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 sha256:9f5a958c87c07351ea9f4049b01dd8c52ed0bd675f70f20cb7205c212b941c98
|
| 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": 0.13039438695415373,
|
| 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": 0.0003203793291256848,
|
| 14 |
+
"grad_norm": 2697.9695951365356,
|
| 15 |
+
"learning_rate": 2.910112359550562e-06,
|
| 16 |
+
"loss": 2.1080955505371093,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0006407586582513696,
|
| 21 |
+
"grad_norm": 150.28792394404246,
|
| 22 |
+
"learning_rate": 3.921348314606742e-06,
|
| 23 |
+
"loss": 1.051735782623291,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0009611379873770544,
|
| 28 |
+
"grad_norm": 11.728004906904669,
|
| 29 |
+
"learning_rate": 4.932584269662922e-06,
|
| 30 |
+
"loss": 0.8452852249145508,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0012815173165027392,
|
| 35 |
+
"grad_norm": 79.82045143025138,
|
| 36 |
+
"learning_rate": 5.943820224719102e-06,
|
| 37 |
+
"loss": 0.8374807357788085,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.001601896645628424,
|
| 42 |
+
"grad_norm": 15.487828868814473,
|
| 43 |
+
"learning_rate": 6.955056179775282e-06,
|
| 44 |
+
"loss": 0.8031275749206543,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0019222759747541088,
|
| 49 |
+
"grad_norm": 26.189373840069653,
|
| 50 |
+
"learning_rate": 7.966292134831462e-06,
|
| 51 |
+
"loss": 0.7619327545166016,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.002242655303879794,
|
| 56 |
+
"grad_norm": 24.510054444181353,
|
| 57 |
+
"learning_rate": 8.977528089887641e-06,
|
| 58 |
+
"loss": 0.788119888305664,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.0025630346330054784,
|
| 63 |
+
"grad_norm": 27.619677813912134,
|
| 64 |
+
"learning_rate": 9.988764044943822e-06,
|
| 65 |
+
"loss": 0.8074554443359375,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0028834139621311634,
|
| 70 |
+
"grad_norm": 53.27561070809039,
|
| 71 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 72 |
+
"loss": 0.7780509948730469,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.003203793291256848,
|
| 77 |
+
"grad_norm": 20.22293994140261,
|
| 78 |
+
"learning_rate": 1.2011235955056182e-05,
|
| 79 |
+
"loss": 0.7639347076416015,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.003524172620382533,
|
| 84 |
+
"grad_norm": 28.55291130912943,
|
| 85 |
+
"learning_rate": 1.3022471910112362e-05,
|
| 86 |
+
"loss": 0.7700469970703125,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0038445519495082176,
|
| 91 |
+
"grad_norm": 7.520305767933223,
|
| 92 |
+
"learning_rate": 1.403370786516854e-05,
|
| 93 |
+
"loss": 0.7953043937683105,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.004164931278633903,
|
| 98 |
+
"grad_norm": 96.08078842951305,
|
| 99 |
+
"learning_rate": 1.504494382022472e-05,
|
| 100 |
+
"loss": 0.8506025314331055,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.004485310607759588,
|
| 105 |
+
"grad_norm": 948.131373093307,
|
| 106 |
+
"learning_rate": 1.60561797752809e-05,
|
| 107 |
+
"loss": 0.8122509002685547,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.004805689936885272,
|
| 112 |
+
"grad_norm": 128.55037969805156,
|
| 113 |
+
"learning_rate": 1.706741573033708e-05,
|
| 114 |
+
"loss": 0.7938695907592773,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.005126069266010957,
|
| 119 |
+
"grad_norm": 25.93618158094234,
|
| 120 |
+
"learning_rate": 1.8078651685393256e-05,
|
| 121 |
+
"loss": 0.8438175201416016,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.005446448595136642,
|
| 126 |
+
"grad_norm": 9.528266971779534,
|
| 127 |
+
"learning_rate": 1.908988764044944e-05,
|
| 128 |
+
"loss": 0.9011789321899414,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.005766827924262327,
|
| 133 |
+
"grad_norm": 4.6353029582495315,
|
| 134 |
+
"learning_rate": 2e-05,
|
| 135 |
+
"loss": 0.8190475463867187,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.006087207253388012,
|
| 140 |
+
"grad_norm": 7.804600561994388,
|
| 141 |
+
"learning_rate": 2e-05,
|
| 142 |
+
"loss": 0.7879319667816163,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.006407586582513696,
|
| 147 |
+
"grad_norm": 8.759193053432988,
|
| 148 |
+
"learning_rate": 2e-05,
|
| 149 |
+
"loss": 0.8373113632202148,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.006727965911639381,
|
| 154 |
+
"grad_norm": 5.7645630188571495,
|
| 155 |
+
"learning_rate": 2e-05,
|
| 156 |
+
"loss": 0.8225456237792969,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.007048345240765066,
|
| 161 |
+
"grad_norm": 2.975476645382291,
|
| 162 |
+
"learning_rate": 2e-05,
|
| 163 |
+
"loss": 0.7898527145385742,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.007368724569890751,
|
| 168 |
+
"grad_norm": 3.5660713690110923,
|
| 169 |
+
"learning_rate": 2e-05,
|
| 170 |
+
"loss": 0.7848415374755859,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.007689103899016435,
|
| 175 |
+
"grad_norm": 3.776542520178389,
|
| 176 |
+
"learning_rate": 2e-05,
|
| 177 |
+
"loss": 0.7830493927001954,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.008009483228142121,
|
| 182 |
+
"grad_norm": 5.652886706438173,
|
| 183 |
+
"learning_rate": 2e-05,
|
| 184 |
+
"loss": 0.7420679092407226,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.008329862557267805,
|
| 189 |
+
"grad_norm": 2.7433625896377887,
|
| 190 |
+
"learning_rate": 2e-05,
|
| 191 |
+
"loss": 0.8062055587768555,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.00865024188639349,
|
| 196 |
+
"grad_norm": 4.993469336501579,
|
| 197 |
+
"learning_rate": 2e-05,
|
| 198 |
+
"loss": 0.7772369384765625,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.008970621215519175,
|
| 203 |
+
"grad_norm": 5.978072597304008,
|
| 204 |
+
"learning_rate": 2e-05,
|
| 205 |
+
"loss": 0.7567183494567871,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.00929100054464486,
|
| 210 |
+
"grad_norm": 20.0019916771175,
|
| 211 |
+
"learning_rate": 2e-05,
|
| 212 |
+
"loss": 0.7632350921630859,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.009611379873770544,
|
| 217 |
+
"grad_norm": 1813.803400221406,
|
| 218 |
+
"learning_rate": 2e-05,
|
| 219 |
+
"loss": 0.7820405006408692,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.00993175920289623,
|
| 224 |
+
"grad_norm": 15.257478921571794,
|
| 225 |
+
"learning_rate": 2e-05,
|
| 226 |
+
"loss": 0.8594757080078125,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.010252138532021914,
|
| 231 |
+
"grad_norm": 26.97548621262987,
|
| 232 |
+
"learning_rate": 2e-05,
|
| 233 |
+
"loss": 0.8610923767089844,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0105725178611476,
|
| 238 |
+
"grad_norm": 21.892790276243126,
|
| 239 |
+
"learning_rate": 2e-05,
|
| 240 |
+
"loss": 0.8686389923095703,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.010892897190273284,
|
| 245 |
+
"grad_norm": 16.06260184223094,
|
| 246 |
+
"learning_rate": 2e-05,
|
| 247 |
+
"loss": 0.8639827728271484,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.011213276519398968,
|
| 252 |
+
"grad_norm": 29.260614803206966,
|
| 253 |
+
"learning_rate": 2e-05,
|
| 254 |
+
"loss": 0.7904685497283935,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.011533655848524654,
|
| 259 |
+
"grad_norm": 16.44688311222318,
|
| 260 |
+
"learning_rate": 2e-05,
|
| 261 |
+
"loss": 0.8587644577026368,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.011854035177650338,
|
| 266 |
+
"grad_norm": 70.37237245987374,
|
| 267 |
+
"learning_rate": 2e-05,
|
| 268 |
+
