1766 |
+
"loss": 0.5774745941162109,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.32233307751343054,
|
| 1771 |
+
"grad_norm": 2.0279978562828735,
|
| 1772 |
+
"learning_rate": 2e-05,
|
| 1773 |
+
"loss": 0.6202166557312012,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.3236121770273727,
|
| 1778 |
+
"grad_norm": 1.5802874918294092,
|
| 1779 |
+
"learning_rate": 2e-05,
|
| 1780 |
+
"loss": 0.6294447898864746,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.3248912765413149,
|
| 1785 |
+
"grad_norm": 0.9906512101701456,
|
| 1786 |
+
"learning_rate": 2e-05,
|
| 1787 |
+
"loss": 0.5678004264831543,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.3261703760552571,
|
| 1792 |
+
"grad_norm": 1.746760187852664,
|
| 1793 |
+
"learning_rate": 2e-05,
|
| 1794 |
+
"loss": 0.6238263607025146,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.3274494755691993,
|
| 1799 |
+
"grad_norm": 2.8108078554248594,
|
| 1800 |
+
"learning_rate": 2e-05,
|
| 1801 |
+
"loss": 0.6203858375549316,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.32872857508314146,
|
| 1806 |
+
"grad_norm": 4.806066554618259,
|
| 1807 |
+
"learning_rate": 2e-05,
|
| 1808 |
+
"loss": 0.6184251308441162,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.33000767459708363,
|
| 1813 |
+
"grad_norm": 2.0113483642991965,
|
| 1814 |
+
"learning_rate": 2e-05,
|
| 1815 |
+
"loss": 0.6477134704589844,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.33128677411102586,
|
| 1820 |
+
"grad_norm": 1.3830922692363252,
|
| 1821 |
+
"learning_rate": 2e-05,
|
| 1822 |
+
"loss": 0.6008767127990723,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.33256587362496803,
|
| 1827 |
+
"grad_norm": 1.2349526734744556,
|
| 1828 |
+
"learning_rate": 2e-05,
|
| 1829 |
+
"loss": 0.624567699432373,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.3338449731389102,
|
| 1834 |
+
"grad_norm": 1.271755174590234,
|
| 1835 |
+
"learning_rate": 2e-05,
|
| 1836 |
+
"loss": 0.6194814205169678,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.3351240726528524,
|
| 1841 |
+
"grad_norm": 1.4926328848591621,
|
| 1842 |
+
"learning_rate": 2e-05,
|
| 1843 |
+
"loss": 0.5909404754638672,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.33640317216679455,
|
| 1848 |
+
"grad_norm": 1.7225071189531411,
|
| 1849 |
+
"learning_rate": 2e-05,
|
| 1850 |
+
"loss": 0.6224935054779053,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.3376822716807368,
|
| 1855 |
+
"grad_norm": 3.5736648482858246,
|
| 1856 |
+
"learning_rate": 2e-05,
|
| 1857 |
+
"loss": 0.6273334503173829,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.33896137119467895,
|
| 1862 |
+
"grad_norm": 1.6199005667039337,
|
| 1863 |
+
"learning_rate": 2e-05,
|
| 1864 |
+
"loss": 0.6213912010192871,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.3402404707086211,
|
| 1869 |
+
"grad_norm": 1.2784430597805259,
|
| 1870 |
+
"learning_rate": 2e-05,
|
| 1871 |
+
"loss": 0.6026556968688965,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.3415195702225633,
|
| 1876 |
+
"grad_norm": 3.8766837411975454,
|
| 1877 |
+
"learning_rate": 2e-05,
|
| 1878 |
+
"loss": 0.5997205257415772,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.3427986697365055,
|
| 1883 |
+
"grad_norm": 1.522149639598022,
|
| 1884 |
+
"learning_rate": 2e-05,
|
| 1885 |
+
"loss": 0.6241069793701172,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.3440777692504477,
|
| 1890 |
+
"grad_norm": 2.7584445375179047,
|
| 1891 |
+
"learning_rate": 2e-05,
|
| 1892 |
+
"loss": 0.6011125564575195,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.34535686876438987,
|
| 1897 |
+
"grad_norm": 4.487541114475743,
|
| 1898 |
+
"learning_rate": 2e-05,
|
| 1899 |
+
"loss": 0.6028533458709717,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.34663596827833204,
|
| 1904 |
+
"grad_norm": 1.1344741201338224,
|
| 1905 |
+
"learning_rate": 2e-05,
|
| 1906 |
+
"loss": 0.6265992164611817,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.3479150677922742,
|
| 1911 |
+
"grad_norm": 1.264736856329463,
|
| 1912 |
+