"loss": 0.8287432670593262,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.012174414506776024,
|
| 273 |
+
"grad_norm": 7.352519778805397,
|
| 274 |
+
"learning_rate": 2e-05,
|
| 275 |
+
"loss": 0.8243340492248535,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.012494793835901708,
|
| 280 |
+
"grad_norm": 7.57180002114558,
|
| 281 |
+
"learning_rate": 2e-05,
|
| 282 |
+
"loss": 0.8488252639770508,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.012815173165027392,
|
| 287 |
+
"grad_norm": 4.028963614155682,
|
| 288 |
+
"learning_rate": 2e-05,
|
| 289 |
+
"loss": 0.800967788696289,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.013135552494153078,
|
| 294 |
+
"grad_norm": 7.604607607710578,
|
| 295 |
+
"learning_rate": 2e-05,
|
| 296 |
+
"loss": 0.8435401916503906,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.013455931823278762,
|
| 301 |
+
"grad_norm": 5.686576942826079,
|
| 302 |
+
"learning_rate": 2e-05,
|
| 303 |
+
"loss": 0.8086394309997559,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.013776311152404446,
|
| 308 |
+
"grad_norm": 29.16934098168494,
|
| 309 |
+
"learning_rate": 2e-05,
|
| 310 |
+
"loss": 0.8092558860778809,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.014096690481530132,
|
| 315 |
+
"grad_norm": 7.838527647928418,
|
| 316 |
+
"learning_rate": 2e-05,
|
| 317 |
+
"loss": 0.7873313903808594,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.014417069810655816,
|
| 322 |
+
"grad_norm": 4.360151373268049,
|
| 323 |
+
"learning_rate": 2e-05,
|
| 324 |
+
"loss": 0.8105262756347656,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.014737449139781502,
|
| 329 |
+
"grad_norm": 3.093592672180748,
|
| 330 |
+
"learning_rate": 2e-05,
|
| 331 |
+
"loss": 0.7954371452331543,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.015057828468907186,
|
| 336 |
+
"grad_norm": 2.8867780320165948,
|
| 337 |
+
"learning_rate": 2e-05,
|
| 338 |
+
"loss": 0.7867568969726563,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.01537820779803287,
|
| 343 |
+
"grad_norm": 2.5525945956595404,
|
| 344 |
+
"learning_rate": 2e-05,
|
| 345 |
+
"loss": 0.7605972290039062,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.015698587127158554,
|
| 350 |
+
"grad_norm": 4.350350397516954,
|
| 351 |
+
"learning_rate": 2e-05,
|
| 352 |
+
"loss": 0.7946616172790527,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.016018966456284242,
|
| 357 |
+
"grad_norm": 2.5601326184252624,
|
| 358 |
+
"learning_rate": 2e-05,
|
| 359 |
+
"loss": 0.7657347679138183,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.016339345785409926,
|
| 364 |
+
"grad_norm": 3.0411596315285943,
|
| 365 |
+
"learning_rate": 2e-05,
|
| 366 |
+
"loss": 0.7469792366027832,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.01665972511453561,
|
| 371 |
+
"grad_norm": 1.6677799044175043,
|
| 372 |
+
"learning_rate": 2e-05,
|
| 373 |
+
"loss": 0.737004280090332,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.016980104443661295,
|
| 378 |
+
"grad_norm": 4.326231014259539,
|
| 379 |
+
"learning_rate": 2e-05,
|
| 380 |
+
"loss": 0.7799790859222412,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.01730048377278698,
|
| 385 |
+
"grad_norm": 2.732975416289203,
|
| 386 |
+
"learning_rate": 2e-05,
|
| 387 |
+
"loss": 0.8164985656738282,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.017620863101912666,
|
| 392 |
+
"grad_norm": 2.1962459439491036,
|
| 393 |
+
"learning_rate": 2e-05,
|
| 394 |
+
"loss": 0.7743217468261718,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.01794124243103835,
|
| 399 |
+
"grad_norm": 4.329667095154606,
|
| 400 |
+
"learning_rate": 2e-05,
|
| 401 |
+
"loss": 0.8150390625,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.018261621760164035,
|
| 406 |
+
"grad_norm": 1.7505089770107007,
|
| 407 |
+
"learning_rate": 2e-05,
|
| 408 |
+
"loss": 0.8161712646484375,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.01858200108928972,
|
| 413 |
+
"grad_norm": 1.94080520422082,
|
| 414 |
+
"learning_rate": 2e-05,
|
| 415 |
+
"loss": 0.7340856075286866,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.018902380418415403,
|
| 420 |
+
"grad_norm": 3.767043072114596,
|
| 421 |
+
"learning_rate": 2e-05,
|
| 422 |
+
"loss": 0.7465179443359375,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.019222759747541087,
|
| 427 |
+
"grad_norm": 2.134739515696804,
|
| 428 |
+
"learning_rate": 2e-05,
|
| 429 |
+
"loss": 0.7511123657226563,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.019543139076666775,
|
| 434 |
+
"grad_norm": 2.6553715328436724,
|
| 435 |
+
"learning_rate": 2e-05,
|
| 436 |
+
"loss": 0.7474520683288575,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.01986351840579246,
|
| 441 |
+
"grad_norm": 23.554705629944248,
|
| 442 |
+
"learning_rate": 2e-05,
|
| 443 |
+
"loss": 0.7547467708587646,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.020183897734918143,
|
| 448 |
+
"grad_norm": 9.663715178820514,
|
| 449 |
+
"learning_rate": 2e-05,
|
| 450 |
+
"loss": 0.7591798782348633,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.020504277064043827,
|
| 455 |
+
"grad_norm": 2.857655614152204,
|
| 456 |
+
"learning_rate": 2e-05,
|
| 457 |
+
"loss": 0.7372615814208985,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.02082465639316951,
|
| 462 |
+
"grad_norm": 19.794443617215876,
|
| 463 |
+
"learning_rate": 2e-05,
|
| 464 |
+
"loss": 0.7473020076751709,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.0211450357222952,
|
| 469 |
+
"grad_norm": 3.4519708452708513,
|
| 470 |
+
"learning_rate": 2e-05,
|
| 471 |
+
"loss": 0.7669941902160644,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.021465415051420883,
|
| 476 |
+
"grad_norm": 4.439475537657204,
|
| 477 |
+
"learning_rate": 2e-05,
|
| 478 |
+
"loss": 0.7708707809448242,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.021785794380546567,
|
| 483 |
+
"grad_norm": 4.288236090232602,
|
| 484 |
+
"learning_rate": 2e-05,
|
| 485 |
+
"loss": 0.75105299949646,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.02210617370967225,
|
| 490 |
+
"grad_norm": 19.247483032059645,
|
| 491 |
+
"learning_rate": 2e-05,
|
| 492 |
+
"loss": 0.7456221580505371,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.022426553038797935,
|
| 497 |
+
"grad_norm": 6.277608344095431,
|
| 498 |
+
"learning_rate": 2e-05,
|
| 499 |
+
"loss": 0.772921371459961,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.022746932367923623,
|
| 504 |
+
"grad_norm": 3.5768536610883284,
|
| 505 |
+
"learning_rate": 2e-05,
|
| 506 |
+
"loss": 0.7416125297546386,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.023067311697049307,
|
| 511 |
+
"grad_norm": 15.05269132810213,
|
| 512 |
+
"learning_rate": 2e-05,
|
| 513 |
+
"loss": 0.7103744506835937,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.02338769102617499,
|
| 518 |
+
"grad_norm": 4.468376174246059,
|
| 519 |
+
"learning_rate": 2e-05,
|
| 520 |
+
"loss": 0.727785873413086,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.023708070355300676,
|
| 525 |
+
"grad_norm": 5.087071299517523,
|
| 526 |
+
"learning_rate": 2e-05,
|
| 527 |
+
"loss": 0.7700984477996826,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.02402844968442636,
|
| 532 |
+
"grad_norm": 11.288253253171709,
|
| 533 |
+
"learning_rate": 2e-05,
|
| 534 |
+
"loss": 0.7581868648529053,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.024348829013552047,
|
| 539 |
+
"grad_norm": 3.27468779155936,
|
| 540 |
+
"learning_rate": 2e-05,
|
| 541 |
+
"loss": 0.7562522888183594,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.02466920834267773,