"learning_rate": 2e-05,
|
| 1913 |
+
"loss": 0.63673095703125,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.34919416730621644,
|
| 1918 |
+
"grad_norm": 0.8044421663432847,
|
| 1919 |
+
"learning_rate": 2e-05,
|
| 1920 |
+
"loss": 0.602764892578125,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.3504732668201586,
|
| 1925 |
+
"grad_norm": 1.0593200577466688,
|
| 1926 |
+
"learning_rate": 2e-05,
|
| 1927 |
+
"loss": 0.5976439952850342,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.3517523663341008,
|
| 1932 |
+
"grad_norm": 1.1434465645557081,
|
| 1933 |
+
"learning_rate": 2e-05,
|
| 1934 |
+
"loss": 0.5878979682922363,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.35303146584804296,
|
| 1939 |
+
"grad_norm": 1.3903295183809865,
|
| 1940 |
+
"learning_rate": 2e-05,
|
| 1941 |
+
"loss": 0.6098363399505615,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.3543105653619852,
|
| 1946 |
+
"grad_norm": 1.282377166346566,
|
| 1947 |
+
"learning_rate": 2e-05,
|
| 1948 |
+
"loss": 0.5990135192871093,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.35558966487592736,
|
| 1953 |
+
"grad_norm": 1.1435337542967667,
|
| 1954 |
+
"learning_rate": 2e-05,
|
| 1955 |
+
"loss": 0.5908602714538574,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.35686876438986953,
|
| 1960 |
+
"grad_norm": 1.170245442472612,
|
| 1961 |
+
"learning_rate": 2e-05,
|
| 1962 |
+
"loss": 0.6214659690856934,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.3581478639038117,
|
| 1967 |
+
"grad_norm": 1.0601914113434607,
|
| 1968 |
+
"learning_rate": 2e-05,
|
| 1969 |
+
"loss": 0.5969385147094727,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.3594269634177539,
|
| 1974 |
+
"grad_norm": 0.8880429422636937,
|
| 1975 |
+
"learning_rate": 2e-05,
|
| 1976 |
+
"loss": 0.5900382041931153,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.3607060629316961,
|
| 1981 |
+
"grad_norm": 1.3291734362533796,
|
| 1982 |
+
"learning_rate": 2e-05,
|
| 1983 |
+
"loss": 0.5984704971313477,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.3619851624456383,
|
| 1988 |
+
"grad_norm": 2.3196798317877803,
|
| 1989 |
+
"learning_rate": 2e-05,
|
| 1990 |
+
"loss": 0.6031060695648194,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.36326426195958045,
|
| 1995 |
+
"grad_norm": 1.4909813232977367,
|
| 1996 |
+
"learning_rate": 2e-05,
|
| 1997 |
+
"loss": 0.5790597438812256,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.3645433614735226,
|
| 2002 |
+
"grad_norm": 2.7690017473286033,
|
| 2003 |
+
"learning_rate": 2e-05,
|
| 2004 |
+
"loss": 0.577203369140625,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.36582246098746485,
|
| 2009 |
+
"grad_norm": 1.2195966201599757,
|
| 2010 |
+
"learning_rate": 2e-05,
|
| 2011 |
+
"loss": 0.6123817443847657,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.367101560501407,
|
| 2016 |
+
"grad_norm": 3.2580341346851034,
|
| 2017 |
+
"learning_rate": 2e-05,
|
| 2018 |
+
"loss": 0.5920998573303222,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.3683806600153492,
|
| 2023 |
+
"grad_norm": 1.191401593758783,
|
| 2024 |
+
"learning_rate": 2e-05,
|
| 2025 |
+
"loss": 0.6038727760314941,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.36965975952929137,
|
| 2030 |
+
"grad_norm": 1.3743110118592492,
|
| 2031 |
+
"learning_rate": 2e-05,
|
| 2032 |
+
"loss": 0.6003240585327149,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.37093885904323354,
|
| 2037 |
+
"grad_norm": 4.509715704890444,
|
| 2038 |
+
"learning_rate": 2e-05,
|
| 2039 |
+
"loss": 0.5917400360107422,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.37221795855717577,
|
| 2044 |
+
"grad_norm": 0.9032779057730358,
|
| 2045 |
+
"learning_rate": 2e-05,
|
| 2046 |
+
"loss": 0.6044122695922851,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.37349705807111794,
|
| 2051 |
+
"grad_norm": 0.9594536006104978,
|
| 2052 |
+
"learning_rate": 2e-05,
|
| 2053 |
+
"loss": 0.6204854488372803,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.3747761575850601,
|
| 2058 |
+
"grad_norm": 2.9729199422688404,
|
| 2059 |
+
"learning_rate": 2e-05,
|
| 2060 |
+
"loss": 0.6077807426452637,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.3760552570990023,
|
| 2065 |
+
"grad_norm": 2.0925145891159795,
|
| 2066 |
+
"learning_rate": 2e-05,
|
| 2067 |
+
"loss": 0.5951478004455566,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.3773343566129445,
|
| 2072 |
+
"grad_norm": 21.76963806945384,
|
| 2073 |
+
"learning_rate": 2e-05,
|
| 2074 |
+
"loss": 0.5952364921569824,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.3786134561268867,
|
| 2079 |
+
"grad_norm": 1.5329185958740825,
|
| 2080 |
+
"learning_rate": 2e-05,
|
| 2081 |
+
"loss": 0.6125658988952637,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.37989255564082886,
|
| 2086 |
+
"grad_norm": 1.4903676249564757,
|
| 2087 |
+
"learning_rate": 2e-05,
|
| 2088 |
+
"loss": 0.5745217800140381,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.381171655154771,
|
| 2093 |
+
"grad_norm": 1.8185242853324988,
|
| 2094 |
+
"learning_rate": 2e-05,
|
| 2095 |
+
"loss": 0.587047004699707,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.3824507546687132,
|
| 2100 |
+
"grad_norm": 1.1127655033537205,
|
| 2101 |
+
"learning_rate": 2e-05,
|
| 2102 |
+
"loss": 0.5825289726257324,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.3837298541826554,
|
| 2107 |
+
"grad_norm": 2.494189103068932,
|
| 2108 |
+
"learning_rate": 2e-05,
|
| 2109 |
+
"loss": 0.6626487731933594,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.3850089536965976,
|
| 2114 |
+
"grad_norm": 1.1353770749693715,
|
| 2115 |
+
"learning_rate": 2e-05,
|
| 2116 |
+
"loss": 0.5981975555419922,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.3862880532105398,
|
| 2121 |
+
"grad_norm": 4.0866893647819476,
|
| 2122 |
+
"learning_rate": 2e-05,
|
| 2123 |
+
"loss": 0.6148322582244873,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.38756715272448194,
|
| 2128 |
+
"grad_norm": 1.4328489841066114,
|
| 2129 |
+
"learning_rate": 2e-05,
|
| 2130 |
+
"loss": 0.5979141235351563,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.3888462522384242,
|
| 2135 |
+
"grad_norm": 0.9580306575907652,
|
| 2136 |
+
"learning_rate": 2e-05,
|
| 2137 |
+
"loss": 0.5795217514038086,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.39012535175236634,
|
| 2142 |
+
"grad_norm": 1.027657289655969,
|
| 2143 |
+
"learning_rate": 2e-05,
|
| 2144 |
+
"loss": 0.593060827255249,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.3914044512663085,
|
| 2149 |
+
"grad_norm": 2.867852167814524,
|
| 2150 |
+
"learning_rate": 2e-05,
|
| 2151 |
+
"loss": 0.649545955657959,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.3926835507802507,
|
| 2156 |
+
"grad_norm": 1.7315278137528691,
|
| 2157 |
+
"learning_rate": 2e-05,
|
| 2158 |
+
"loss": 0.5818486690521241,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.39396265029419286,
|
| 2163 |
+
"grad_norm": 18.51948790045596,
|
| 2164 |
+
"learning_rate": 2e-05,
|
| 2165 |
+
"loss": 0.6028565883636474,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.3952417498081351,
|
| 2170 |
+
"grad_norm": 1.0814514467211724,
|
| 2171 |
+
"learning_rate": 2e-05,
|
| 2172 |
+
"loss": 0.6369311809539795,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.39652084932207726,
|
| 2177 |
+
"grad_norm": 0.8365819549840845,
|
| 2178 |
+
"learning_rate": 2e-05,
|
| 2179 |
+
"loss": 0.5891385555267334,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.39779994883601943,
|
| 2184 |
+
"grad_norm": 1.1643014580060915,
|
| 2185 |
+
"learning_rate": 2e-05,
|
| 2186 |
+
"loss": 0.620572280883789,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.3990790483499616,
|
| 2191 |
+
"grad_norm": 1.3853634092376967,
|
| 2192 |
+
"learning_rate": 2e-05,
|
| 2193 |
+
"loss": 0.5695016384124756,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.40035814786390383,
|
| 2198 |
+
"grad_norm": 0.941601759431241,
|
| 2199 |
+
"learning_rate": 2e-05,
|
| 2200 |
+
"loss": 0.5876264095306396,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.401637247377846,
|
| 2205 |
+
"grad_norm": 0.9057064185186952,