|
| 546 |
+
"grad_norm": 7.595749246101231,
|
| 547 |
+
"learning_rate": 2e-05,
|
| 548 |
+
"loss": 0.7546162605285645,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.024989587671803416,
|
| 553 |
+
"grad_norm": 2.1701956042711728,
|
| 554 |
+
"learning_rate": 2e-05,
|
| 555 |
+
"loss": 0.7503818511962891,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.0253099670009291,
|
| 560 |
+
"grad_norm": 7.551313750149006,
|
| 561 |
+
"learning_rate": 2e-05,
|
| 562 |
+
"loss": 0.7194967269897461,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.025630346330054784,
|
| 567 |
+
"grad_norm": 3.0982279258014085,
|
| 568 |
+
"learning_rate": 2e-05,
|
| 569 |
+
"loss": 0.7183728218078613,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.025950725659180468,
|
| 574 |
+
"grad_norm": 2.2319456779746507,
|
| 575 |
+
"learning_rate": 2e-05,
|
| 576 |
+
"loss": 0.741051197052002,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.026271104988306156,
|
| 581 |
+
"grad_norm": 2.1272967612583904,
|
| 582 |
+
"learning_rate": 2e-05,
|
| 583 |
+
"loss": 0.7074191093444824,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.02659148431743184,
|
| 588 |
+
"grad_norm": 2.3756745796722125,
|
| 589 |
+
"learning_rate": 2e-05,
|
| 590 |
+
"loss": 0.7466435909271241,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.026911863646557524,
|
| 595 |
+
"grad_norm": 3.9008908753860925,
|
| 596 |
+
"learning_rate": 2e-05,
|
| 597 |
+
"loss": 0.7262076377868653,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.027232242975683208,
|
| 602 |
+
"grad_norm": 2.0561831766929806,
|
| 603 |
+
"learning_rate": 2e-05,
|
| 604 |
+
"loss": 0.7282081604003906,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.027552622304808892,
|
| 609 |
+
"grad_norm": 5.818198655017998,
|
| 610 |
+
"learning_rate": 2e-05,
|
| 611 |
+
"loss": 0.7566670894622802,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.02787300163393458,
|
| 616 |
+
"grad_norm": 2.1494490230827337,
|
| 617 |
+
"learning_rate": 2e-05,
|
| 618 |
+
"loss": 0.7288908958435059,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.028193380963060264,
|
| 623 |
+
"grad_norm": 1.409242496714709,
|
| 624 |
+
"learning_rate": 2e-05,
|
| 625 |
+
"loss": 0.7215320587158203,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.028513760292185948,
|
| 630 |
+
"grad_norm": 7.0920590147047164,
|
| 631 |
+
"learning_rate": 2e-05,
|
| 632 |
+
"loss": 0.7246700286865234,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.028834139621311632,
|
| 637 |
+
"grad_norm": 1.5684095793164903,
|
| 638 |
+
"learning_rate": 2e-05,
|
| 639 |
+
"loss": 0.7144109725952148,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.029154518950437316,
|
| 644 |
+
"grad_norm": 1.3316395400018082,
|
| 645 |
+
"learning_rate": 2e-05,
|
| 646 |
+
"loss": 0.7477982997894287,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.029474898279563004,
|
| 651 |
+
"grad_norm": 1.3602729023468434,
|
| 652 |
+
"learning_rate": 2e-05,
|
| 653 |
+
"loss": 0.7173037528991699,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.029795277608688688,
|
| 658 |
+
"grad_norm": 49.53643391306756,
|
| 659 |
+
"learning_rate": 2e-05,
|
| 660 |
+
"loss": 0.7678704738616944,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.030115656937814372,
|
| 665 |
+
"grad_norm": 4.340394359223178,
|
| 666 |
+
"learning_rate": 2e-05,
|
| 667 |
+
"loss": 0.7031218528747558,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.030436036266940057,
|
| 672 |
+
"grad_norm": 2.1203555620583736,
|
| 673 |
+
"learning_rate": 2e-05,
|
| 674 |
+
"loss": 0.7084536552429199,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.03075641559606574,
|
| 679 |
+
"grad_norm": 1.3371075573755178,
|
| 680 |
+
"learning_rate": 2e-05,
|
| 681 |
+
"loss": 0.7464202880859375,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.03107679492519143,
|
| 686 |
+
"grad_norm": 1.6473009838115498,
|
| 687 |
+
"learning_rate": 2e-05,
|
| 688 |
+
"loss": 0.7116725444793701,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.03139717425431711,
|
| 693 |
+
"grad_norm": 3.385121877804678,
|
| 694 |
+
"learning_rate": 2e-05,
|
| 695 |
+
"loss": 0.738076114654541,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.0317175535834428,
|
| 700 |
+
"grad_norm": 1.7147230699245448,
|
| 701 |
+
"learning_rate": 2e-05,
|
| 702 |
+
"loss": 0.7158563137054443,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.032037932912568484,
|
| 707 |
+
"grad_norm": 3.3445220753393468,
|
| 708 |
+
"learning_rate": 2e-05,
|
| 709 |
+
"loss": 0.8030566215515137,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.032358312241694165,
|
| 714 |
+
"grad_norm": 2.077502109986157,
|
| 715 |
+
"learning_rate": 2e-05,
|
| 716 |
+
"loss": 0.7472703456878662,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.03267869157081985,
|
| 721 |
+
"grad_norm": 17.329143506946714,
|
| 722 |
+
"learning_rate": 2e-05,
|
| 723 |
+
"loss": 0.7366230964660645,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.03299907089994553,
|
| 728 |
+
"grad_norm": 2.6268238025254735,
|
| 729 |
+
"learning_rate": 2e-05,
|
| 730 |
+
"loss": 0.7404189109802246,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.03331945022907122,
|
| 735 |
+
"grad_norm": 1.977866371730136,
|
| 736 |
+
"learning_rate": 2e-05,
|
| 737 |
+
"loss": 0.7236545085906982,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.03363982955819691,
|
| 742 |
+
"grad_norm": 1.498659413315871,
|
| 743 |
+
"learning_rate": 2e-05,
|
| 744 |
+
"loss": 0.7335420608520508,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.03396020888732259,
|
| 749 |
+
"grad_norm": 1.7062214656423396,
|
| 750 |
+
"learning_rate": 2e-05,
|
| 751 |
+
"loss": 0.7051788330078125,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.03428058821644828,
|
| 756 |
+
"grad_norm": 1.3397922877558162,
|
| 757 |
+
"learning_rate": 2e-05,
|
| 758 |
+
"loss": 0.7626142024993896,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.03460096754557396,
|
| 763 |
+
"grad_norm": 3.93050291862445,
|
| 764 |
+
"learning_rate": 2e-05,
|
| 765 |
+
"loss": 0.7514806270599366,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.034921346874699645,
|
| 770 |
+
"grad_norm": 3.398990137450838,
|
| 771 |
+
"learning_rate": 2e-05,
|
| 772 |
+
"loss": 0.7132882595062255,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.03524172620382533,
|
| 777 |
+
"grad_norm": 1.4600798546048035,
|
| 778 |
+
"learning_rate": 2e-05,
|
| 779 |
+
"loss": 0.716385793685913,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.03556210553295101,
|
| 784 |
+
"grad_norm": 1.3160301135986112,
|
| 785 |
+
"learning_rate": 2e-05,
|
| 786 |
+
"loss": 0.7191272735595703,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.0358824848620767,
|
| 791 |
+
"grad_norm": 1.2481731524679547,
|
| 792 |
+
"learning_rate": 2e-05,
|
| 793 |
+
"loss": 0.7158429145812988,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.03620286419120238,
|
| 798 |
+
"grad_norm": 2.8179317711592504,
|
| 799 |
+
"learning_rate": 2e-05,
|
| 800 |
+
"loss": 0.7038905143737793,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.03652324352032807,
|
| 805 |
+
"grad_norm": 1.8284267025580432,
|
| 806 |
+
"learning_rate": 2e-05,
|
| 807 |
+
"loss": 0.7065417289733886,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.03684362284945375,
|
| 812 |
+
"grad_norm": 2.212892645914091,
|
| 813 |
+
"learning_rate": 2e-05,
|
| 814 |
+
"loss": 0.7265023708343505,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.03716400217857944,
|
| 819 |
+
"grad_norm": 2.2597923858522404,