|
| 2206 |
+
"learning_rate": 2e-05,
|
| 2207 |
+
"loss": 0.635837173461914,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.4029163468917882,
|
| 2212 |
+
"grad_norm": 1.0678199968255322,
|
| 2213 |
+
"learning_rate": 2e-05,
|
| 2214 |
+
"loss": 0.5842601776123046,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.40419544640573035,
|
| 2219 |
+
"grad_norm": 2.353208944812402,
|
| 2220 |
+
"learning_rate": 2e-05,
|
| 2221 |
+
"loss": 0.6364971160888672,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.4054745459196725,
|
| 2226 |
+
"grad_norm": 2.2056815946147945,
|
| 2227 |
+
"learning_rate": 2e-05,
|
| 2228 |
+
"loss": 0.5831233024597168,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.40675364543361475,
|
| 2233 |
+
"grad_norm": 1.0432213491886457,
|
| 2234 |
+
"learning_rate": 2e-05,
|
| 2235 |
+
"loss": 0.6237380981445313,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.4080327449475569,
|
| 2240 |
+
"grad_norm": 1.7016360850124244,
|
| 2241 |
+
"learning_rate": 2e-05,
|
| 2242 |
+
"loss": 0.5941524505615234,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.4093118444614991,
|
| 2247 |
+
"grad_norm": 17.46618090659867,
|
| 2248 |
+
"learning_rate": 2e-05,
|
| 2249 |
+
"loss": 0.6070680141448974,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.41059094397544127,
|
| 2254 |
+
"grad_norm": 1.9361701078093683,
|
| 2255 |
+
"learning_rate": 2e-05,
|
| 2256 |
+
"loss": 0.5642125129699707,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.4118700434893835,
|
| 2261 |
+
"grad_norm": 1.9629421677603713,
|
| 2262 |
+
"learning_rate": 2e-05,
|
| 2263 |
+
"loss": 0.5952259063720703,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.41314914300332567,
|
| 2268 |
+
"grad_norm": 3.65571115063658,
|
| 2269 |
+
"learning_rate": 2e-05,
|
| 2270 |
+
"loss": 0.5965877056121827,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.41442824251726784,
|
| 2275 |
+
"grad_norm": 1.3605239806487905,
|
| 2276 |
+
"learning_rate": 2e-05,
|
| 2277 |
+
"loss": 0.5746373653411865,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.41570734203121,
|
| 2282 |
+
"grad_norm": 1.0451824030026722,
|
| 2283 |
+
"learning_rate": 2e-05,
|
| 2284 |
+
"loss": 0.5859414577484131,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.4169864415451522,
|
| 2289 |
+
"grad_norm": 1.7463120327939081,
|
| 2290 |
+
"learning_rate": 2e-05,
|
| 2291 |
+
"loss": 0.5620139122009278,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.4182655410590944,
|
| 2296 |
+
"grad_norm": 2.138413071925528,
|
| 2297 |
+
"learning_rate": 2e-05,
|
| 2298 |
+
"loss": 0.5906031608581543,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.4195446405730366,
|
| 2303 |
+
"grad_norm": 2.7608510899833103,
|
| 2304 |
+
"learning_rate": 2e-05,
|
| 2305 |
+
"loss": 0.5822212219238281,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.42082374008697876,
|
| 2310 |
+
"grad_norm": 0.9893407948872976,
|
| 2311 |
+
"learning_rate": 2e-05,
|
| 2312 |
+
"loss": 0.6076113700866699,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.42210283960092093,
|
| 2317 |
+
"grad_norm": 3.415050230981063,
|
| 2318 |
+
"learning_rate": 2e-05,
|
| 2319 |
+
"loss": 0.5842854499816894,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.42338193911486316,
|
| 2324 |
+
"grad_norm": 1.2831244261530679,
|
| 2325 |
+
"learning_rate": 2e-05,
|
| 2326 |
+
"loss": 0.6099458694458008,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.42466103862880533,
|
| 2331 |
+
"grad_norm": 1.463679574003703,
|
| 2332 |
+
"learning_rate": 2e-05,
|
| 2333 |
+
"loss": 0.5927666664123535,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.4259401381427475,
|
| 2338 |
+
"grad_norm": 1.2948575443800137,
|
| 2339 |
+
"learning_rate": 2e-05,
|
| 2340 |
+
"loss": 0.6183608055114747,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.4272192376566897,
|
| 2345 |
+
"grad_norm": 1.7719649112060292,
|
| 2346 |
+
"learning_rate": 2e-05,
|
| 2347 |
+
"loss": 0.5920117855072021,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.42849833717063185,
|
| 2352 |
+