|
| 820 |
+
"learning_rate": 2e-05,
|
| 821 |
+
"loss": 0.7199366569519043,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.037484381507705125,
|
| 826 |
+
"grad_norm": 2.0160695548330936,
|
| 827 |
+
"learning_rate": 2e-05,
|
| 828 |
+
"loss": 0.7658578395843506,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.037804760836830806,
|
| 833 |
+
"grad_norm": 2.5572488896871843,
|
| 834 |
+
"learning_rate": 2e-05,
|
| 835 |
+
"loss": 0.7220034599304199,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.03812514016595649,
|
| 840 |
+
"grad_norm": 2.459122478774182,
|
| 841 |
+
"learning_rate": 2e-05,
|
| 842 |
+
"loss": 0.6983201026916503,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.038445519495082174,
|
| 847 |
+
"grad_norm": 5.318818273924634,
|
| 848 |
+
"learning_rate": 2e-05,
|
| 849 |
+
"loss": 0.7152614116668701,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.03876589882420786,
|
| 854 |
+
"grad_norm": 4.348661337562832,
|
| 855 |
+
"learning_rate": 2e-05,
|
| 856 |
+
"loss": 0.7531236171722412,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.03908627815333355,
|
| 861 |
+
"grad_norm": 3.5073096880649666,
|
| 862 |
+
"learning_rate": 2e-05,
|
| 863 |
+
"loss": 0.7070003509521484,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.03940665748245923,
|
| 868 |
+
"grad_norm": 6.250765626260255,
|
| 869 |
+
"learning_rate": 2e-05,
|
| 870 |
+
"loss": 0.7422648429870605,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.03972703681158492,
|
| 875 |
+
"grad_norm": 2.1134272476858693,
|
| 876 |
+
"learning_rate": 2e-05,
|
| 877 |
+
"loss": 0.7167387962341308,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.0400474161407106,
|
| 882 |
+
"grad_norm": 2.728613864385018,
|
| 883 |
+
"learning_rate": 2e-05,
|
| 884 |
+
"loss": 0.7613626480102539,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.040367795469836286,
|
| 889 |
+
"grad_norm": 2.8289243765180094,
|
| 890 |
+
"learning_rate": 2e-05,
|
| 891 |
+
"loss": 0.7406154155731202,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.040688174798961974,
|
| 896 |
+
"grad_norm": 1.460978022101602,
|
| 897 |
+
"learning_rate": 2e-05,
|
| 898 |
+
"loss": 0.7171460628509522,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.041008554128087654,
|
| 903 |
+
"grad_norm": 11.187948658326427,
|
| 904 |
+
"learning_rate": 2e-05,
|
| 905 |
+
"loss": 0.7119218349456787,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.04132893345721334,
|
| 910 |
+
"grad_norm": 1.5366780655953838,
|
| 911 |
+
"learning_rate": 2e-05,
|
| 912 |
+
"loss": 0.7250003337860107,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.04164931278633902,
|
| 917 |
+
"grad_norm": 2.6451401695263392,
|
| 918 |
+
"learning_rate": 2e-05,
|
| 919 |
+
"loss": 0.7535340785980225,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.04196969211546471,
|
| 924 |
+
"grad_norm": 5.247231775991024,
|
| 925 |
+
"learning_rate": 2e-05,
|
| 926 |
+
"loss": 0.7120182991027832,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.0422900714445904,
|
| 931 |
+
"grad_norm": 1.265414487137248,
|
| 932 |
+
"learning_rate": 2e-05,
|
| 933 |
+
"loss": 0.7114714622497559,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.04261045077371608,
|
| 938 |
+
"grad_norm": 32.52328999465322,
|
| 939 |
+
"learning_rate": 2e-05,
|
| 940 |
+
"loss": 0.6845267295837403,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.042930830102841766,
|
| 945 |
+
"grad_norm": 6.1768247607509394,
|
| 946 |
+
"learning_rate": 2e-05,
|
| 947 |
+
"loss": 0.702857780456543,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.04325120943196745,
|
| 952 |
+
"grad_norm": 7.1933410968196725,
|
| 953 |
+
"learning_rate": 2e-05,
|
| 954 |
+
"loss": 0.72010817527771,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.043571588761093134,
|
| 959 |
+
"grad_norm": 1.1755562476246728,
|
| 960 |
+
"learning_rate": 2e-05,
|
| 961 |
+
"loss": 0.6978877067565918,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.04389196809021882,
|
| 966 |
+
"grad_norm": 2.8540830213502844,
|
| 967 |
+
"learning_rate": 2e-05,
|
| 968 |
+
"loss": 0.7204657554626465,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.0442123474193445,
|
| 973 |
+
"grad_norm": 1.1871852286932287,
|
| 974 |
+
"learning_rate": 2e-05,
|
| 975 |
+
"loss": 0.7182430267333985,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.04453272674847019,
|
| 980 |
+
"grad_norm": 1.0566209393736636,
|
| 981 |
+
"learning_rate": 2e-05,
|
| 982 |
+
"loss": 0.7123696804046631,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.04485310607759587,
|
| 987 |
+
"grad_norm": 4.237024261846737,
|
| 988 |
+
"learning_rate": 2e-05,
|
| 989 |
+
"loss": 0.6963154792785644,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.04517348540672156,
|
| 994 |
+
"grad_norm": 2.008976090493564,
|
| 995 |
+
"learning_rate": 2e-05,
|
| 996 |
+
"loss": 0.6849660396575927,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.045493864735847246,
|
| 1001 |
+
"grad_norm": 1.3493113476458192,
|
| 1002 |
+
"learning_rate": 2e-05,
|
| 1003 |
+
"loss": 0.7087863922119141,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.04581424406497293,
|
| 1008 |
+
"grad_norm": 1.0453265260241618,
|
| 1009 |
+
"learning_rate": 2e-05,
|
| 1010 |
+
"loss": 0.6820076942443848,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.046134623394098614,
|
| 1015 |
+
"grad_norm": 1.367769743280532,
|
| 1016 |
+
"learning_rate": 2e-05,
|
| 1017 |
+
"loss": 0.7149313926696778,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.046455002723224295,
|
| 1022 |
+
"grad_norm": 1.5424097504581287,
|
| 1023 |
+
"learning_rate": 2e-05,
|
| 1024 |
+
"loss": 0.7253753185272217,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.04677538205234998,
|
| 1029 |
+
"grad_norm": 2.3372729347553025,
|
| 1030 |
+
"learning_rate": 2e-05,
|
| 1031 |
+
"loss": 0.7671535015106201,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.04709576138147567,
|
| 1036 |
+
"grad_norm": 1.8742828809078558,
|
| 1037 |
+
"learning_rate": 2e-05,
|
| 1038 |
+
"loss": 0.7182416915893555,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.04741614071060135,
|
| 1043 |
+
"grad_norm": 6.87172572472952,
|
| 1044 |
+
"learning_rate": 2e-05,
|
| 1045 |
+
"loss": 0.7154425621032715,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.04773652003972704,
|
| 1050 |
+
"grad_norm": 2.4729840403642545,
|
| 1051 |
+
"learning_rate": 2e-05,
|
| 1052 |
+
"loss": 0.7049578666687012,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.04805689936885272,
|
| 1057 |
+
"grad_norm": 1.3804743410158802,
|
| 1058 |
+
"learning_rate": 2e-05,
|
| 1059 |
+
"loss": 0.7238261222839355,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.04837727869797841,
|
| 1064 |
+
"grad_norm": 1.2556616325656202,
|
| 1065 |
+
"learning_rate": 2e-05,
|
| 1066 |
+
"loss": 0.7582721710205078,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.048697658027104095,
|
| 1071 |
+
"grad_norm": 1.4068662819395166,
|
| 1072 |
+
"learning_rate": 2e-05,
|
| 1073 |
+
"loss": 0.6943454742431641,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.049018037356229775,
|
| 1078 |
+
"grad_norm": 1.5191574570193485,
|
| 1079 |
+
"learning_rate": 2e-05,
|
| 1080 |
+
"loss": 0.7089879512786865,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.04933841668535546,
|
| 1085 |
+
"grad_norm": 1.1100647663489502,
|
| 1086 |
+
"learning_rate": 2e-05,
|
| 1087 |
+
"loss": 0.6814855098724365,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.049658796014481144,
|
| 1092 |
+
"grad_norm": 1.2368089337986163,
|
| 1093 |
+
"learning_rate": 2e-05,
|
| 1094 |
+
"loss": 0.7276053428649902,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.04997917534360683,
|
| 1099 |
+
"grad_norm": 1.596887157213148,
|
| 1100 |
+
"learning_rate": 2e-05,
|
| 1101 |
+
"loss": 0.7387615203857422,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.05029955467273251,
|
| 1106 |
+
"grad_norm": 1.3857783026385584,
|
| 1107 |
+
"learning_rate": 2e-05,
|
| 1108 |
+
"loss": 0.6814603805541992,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.0506199340018582,
|
| 1113 |
+
"grad_norm": 1.5717792680161595,
|
| 1114 |
+
"learning_rate": 2e-05,
|
| 1115 |
+
"loss": 0.6634132385253906,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.05094031333098389,
|
| 1120 |
+
"grad_norm": 5.076622025769138,
|
| 1121 |
+
"learning_rate": 2e-05,
|
| 1122 |
+
"loss": 0.7170121669769287,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.05126069266010957,
|
| 1127 |
+
"grad_norm": 3.075120261257635,
|
| 1128 |
+
"learning_rate": 2e-05,
|
| 1129 |
+
"loss": 0.6901824951171875,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.051581071989235255,
|
| 1134 |
+
"grad_norm": 3.1443073556392838,
|
| 1135 |
+
"learning_rate": 2e-05,
|
| 1136 |
+
"loss": 0.7126097679138184,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.051901451318360936,
|
| 1141 |
+
"grad_norm": 1.3279065854265553,
|
| 1142 |
+
"learning_rate": 2e-05,
|
| 1143 |
+
"loss": 0.696572208404541,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.052221830647486624,
|
| 1148 |
+
"grad_norm": 2.2922495579972213,
|
| 1149 |
+
"learning_rate": 2e-05,
|
| 1150 |
+
"loss": 0.7236587047576905,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.05254220997661231,
|
| 1155 |
+
"grad_norm": 1.853599288786833,
|
| 1156 |
+
"learning_rate": 2e-05,
|
| 1157 |
+
"loss": 0.6959040641784668,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.05286258930573799,
|
| 1162 |
+
"grad_norm": 1.6356938422959257,
|
| 1163 |
+
"learning_rate": 2e-05,
|
| 1164 |
+
"loss": 0.6883045196533203,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.05318296863486368,
|
| 1169 |
+
"grad_norm": 1.4258158618988481,
|
| 1170 |
+
"learning_rate": 2e-05,
|
| 1171 |
+
"loss": 0.7095949172973632,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.05350334796398936,
|
| 1176 |
+
"grad_norm": 1.3818626048983265,
|
| 1177 |
+
"learning_rate": 2e-05,
|
| 1178 |
+
"loss": 0.7279641628265381,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.05382372729311505,
|
| 1183 |
+
"grad_norm": 1.3148725391903133,
|
| 1184 |
+
"learning_rate": 2e-05,
|
| 1185 |
+
"loss": 0.7071706295013428,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.054144106622240736,
|
| 1190 |
+
"grad_norm": 1.3205843385485543,
|
| 1191 |
+
"learning_rate": 2e-05,
|
| 1192 |
+
"loss": 0.7102875709533691,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.054464485951366416,
|
| 1197 |
+
"grad_norm": 1.0245118537391376,
|
| 1198 |
+
"learning_rate": 2e-05,
|
| 1199 |
+
"loss": 0.7355536460876465,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.054784865280492104,
|
| 1204 |
+
"grad_norm": 0.9830319826014532,
|
| 1205 |
+
"learning_rate": 2e-05,
|
| 1206 |
+
"loss": 0.711072301864624,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.055105244609617784,
|
| 1211 |
+
"grad_norm": 2.2427073223135,
|
| 1212 |
+
"learning_rate": 2e-05,
|
| 1213 |
+
"loss": 0.7257272720336914,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.05542562393874347,
|
| 1218 |
+
"grad_norm": 1.7432652030049554,
|
| 1219 |
+
"learning_rate": 2e-05,
|
| 1220 |
+
"loss": 0.7266391754150391,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.05574600326786916,
|
| 1225 |
+
"grad_norm": 7.3855085380936965,
|
| 1226 |
+
"learning_rate": 2e-05,
|
| 1227 |
+
"loss": 0.7229394435882568,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.05606638259699484,
|
| 1232 |
+
"grad_norm": 2.1417039515536542,
|
| 1233 |
+
"learning_rate": 2e-05,
|
| 1234 |
+
"loss": 0.6833599090576172,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.05638676192612053,
|
| 1239 |
+
"grad_norm": 2.4468519837946165,
|
| 1240 |
+
"learning_rate": 2e-05,
|
| 1241 |
+
"loss": 0.7663132667541503,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.05670714125524621,
|
| 1246 |
+
"grad_norm": 3.0137174378050293,
|
| 1247 |
+
"learning_rate": 2e-05,
|
| 1248 |
+
"loss": 0.7224421977996827,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.057027520584371896,
|
| 1253 |
+
"grad_norm": 2.2523438896707706,
|
| 1254 |
+
"learning_rate": 2e-05,
|
| 1255 |
+
"loss": 0.7207423210144043,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.057347899913497584,
|
| 1260 |
+
"grad_norm": 2.919705865322085,
|
| 1261 |
+
"learning_rate": 2e-05,
|
| 1262 |
+
"loss": 0.6854898452758789,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.057668279242623265,
|
| 1267 |
+
"grad_norm": 3.984396788478948,
|
| 1268 |
+
"learning_rate": 2e-05,
|
| 1269 |
+
"loss": 0.7234033107757568,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.05798865857174895,
|
| 1274 |
+
"grad_norm": 6.67851094941151,
|
| 1275 |
+
"learning_rate": 2e-05,
|
| 1276 |
+
"loss": 0.7022351264953614,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.05830903790087463,
|
| 1281 |
+
"grad_norm": 1.6866126836675677,
|
| 1282 |
+
"learning_rate": 2e-05,
|
| 1283 |
+
"loss": 0.7415154457092286,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.05862941723000032,
|
| 1288 |
+
"grad_norm": 3.218820869797509,
|
| 1289 |
+
"learning_rate": 2e-05,
|
| 1290 |
+
"loss": 0.7005694389343262,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.05894979655912601,
|
| 1295 |
+
"grad_norm": 3.0737908739781106,
|
| 1296 |
+
"learning_rate": 2e-05,
|
| 1297 |
+
"loss": 0.6981851100921631,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.05927017588825169,
|
| 1302 |
+
"grad_norm": 2.2327032848527795,
|
| 1303 |
+
"learning_rate": 2e-05,
|
| 1304 |
+
"loss": 0.7604454040527344,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.059590555217377376,
|
| 1309 |
+
"grad_norm": 2.8761311961580973,
|
| 1310 |
+
"learning_rate": 2e-05,
|
| 1311 |
+
"loss": 0.6888795852661133,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.05991093454650306,
|
| 1316 |
+
"grad_norm": 2.2317201691382533,
|
| 1317 |
+
"learning_rate": 2e-05,
|
| 1318 |
+
"loss": 0.708838415145874,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.060231313875628745,
|
| 1323 |
+
"grad_norm": 2.0355186191234806,
|
| 1324 |
+
"learning_rate": 2e-05,
|
| 1325 |
+
"loss": 0.7040927886962891,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.06055169320475443,
|
| 1330 |
+
"grad_norm": 2.014047361446366,
|
| 1331 |
+
"learning_rate": 2e-05,
|
| 1332 |
+
"loss": 0.7197917461395263,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.06087207253388011,
|
| 1337 |
+
"grad_norm": 1.977104681851146,
|
| 1338 |
+
"learning_rate": 2e-05,
|
| 1339 |
+
"loss": 0.6776611328125,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.0611924518630058,
|
| 1344 |
+
"grad_norm": 13.901447338457938,
|
| 1345 |
+
"learning_rate": 2e-05,
|
| 1346 |
+
"loss": 0.6899932861328125,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.06151283119213148,
|
| 1351 |
+
"grad_norm": 2.1123993495266435,
|
| 1352 |
+
"learning_rate": 2e-05,
|
| 1353 |
+
"loss": 0.6683700084686279,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.06183321052125717,
|
| 1358 |
+
"grad_norm": 2.3306798040564676,
|
| 1359 |
+
"learning_rate": 2e-05,
|
| 1360 |
+
"loss": 0.691620922088623,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.06215358985038286,
|