"grad_norm": 2.5338780912419314,
|
| 2353 |
+
"learning_rate": 2e-05,
|
| 2354 |
+
"loss": 0.5885765075683593,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.4297774366845741,
|
| 2359 |
+
"grad_norm": 2.1104264661059933,
|
| 2360 |
+
"learning_rate": 2e-05,
|
| 2361 |
+
"loss": 0.5881740093231201,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.43105653619851625,
|
| 2366 |
+
"grad_norm": 3.062281823976241,
|
| 2367 |
+
"learning_rate": 2e-05,
|
| 2368 |
+
"loss": 0.5795594692230225,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.4323356357124584,
|
| 2373 |
+
"grad_norm": 1.1334170743304055,
|
| 2374 |
+
"learning_rate": 2e-05,
|
| 2375 |
+
"loss": 0.6271108150482178,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.4336147352264006,
|
| 2380 |
+
"grad_norm": 1.3336335739798473,
|
| 2381 |
+
"learning_rate": 2e-05,
|
| 2382 |
+
"loss": 0.602003812789917,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.4348938347403428,
|
| 2387 |
+
"grad_norm": 1.3227068900163543,
|
| 2388 |
+
"learning_rate": 2e-05,
|
| 2389 |
+
"loss": 0.6033557891845703,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.436172934254285,
|
| 2394 |
+
"grad_norm": 1.1445072235885272,
|
| 2395 |
+
"learning_rate": 2e-05,
|
| 2396 |
+
"loss": 0.6056077003479003,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.43745203376822717,
|
| 2401 |
+
"grad_norm": 2.8759382031453025,
|
| 2402 |
+
"learning_rate": 2e-05,
|
| 2403 |
+
"loss": 0.5847073554992676,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.43873113328216934,
|
| 2408 |
+
"grad_norm": 2.40286580704749,
|
| 2409 |
+
"learning_rate": 2e-05,
|
| 2410 |
+
"loss": 0.6021082878112793,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.4400102327961115,
|
| 2415 |
+
"grad_norm": 1.058490597268109,
|
| 2416 |
+
"learning_rate": 2e-05,
|
| 2417 |
+
"loss": 0.5841891765594482,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.44128933231005374,
|
| 2422 |
+
"grad_norm": 1.304059233417537,
|
| 2423 |
+
"learning_rate": 2e-05,
|
| 2424 |
+
"loss": 0.6094838619232178,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.4425684318239959,
|
| 2429 |
+
"grad_norm": 0.8060832967578174,
|
| 2430 |
+
"learning_rate": 2e-05,
|
| 2431 |
+
"loss": 0.5862702846527099,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.4438475313379381,
|
| 2436 |
+
"grad_norm": 0.840500573095063,
|
| 2437 |
+
"learning_rate": 2e-05,
|
| 2438 |
+
"loss": 0.5949713706970214,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.44512663085188026,
|
| 2443 |
+
"grad_norm": 4.542032451860691,
|
| 2444 |
+
"learning_rate": 2e-05,
|
| 2445 |
+
"loss": 0.6270683288574219,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.4464057303658225,
|
| 2450 |
+
"grad_norm": 1.0368756465623938,
|
| 2451 |
+
"learning_rate": 2e-05,
|
| 2452 |
+
"loss": 0.606722640991211,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.44768482987976466,
|
| 2457 |
+
"grad_norm": 0.947207216589235,
|
| 2458 |
+
"learning_rate": 2e-05,
|
| 2459 |
+
"loss": 0.5707377433776856,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.44896392939370683,
|
| 2464 |
+
"grad_norm": 1.62305712819688,
|
| 2465 |
+
"learning_rate": 2e-05,
|
| 2466 |
+
"loss": 0.5990262508392334,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.450243028907649,
|
| 2471 |
+
"grad_norm": 1.4535680237402864,
|
| 2472 |
+
"learning_rate": 2e-05,
|
| 2473 |
+
"loss": 0.6101785182952881,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.4515221284215912,
|
| 2478 |
+
"grad_norm": 1.0679867768799687,
|
| 2479 |
+
"learning_rate": 2e-05,
|
| 2480 |
+
"loss": 0.5582299709320069,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.4528012279355334,
|
| 2485 |
+
"grad_norm": 2.029254450147691,
|
| 2486 |
+
"learning_rate": 2e-05,
|
| 2487 |
+
"loss": 0.6040618896484375,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.4540803274494756,
|
| 2492 |
+
"grad_norm": 3.4693073257949103,
|
| 2493 |
+
"learning_rate": 2e-05,
|
| 2494 |
+
"loss": 0.5946227073669433,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.45535942696341775,
|
| 2499 |
+