| 1365 |
+
"grad_norm": 13.395393364218384,
|
| 1366 |
+
"learning_rate": 2e-05,
|
| 1367 |
+
"loss": 0.7234976768493653,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.06247396917950854,
|
| 1372 |
+
"grad_norm": 7.087449907589583,
|
| 1373 |
+
"learning_rate": 2e-05,
|
| 1374 |
+
"loss": 0.7050728797912598,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.06279434850863422,
|
| 1379 |
+
"grad_norm": 2.912496650205425,
|
| 1380 |
+
"learning_rate": 2e-05,
|
| 1381 |
+
"loss": 0.6966752529144287,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.06311472783775991,
|
| 1386 |
+
"grad_norm": 8.623415554665941,
|
| 1387 |
+
"learning_rate": 2e-05,
|
| 1388 |
+
"loss": 0.67595534324646,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.0634351071668856,
|
| 1393 |
+
"grad_norm": 3.239938193275378,
|
| 1394 |
+
"learning_rate": 2e-05,
|
| 1395 |
+
"loss": 0.6972695350646972,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.06375548649601127,
|
| 1400 |
+
"grad_norm": 4.165992500546785,
|
| 1401 |
+
"learning_rate": 2e-05,
|
| 1402 |
+
"loss": 0.7464128017425538,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.06407586582513697,
|
| 1407 |
+
"grad_norm": 3.308146075178026,
|
| 1408 |
+
"learning_rate": 2e-05,
|
| 1409 |
+
"loss": 0.7509006023406982,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.06439624515426265,
|
| 1414 |
+
"grad_norm": 3.4969526861271802,
|
| 1415 |
+
"learning_rate": 2e-05,
|
| 1416 |
+
"loss": 0.7300586700439453,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.06471662448338833,
|
| 1421 |
+
"grad_norm": 4.743378784447764,
|
| 1422 |
+
"learning_rate": 2e-05,
|
| 1423 |
+
"loss": 0.6886860847473144,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.06503700381251401,
|
| 1428 |
+
"grad_norm": 1.9438372463888733,
|
| 1429 |
+
"learning_rate": 2e-05,
|
| 1430 |
+
"loss": 0.705808448791504,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.0653573831416397,
|
| 1435 |
+
"grad_norm": 11.001500428867326,
|
| 1436 |
+
"learning_rate": 2e-05,
|
| 1437 |
+
"loss": 0.698797607421875,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.06567776247076539,
|
| 1442 |
+
"grad_norm": 1.7716709742858978,
|
| 1443 |
+
"learning_rate": 2e-05,
|
| 1444 |
+
"loss": 0.7007397651672364,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.06599814179989107,
|
| 1449 |
+
"grad_norm": 2.0422284531425343,
|
| 1450 |
+
"learning_rate": 2e-05,
|
| 1451 |
+
"loss": 0.7145570755004883,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.06631852112901676,
|
| 1456 |
+
"grad_norm": 5.491456733860842,
|
| 1457 |
+
"learning_rate": 2e-05,
|
| 1458 |
+
"loss": 0.669373893737793,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.06663890045814244,
|
| 1463 |
+
"grad_norm": 17.321502365857935,
|
| 1464 |
+
"learning_rate": 2e-05,
|
| 1465 |
+
"loss": 0.7140804290771484,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.06695927978726812,
|
| 1470 |
+
"grad_norm": 1.6192456244078586,
|
| 1471 |
+
"learning_rate": 2e-05,
|
| 1472 |
+
"loss": 0.7083226680755615,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.06727965911639382,
|
| 1477 |
+
"grad_norm": 1.70128491897758,
|
| 1478 |
+
"learning_rate": 2e-05,
|
| 1479 |
+
"loss": 0.696595573425293,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.0676000384455195,
|
| 1484 |
+
"grad_norm": 1.6411121971094411,
|
| 1485 |
+
"learning_rate": 2e-05,
|
| 1486 |
+
"loss": 0.6904148101806641,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.06792041777464518,
|
| 1491 |
+
"grad_norm": 1.2162810019964114,
|
| 1492 |
+
"learning_rate": 2e-05,
|
| 1493 |
+
"loss": 0.7027925014495849,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.06824079710377086,
|
| 1498 |
+
"grad_norm": 4.319041953230064,
|
| 1499 |
+
"learning_rate": 2e-05,
|
| 1500 |
+
"loss": 0.8154801368713379,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.06856117643289655,
|
| 1505 |
+
"grad_norm": 6.8100351134727335,
|
| 1506 |
+
"learning_rate": 2e-05,
|
| 1507 |
+
"loss": 0.7859745025634766,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.06888155576202223,
|
| 1512 |
+
"grad_norm": 41.32429680397114,
|
| 1513 |
+
"learning_rate": 2e-05,
|
| 1514 |
+
"loss": 0.8241888046264648,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.06920193509114791,
|
| 1519 |
+
"grad_norm": 7.447130807285079,
|
| 1520 |
+
"learning_rate": 2e-05,
|
| 1521 |
+
"loss": 0.738440465927124,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.06952231442027361,
|
| 1526 |
+
"grad_norm": 2.4311779687133095,
|
| 1527 |
+
"learning_rate": 2e-05,
|
| 1528 |
+
"loss": 0.7003934860229493,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.06984269374939929,
|
| 1533 |
+
"grad_norm": 1.888452194069137,
|
| 1534 |
+
"learning_rate": 2e-05,
|
| 1535 |
+
"loss": 0.6981853485107422,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.07016307307852497,
|
| 1540 |
+
"grad_norm": 1.3884978841961744,
|
| 1541 |
+
"learning_rate": 2e-05,
|
| 1542 |
+
"loss": 0.6570146560668946,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.07048345240765067,
|
| 1547 |
+
"grad_norm": 1.90260673711977,
|
| 1548 |
+
"learning_rate": 2e-05,
|
| 1549 |
+
"loss": 0.6660157680511475,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.07080383173677635,
|
| 1554 |
+
"grad_norm": 2.313870497589669,
|
| 1555 |
+
"learning_rate": 2e-05,
|
| 1556 |
+
"loss": 0.7752554893493653,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.07112421106590203,
|
| 1561 |
+
"grad_norm": 17.868103729169547,
|
| 1562 |
+
"learning_rate": 2e-05,
|
| 1563 |
+
"loss": 0.745037841796875,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.07144459039502771,
|
| 1568 |
+
"grad_norm": 2.3214674515548737,
|
| 1569 |
+
"learning_rate": 2e-05,
|
| 1570 |
+
"loss": 0.7210993766784668,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.0717649697241534,
|
| 1575 |
+
"grad_norm": 3.1963888308689614,
|
| 1576 |
+
"learning_rate": 2e-05,
|
| 1577 |
+
"loss": 0.6749308586120606,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.07208534905327908,
|
| 1582 |
+
"grad_norm": 3.3505276313125614,
|
| 1583 |
+
"learning_rate": 2e-05,
|
| 1584 |
+
"loss": 0.6825098991394043,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.07240572838240476,
|
| 1589 |
+
"grad_norm": 1.7104131993752278,
|
| 1590 |
+
"learning_rate": 2e-05,
|
| 1591 |
+
"loss": 0.7080960750579834,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.07272610771153046,
|
| 1596 |
+
"grad_norm": 1.4805583634540915,
|
| 1597 |
+
"learning_rate": 2e-05,
|
| 1598 |
+
"loss": 0.6782462120056152,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.07304648704065614,
|
| 1603 |
+
"grad_norm": 1.3197929147416767,
|
| 1604 |
+
"learning_rate": 2e-05,
|
| 1605 |
+
"loss": 0.6485519409179688,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.07336686636978182,
|
| 1610 |
+
"grad_norm": 1.221515239216845,
|
| 1611 |
+
"learning_rate": 2e-05,
|
| 1612 |
+
"loss": 0.6853775024414063,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.0736872456989075,
|
| 1617 |
+
"grad_norm": 30.204078348726764,
|
| 1618 |
+
"learning_rate": 2e-05,
|
| 1619 |
+
"loss": 0.6834779262542725,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.0740076250280332,
|
| 1624 |
+
"grad_norm": 1.4517575091324884,
|
| 1625 |
+
"learning_rate": 2e-05,
|
| 1626 |
+
"loss": 0.6975203514099121,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.07432800435715887,
|
| 1631 |
+
"grad_norm": 1.3422521684581996,
|
| 1632 |
+
"learning_rate": 2e-05,
|
| 1633 |
+
"loss": 0.759161901473999,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.07464838368628456,
|
| 1638 |
+
"grad_norm": 2.3245979856998127,
|
| 1639 |
+
"learning_rate": 2e-05,
|
| 1640 |
+
"loss": 0.7400575637817383,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.07496876301541025,
|
| 1645 |
+
"grad_norm": 2.4522055207504607,
|
| 1646 |
+
"learning_rate": 2e-05,
|
| 1647 |
+
"loss": 0.6582594871520996,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.07528914234453593,
|
| 1652 |
+
"grad_norm": 3.9583378563876668,
|
| 1653 |
+
"learning_rate": 2e-05,
|
| 1654 |
+
"loss": 0.7118147850036621,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.07560952167366161,
|
| 1659 |
+
"grad_norm": 1.3241532087375179,
|
| 1660 |
+
"learning_rate": 2e-05,
|
| 1661 |
+
"loss": 0.750539493560791,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.0759299010027873,
|
| 1666 |
+
"grad_norm": 1.6871548656996915,
|
| 1667 |
+
"learning_rate": 2e-05,
|
| 1668 |
+
"loss": 0.7293619155883789,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.07625028033191299,
|
| 1673 |
+
"grad_norm": 1.9502293710817282,
|
| 1674 |
+
"learning_rate": 2e-05,
|
| 1675 |
+
"loss": 0.7126898765563965,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.07657065966103867,
|
| 1680 |
+
"grad_norm": 2.5257667368767187,
|
| 1681 |
+
"learning_rate": 2e-05,
|
| 1682 |
+
"loss": 0.6581045150756836,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.07689103899016435,
|
| 1687 |
+
"grad_norm": 2.4638908736573693,
|
| 1688 |
+
"learning_rate": 2e-05,
|
| 1689 |
+
"loss": 0.7101143836975098,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.07721141831929004,
|
| 1694 |
+
"grad_norm": 2.047332045907796,
|
| 1695 |
+
"learning_rate": 2e-05,
|
| 1696 |
+
"loss": 0.6816272735595703,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.07753179764841572,
|
| 1701 |
+
"grad_norm": 1.0750576692726481,
|
| 1702 |
+
"learning_rate": 2e-05,
|
| 1703 |
+
"loss": 0.7101527214050293,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.0778521769775414,
|
| 1708 |
+
"grad_norm": 1.3667366477929122,
|
| 1709 |
+
"learning_rate": 2e-05,
|
| 1710 |
+
"loss": 0.680915641784668,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.0781725563066671,
|
| 1715 |
+
"grad_norm": 1.9499465487529435,
|
| 1716 |
+
"learning_rate": 2e-05,
|
| 1717 |
+
"loss": 0.67269287109375,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.07849293563579278,
|
| 1722 |
+
"grad_norm": 9.520772501016044,
|
| 1723 |
+
"learning_rate": 2e-05,
|
| 1724 |
+
"loss": 0.7161269187927246,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.07881331496491846,
|
| 1729 |
+
"grad_norm": 1.2730224537721924,
|
| 1730 |
+
"learning_rate": 2e-05,
|
| 1731 |
+
"loss": 0.6880199432373046,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.07913369429404415,
|
| 1736 |
+
"grad_norm": 4.122823755612335,
|
| 1737 |
+
"learning_rate": 2e-05,
|
| 1738 |
+
"loss": 0.7019095420837402,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.07945407362316984,
|
| 1743 |
+
"grad_norm": 1.2350902700761086,
|
| 1744 |
+
"learning_rate": 2e-05,
|
| 1745 |
+
"loss": 0.7402851104736328,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.07977445295229552,
|
| 1750 |
+
"grad_norm": 1.6557513951573315,
|
| 1751 |
+
"learning_rate": 2e-05,
|
| 1752 |
+
"loss": 0.6856629371643066,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.0800948322814212,
|
| 1757 |
+
"grad_norm": 2.76250457287939,
|
| 1758 |
+
"learning_rate": 2e-05,
|
| 1759 |
+
"loss": 0.7310905456542969,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.08041521161054689,
|
| 1764 |
+
"grad_norm": 10.355727739658676,
|
| 1765 |
+
"learning_rate": 2e-05,
|
| 1766 |
+
"loss": 0.769098424911499,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.08073559093967257,
|
| 1771 |
+
"grad_norm": 1.505013422609845,
|
| 1772 |
+
"learning_rate": 2e-05,
|
| 1773 |
+
"loss": 0.699889850616455,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.08105597026879825,
|
| 1778 |
+
"grad_norm": 1.1570475749103184,
|
| 1779 |
+
"learning_rate": 2e-05,
|
| 1780 |
+
"loss": 0.6950805187225342,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.08137634959792395,
|
| 1785 |
+
"grad_norm": 2.1024924495057995,
|
| 1786 |
+
"learning_rate": 2e-05,
|
| 1787 |
+
"loss": 0.701189374923706,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.08169672892704963,
|
| 1792 |
+
"grad_norm": 1.7444425666233216,
|
| 1793 |
+
"learning_rate": 2e-05,
|
| 1794 |
+
"loss": 0.7059882164001465,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.08201710825617531,
|
| 1799 |
+
"grad_norm": 1.0322806923298704,
|
| 1800 |
+
"learning_rate": 2e-05,
|
| 1801 |
+
"loss": 0.8018281936645508,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.082337487585301,
|
| 1806 |
+
"grad_norm": 2.6526418970250307,
|
| 1807 |
+
"learning_rate": 2e-05,
|
| 1808 |
+
"loss": 0.6831948757171631,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.08265786691442668,
|
| 1813 |
+
"grad_norm": 2.8515124039952076,
|
| 1814 |
+
"learning_rate": 2e-05,
|
| 1815 |
+
"loss": 0.7252332687377929,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.08297824624355236,
|
| 1820 |
+
"grad_norm": 1.5075998819752536,
|
| 1821 |
+
"learning_rate": 2e-05,
|
| 1822 |
+
"loss": 0.6708373069763184,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.08329862557267805,
|
| 1827 |
+
"grad_norm": 2.4918434345399083,
|
| 1828 |
+
"learning_rate": 2e-05,
|
| 1829 |
+
"loss": 0.6856244087219239,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.08361900490180374,
|
| 1834 |
+
"grad_norm": 2.9895550752304025,
|
| 1835 |
+
"learning_rate": 2e-05,
|
| 1836 |
+
"loss": 0.6946100234985352,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.08393938423092942,
|
| 1841 |
+
"grad_norm": 1.2195098935112956,
|
| 1842 |
+
"learning_rate": 2e-05,
|
| 1843 |
+
"loss": 0.6571969509124755,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.0842597635600551,
|
| 1848 |
+
"grad_norm": 1.5659331888464,
|
| 1849 |
+
"learning_rate": 2e-05,
|
| 1850 |
+
"loss": 0.6633186340332031,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.0845801428891808,
|
| 1855 |
+
"grad_norm": 1.2623772678903336,
|
| 1856 |
+
"learning_rate": 2e-05,
|
| 1857 |
+
"loss": 0.7006426811218261,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.08490052221830648,
|
| 1862 |
+
"grad_norm": 53.51607876951121,
|
| 1863 |
+
"learning_rate": 2e-05,
|
| 1864 |
+
"loss": 0.7465255260467529,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.08522090154743216,
|
| 1869 |
+
"grad_norm": 7.141894169140511,
|
| 1870 |
+
"learning_rate": 2e-05,
|
| 1871 |
+
"loss": 0.6905827999114991,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.08554128087655785,
|
| 1876 |
+
"grad_norm": 12.021818224070394,
|
| 1877 |
+
"learning_rate": 2e-05,
|
| 1878 |
+
"loss": 0.7053852081298828,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.08586166020568353,
|
| 1883 |
+
"grad_norm": 1.9077255089530207,
|
| 1884 |
+
"learning_rate": 2e-05,
|
| 1885 |
+
"loss": 0.6775362968444825,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.08618203953480921,
|
| 1890 |
+
"grad_norm": 3.124384881028629,
|
| 1891 |
+
"learning_rate": 2e-05,
|
| 1892 |
+
"loss": 0.6948652267456055,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.0865024188639349,
|
| 1897 |
+
"grad_norm": 1.3550553388372282,
|
| 1898 |
+
"learning_rate": 2e-05,
|
| 1899 |
+
"loss": 0.6418863296508789,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.08682279819306059,
|
| 1904 |
+
"grad_norm": 1.6996387488097904,
|
| 1905 |
+
"learning_rate": 2e-05,
|
| 1906 |
+