"grad_norm": 1.8264567925108544,
|
| 2500 |
+
"learning_rate": 2e-05,
|
| 2501 |
+
"loss": 0.6080288887023926,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.4566385264773599,
|
| 2506 |
+
"grad_norm": 1.0695415170895906,
|
| 2507 |
+
"learning_rate": 2e-05,
|
| 2508 |
+
"loss": 0.6208341598510743,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.45791762599130215,
|
| 2513 |
+
"grad_norm": 1.1090984471019585,
|
| 2514 |
+
"learning_rate": 2e-05,
|
| 2515 |
+
"loss": 0.5863708019256592,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.4591967255052443,
|
| 2520 |
+
"grad_norm": 1.6688268457920834,
|
| 2521 |
+
"learning_rate": 2e-05,
|
| 2522 |
+
"loss": 0.591333532333374,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.4604758250191865,
|
| 2527 |
+
"grad_norm": 1.8755299547963225,
|
| 2528 |
+
"learning_rate": 2e-05,
|
| 2529 |
+
"loss": 0.5972795486450195,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.46175492453312866,
|
| 2534 |
+
"grad_norm": 1.991568535455219,
|
| 2535 |
+
"learning_rate": 2e-05,
|
| 2536 |
+
"loss": 0.5916287899017334,
|
| 2537 |
+
"step": 3610
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.46303402404707084,
|
| 2541 |
+
"grad_norm": 5.633434865013647,
|
| 2542 |
+
"learning_rate": 2e-05,
|
| 2543 |
+
"loss": 0.5936531066894531,
|
| 2544 |
+
"step": 3620
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.46431312356101306,
|
| 2548 |
+
"grad_norm": 2.7813966842720927,
|
| 2549 |
+
"learning_rate": 2e-05,
|
| 2550 |
+
"loss": 0.6172687530517578,
|
| 2551 |
+
"step": 3630
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.46559222307495524,
|
| 2555 |
+
"grad_norm": 1.1795683170771842,
|
| 2556 |
+
"learning_rate": 2e-05,
|
| 2557 |
+
"loss": 0.608796501159668,
|
| 2558 |
+
"step": 3640
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.4668713225888974,
|
| 2562 |
+
"grad_norm": 2.1898796237116938,
|
| 2563 |
+
"learning_rate": 2e-05,
|
| 2564 |
+
"loss": 0.5805678367614746,
|
| 2565 |
+
"step": 3650
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.4681504221028396,
|
| 2569 |
+
"grad_norm": 1.5027857186055826,
|
| 2570 |
+
"learning_rate": 2e-05,
|
| 2571 |
+
"loss": 0.5937762260437012,
|
| 2572 |
+
"step": 3660
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.4694295216167818,
|
| 2576 |
+
"grad_norm": 2.2064227907053877,
|
| 2577 |
+
"learning_rate": 2e-05,
|
| 2578 |
+
"loss": 0.5972755432128907,
|
| 2579 |
+
"step": 3670
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.470708621130724,
|
| 2583 |
+
"grad_norm": 3.307339050548398,
|
| 2584 |
+
"learning_rate": 2e-05,
|
| 2585 |
+
"loss": 0.6013876914978027,
|
| 2586 |
+
"step": 3680
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.47198772064466615,
|
| 2590 |
+
"grad_norm": 1.7538816105627706,
|
| 2591 |
+
"learning_rate": 2e-05,
|
| 2592 |
+
"loss": 0.5830701351165771,
|
| 2593 |
+
"step": 3690
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.4732668201586083,
|
| 2597 |
+
"grad_norm": 1.1541330208039735,
|
| 2598 |
+
"learning_rate": 2e-05,
|
| 2599 |
+
"loss": 0.5992613792419433,
|
| 2600 |
+
"step": 3700
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.4745459196725505,
|
| 2604 |
+
"grad_norm": 2.433370390394683,
|
| 2605 |
+
"learning_rate": 2e-05,
|
| 2606 |
+
"loss": 0.6018333435058594,
|
| 2607 |
+
"step": 3710
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.4758250191864927,
|
| 2611 |
+
"grad_norm": 8.05880675122322,
|
| 2612 |
+
"learning_rate": 2e-05,
|
| 2613 |
+
"loss": 0.5928830146789551,
|
| 2614 |
+
"step": 3720
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.4771041187004349,
|
| 2618 |
+
"grad_norm": 1.4689043424612738,
|
| 2619 |
+
"learning_rate": 2e-05,
|
| 2620 |
+
"loss": 0.5966341972351075,
|
| 2621 |
+
"step": 3730
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.47838321821437707,
|
| 2625 |
+
"grad_norm": 1.3562110418457431,
|
| 2626 |
+
"learning_rate": 2e-05,
|
| 2627 |
+
"loss": 0.5890920162200928,
|
| 2628 |
+
"step": 3740
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.47966231772831924,
|
| 2632 |
+
"grad_norm": 1.878872514143756,
|
| 2633 |
+
"learning_rate": 2e-05,
|
| 2634 |
+
"loss": 0.5998680114746093,
|
| 2635 |
+
"step": 3750
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 0.48094141724226147,
|
| 2639 |
+
"grad_norm": 1.1196343435761942,
|
| 2640 |
+
"learning_rate": 2e-05,
|
| 2641 |
+
"loss": 0.5796008110046387,
|
| 2642 |
+
"step": 3760
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.48222051675620364,
|
| 2646 |
+
"grad_norm": 1.1156802634346448,
|
| 2647 |
+
"learning_rate": 2e-05,
|
| 2648 |
+
"loss": 0.5972362518310547,
|
| 2649 |
+
"step": 3770
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 0.4834996162701458,
|
| 2653 |
+
"grad_norm": 2.1220659692240647,
|
| 2654 |
+
"learning_rate": 2e-05,
|
| 2655 |
+
"loss": 0.5801189422607422,
|
| 2656 |
+
"step": 3780
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.484778715784088,
|
| 2660 |
+
"grad_norm": 0.868961971462684,
|
| 2661 |
+
"learning_rate": 2e-05,
|
| 2662 |
+
"loss": 0.5691156387329102,
|
| 2663 |
+
"step": 3790
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.48605781529803016,
|
| 2667 |
+
"grad_norm": 0.9336123344027439,
|
| 2668 |
+
"learning_rate": 2e-05,
|
| 2669 |
+
"loss": 0.5900723934173584,
|
| 2670 |
+
"step": 3800
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.4873369148119724,
|
| 2674 |
+
"grad_norm": 0.9906753203156491,
|
| 2675 |
+
"learning_rate": 2e-05,
|
| 2676 |
+
"loss": 0.5864486694335938,
|
| 2677 |
+
"step": 3810
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 0.48861601432591456,
|
| 2681 |
+
"grad_norm": 1.0851288805526387,
|
| 2682 |
+
"learning_rate": 2e-05,
|
| 2683 |
+
"loss": 0.5898387908935547,
|
| 2684 |
+
"step": 3820
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.48989511383985673,
|
| 2688 |
+
"grad_norm": 3.5916418699724537,
|
| 2689 |
+
"learning_rate": 2e-05,
|
| 2690 |
+
"loss": 0.5957374572753906,
|
| 2691 |
+
"step": 3830
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 0.4911742133537989,
|
| 2695 |
+
"grad_norm": 0.7967894190370447,
|
| 2696 |
+
"learning_rate": 2e-05,
|
| 2697 |
+
"loss": 0.5781919002532959,
|
| 2698 |
+
"step": 3840
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 0.49245331286774113,
|
| 2702 |
+
"grad_norm": 1.2068188085171687,
|
| 2703 |
+
"learning_rate": 2e-05,
|
| 2704 |
+
"loss": 0.5645539283752441,
|
| 2705 |
+
"step": 3850
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.4937324123816833,
|
| 2709 |
+
"grad_norm": 0.8042966972737443,
|
| 2710 |
+
"learning_rate": 2e-05,
|
| 2711 |
+
"loss": 0.5781976222991944,
|
| 2712 |
+
"step": 3860
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 0.4950115118956255,
|
| 2716 |
+
"grad_norm": 2.9731362460553186,
|
| 2717 |
+
"learning_rate": 2e-05,
|
| 2718 |
+
"loss": 0.6096286296844482,
|
| 2719 |
+
"step": 3870
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 0.49629061140956765,
|
| 2723 |
+
"grad_norm": 1.8855978838078042,
|
| 2724 |
+
"learning_rate": 2e-05,
|
| 2725 |
+
"loss": 0.6072882652282715,
|
| 2726 |
+
"step": 3880
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.4975697109235098,
|
| 2730 |
+
"grad_norm": 2.3279484427764148,
|
| 2731 |
+
"learning_rate": 2e-05,
|
| 2732 |
+
"loss": 0.593841552734375,
|
| 2733 |
+
"step": 3890
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 0.49884881043745205,
|
| 2737 |
+
"grad_norm": 10.335316468198988,
|
| 2738 |
+
"learning_rate": 2e-05,
|
| 2739 |
+
"loss": 0.5867321968078614,
|
| 2740 |
+
"step": 3900
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 0.5001279099513942,
|
| 2744 |
+
"grad_norm": 1.151612922573664,
|
| 2745 |
+
"learning_rate": 2e-05,
|
| 2746 |
+
"loss": 0.5656989574432373,
|
| 2747 |
+
"step": 3910
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 0.5014070094653365,
|
| 2751 |
+
"grad_norm": 1.285187483922441,
|
| 2752 |
+
"learning_rate": 2e-05,
|
| 2753 |
+
"loss": 0.5880422115325927,
|
| 2754 |
+
"step": 3920
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 0.5026861089792786,
|
| 2758 |
+
"grad_norm": 2.7311402622059107,
|
| 2759 |
+
"learning_rate": 2e-05,
|
| 2760 |
+
"loss": 0.6022061824798584,
|
| 2761 |
+
"step": 3930
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 0.5039652084932208,
|
| 2765 |
+
"grad_norm": 1.0709412627660333,
|
| 2766 |
+
"learning_rate": 2e-05,
|
| 2767 |
+
"loss": 0.5973441123962402,
|
| 2768 |
+
"step": 3940
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 0.5052443080071629,
|
| 2772 |
+
"grad_norm": 14.145026244622716,
|
| 2773 |
+
"learning_rate": 2e-05,
|
| 2774 |
+
"loss": 0.6096267700195312,
|
| 2775 |
+
"step": 3950
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 0.5065234075211051,
|
| 2779 |
+
"grad_norm": 3.9916995890687144,
|
| 2780 |
+
"learning_rate": 2e-05,
|
| 2781 |
+
"loss": 0.5941912651062011,
|
| 2782 |
+
"step": 3960
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 0.5078025070350474,
|
| 2786 |
+
"grad_norm": 1.536315111968428,
|
| 2787 |
+
"learning_rate": 2e-05,
|
| 2788 |
+
"loss": 0.5798390865325928,
|
| 2789 |
+
"step": 3970
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 0.5090816065489895,
|
| 2793 |
+
"grad_norm": 3.038710043449597,
|
| 2794 |
+
"learning_rate": 2e-05,
|
| 2795 |
+
"loss": 0.5881974220275878,
|
| 2796 |
+
"step": 3980
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 0.5103607060629317,
|
| 2800 |
+
"grad_norm": 1.7437382040110692,
|
| 2801 |
+
"learning_rate": 2e-05,
|
| 2802 |
+
"loss": 0.597667646408081,
|
| 2803 |
+
"step": 3990
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 0.5116398055768738,
|
| 2807 |
+
"grad_norm": 2.2086375022016718,
|
| 2808 |
+
"learning_rate": 2e-05,
|
| 2809 |
+
"loss": 0.6139577388763428,
|
| 2810 |
+
"step": 4000
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 0.5129189050908161,
|
| 2814 |
+
"grad_norm": 1.1094758092405541,
|
| 2815 |
+
"learning_rate": 2e-05,
|
| 2816 |
+
"loss": 0.5899992942810058,
|
| 2817 |
+
"step": 4010
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 0.5141980046047583,
|
| 2821 |
+
"grad_norm": 3.191819345415452,
|
| 2822 |
+
"learning_rate": 2e-05,
|
| 2823 |
+
"loss": 0.5848158836364746,
|
| 2824 |
+
"step": 4020
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 0.5154771041187004,
|
| 2828 |
+
"grad_norm": 18.343464601651103,
|
| 2829 |
+
"learning_rate": 2e-05,
|
| 2830 |
+
"loss": 0.6113421440124511,
|
| 2831 |
+
"step": 4030
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 0.5167562036326426,
|
| 2835 |
+
"grad_norm": 8.151646256379413,
|
| 2836 |
+
"learning_rate": 2e-05,
|
| 2837 |
+
"loss": 0.6080557823181152,
|
| 2838 |
+
"step": 4040
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 0.5180353031465849,
|
| 2842 |
+
"grad_norm": 1.4743567384823875,
|
| 2843 |
+
"learning_rate": 2e-05,
|
| 2844 |
+
"loss": 0.5929433345794678,
|
| 2845 |
+
"step": 4050
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 0.519314402660527,
|
| 2849 |
+
"grad_norm": 1.6244297406185133,
|
| 2850 |
+
"learning_rate": 2e-05,
|
| 2851 |
+
"loss": 0.6228420257568359,
|
| 2852 |
+
"step": 4060
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 0.5205935021744692,
|
| 2856 |
+
"grad_norm": 1.8964666797465644,
|
| 2857 |
+
"learning_rate": 2e-05,
|
| 2858 |
+
"loss": 0.6942201137542725,
|
| 2859 |
+
"step": 4070
|
| 2860 |
+
}
|
| 2861 |
+
],
|
| 2862 |
+
"logging_steps": 10,
|
| 2863 |
+
"max_steps": 5935,
|
| 2864 |
+
"num_input_tokens_seen": 0,
|
| 2865 |
+
"num_train_epochs": 1,
|
| 2866 |
+
"save_steps": 2035,
|
| 2867 |
+
"stateful_callbacks": {
|
| 2868 |
+
"TrainerControl": {
|
| 2869 |
+
"args": {
|
| 2870 |
+
"should_epoch_stop": false,
|
| 2871 |
+
"should_evaluate": false,
|
| 2872 |
+
"should_log": false,
|
| 2873 |
+
"should_save": true,
|
| 2874 |
+
"should_training_stop": false
|
| 2875 |
+
},
|
| 2876 |
+
"attributes": {}
|
| 2877 |
+
}
|
| 2878 |
+
},
|
| 2879 |
+
"total_flos": 1927226987642880.0,
|
| 2880 |
+
"train_batch_size": 1,
|
| 2881 |
+
"trial_name": null,
|
| 2882 |
+
"trial_params": null
|
| 2883 |
+
}
|
400M/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f3a738f3e847c4a90be977945725b87956d0b276f6f6945167308e80d6c5092
|
| 3 |
+
size 7313
|