"loss": 0.7217937469482422,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.08714317752218627,
|
| 1911 |
+
"grad_norm": 4.721987876312749,
|
| 1912 |
+
"learning_rate": 2e-05,
|
| 1913 |
+
"loss": 0.678973388671875,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.08746355685131195,
|
| 1918 |
+
"grad_norm": 207.96463050114306,
|
| 1919 |
+
"learning_rate": 2e-05,
|
| 1920 |
+
"loss": 0.7110597610473632,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.08778393618043764,
|
| 1925 |
+
"grad_norm": 3.3362197281306725,
|
| 1926 |
+
"learning_rate": 2e-05,
|
| 1927 |
+
"loss": 0.7547786235809326,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.08810431550956332,
|
| 1932 |
+
"grad_norm": 1.7625372241440926,
|
| 1933 |
+
"learning_rate": 2e-05,
|
| 1934 |
+
"loss": 0.678510046005249,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.088424694838689,
|
| 1939 |
+
"grad_norm": 1.7993132875303999,
|
| 1940 |
+
"learning_rate": 2e-05,
|
| 1941 |
+
"loss": 0.6995362758636474,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.08874507416781469,
|
| 1946 |
+
"grad_norm": 7.076835439735431,
|
| 1947 |
+
"learning_rate": 2e-05,
|
| 1948 |
+
"loss": 0.6843991279602051,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.08906545349694038,
|
| 1953 |
+
"grad_norm": 1.6043720642575496,
|
| 1954 |
+
"learning_rate": 2e-05,
|
| 1955 |
+
"loss": 0.6938441276550293,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.08938583282606606,
|
| 1960 |
+
"grad_norm": 1.3640725070465638,
|
| 1961 |
+
"learning_rate": 2e-05,
|
| 1962 |
+
"loss": 0.7047198295593262,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.08970621215519174,
|
| 1967 |
+
"grad_norm": 2.562596485102053,
|
| 1968 |
+
"learning_rate": 2e-05,
|
| 1969 |
+
"loss": 0.6708857536315918,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.09002659148431744,
|
| 1974 |
+
"grad_norm": 1.9416658533084818,
|
| 1975 |
+
"learning_rate": 2e-05,
|
| 1976 |
+
"loss": 0.6964911937713623,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.09034697081344312,
|
| 1981 |
+
"grad_norm": 1.5822864532309853,
|
| 1982 |
+
"learning_rate": 2e-05,
|
| 1983 |
+
"loss": 0.6741170406341552,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.0906673501425688,
|
| 1988 |
+
"grad_norm": 2.492955276963355,
|
| 1989 |
+
"learning_rate": 2e-05,
|
| 1990 |
+
"loss": 0.6612090110778809,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.09098772947169449,
|
| 1995 |
+
"grad_norm": 12.577880739333416,
|
| 1996 |
+
"learning_rate": 2e-05,
|
| 1997 |
+
"loss": 0.6880275249481201,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.09130810880082017,
|
| 2002 |
+
"grad_norm": 2.498526802339318,
|
| 2003 |
+
"learning_rate": 2e-05,
|
| 2004 |
+
"loss": 0.6688164710998535,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.09162848812994585,
|
| 2009 |
+
"grad_norm": 2.480207709776554,
|
| 2010 |
+
"learning_rate": 2e-05,
|
| 2011 |
+
"loss": 0.8187369346618653,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.09194886745907153,
|
| 2016 |
+
"grad_norm": 1.7387835479887794,
|
| 2017 |
+
"learning_rate": 2e-05,
|
| 2018 |
+
"loss": 0.6645703315734863,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.09226924678819723,
|
| 2023 |
+
"grad_norm": 2.463777428019542,
|
| 2024 |
+
"learning_rate": 2e-05,
|
| 2025 |
+
"loss": 0.680314016342163,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.09258962611732291,
|
| 2030 |
+
"grad_norm": 2.1125181898998457,
|
| 2031 |
+
"learning_rate": 2e-05,
|
| 2032 |
+
"loss": 0.7115361213684082,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.09291000544644859,
|
| 2037 |
+
"grad_norm": 3.6819788201167962,
|
| 2038 |
+
"learning_rate": 2e-05,
|
| 2039 |
+
"loss": 0.7376339912414551,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.09323038477557428,
|
| 2044 |
+
"grad_norm": 5.364519716326931,
|
| 2045 |
+
"learning_rate": 2e-05,
|
| 2046 |
+
"loss": 0.7281315803527832,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.09355076410469997,
|
| 2051 |
+
"grad_norm": 5.2868820878632246,
|
| 2052 |
+
"learning_rate": 2e-05,
|
| 2053 |
+
"loss": 0.6780884742736817,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.09387114343382565,
|
| 2058 |
+
"grad_norm": 2.569546482429642,
|
| 2059 |
+
"learning_rate": 2e-05,
|
| 2060 |
+
"loss": 0.6810887813568115,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.09419152276295134,
|
| 2065 |
+
"grad_norm": 3.1590297185599887,
|
| 2066 |
+
"learning_rate": 2e-05,
|
| 2067 |
+
"loss": 0.6913040161132813,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.09451190209207702,
|
| 2072 |
+
"grad_norm": 2.2404038565045,
|
| 2073 |
+
"learning_rate": 2e-05,
|
| 2074 |
+
"loss": 0.639086103439331,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.0948322814212027,
|
| 2079 |
+
"grad_norm": 1.6593714513437243,
|
| 2080 |
+
"learning_rate": 2e-05,
|
| 2081 |
+
"loss": 0.6881752014160156,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.09515266075032838,
|
| 2086 |
+
"grad_norm": 1.9761898426256463,
|
| 2087 |
+
"learning_rate": 2e-05,
|
| 2088 |
+
"loss": 0.6969388484954834,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.09547304007945408,
|
| 2093 |
+
"grad_norm": 1.5176347302651876,
|
| 2094 |
+
"learning_rate": 2e-05,
|
| 2095 |
+
"loss": 0.6573243141174316,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.09579341940857976,
|
| 2100 |
+
"grad_norm": 1.5490982461158973,
|
| 2101 |
+
"learning_rate": 2e-05,
|
| 2102 |
+
"loss": 0.6932000160217285,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.09611379873770544,
|
| 2107 |
+
"grad_norm": 3.9352475270631273,
|
| 2108 |
+
"learning_rate": 2e-05,
|
| 2109 |
+
"loss": 0.649666690826416,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.09643417806683113,
|
| 2114 |
+
"grad_norm": 3.0813498769346834,
|
| 2115 |
+
"learning_rate": 2e-05,
|
| 2116 |
+
"loss": 0.6733899593353272,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.09675455739595681,
|
| 2121 |
+
"grad_norm": 1.4582486886298482,
|
| 2122 |
+
"learning_rate": 2e-05,
|
| 2123 |
+
"loss": 0.6860736846923828,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.0970749367250825,
|
| 2128 |
+
"grad_norm": 2.1316688325298223,
|
| 2129 |
+
"learning_rate": 2e-05,
|
| 2130 |
+
"loss": 0.6471640586853027,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.09739531605420819,
|
| 2135 |
+
"grad_norm": 1.624813158041112,
|
| 2136 |
+
"learning_rate": 2e-05,
|
| 2137 |
+
"loss": 0.6769994258880615,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.09771569538333387,
|
| 2142 |
+
"grad_norm": 1.1991400569586823,
|
| 2143 |
+
"learning_rate": 2e-05,
|
| 2144 |
+
"loss": 0.7090652465820313,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.09803607471245955,
|
| 2149 |
+
"grad_norm": 1.8553248908588058,
|
| 2150 |
+
"learning_rate": 2e-05,
|
| 2151 |
+
"loss": 0.6960935115814209,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.09835645404158523,
|
| 2156 |
+
"grad_norm": 1.0947674863297467,
|
| 2157 |
+
"learning_rate": 2e-05,
|
| 2158 |
+
"loss": 0.7006158828735352,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.09867683337071093,
|
| 2163 |
+
"grad_norm": 1.9345154791117662,
|
| 2164 |
+
"learning_rate": 2e-05,
|
| 2165 |
+
"loss": 0.6712536811828613,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.0989972126998366,
|
| 2170 |
+
"grad_norm": 1.206110040611478,
|
| 2171 |
+
"learning_rate": 2e-05,
|
| 2172 |
+
"loss": 0.6646307945251465,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.09931759202896229,
|
| 2177 |
+
"grad_norm": 1.5706700337393924,
|
| 2178 |
+
"learning_rate": 2e-05,
|
| 2179 |
+
"loss": 0.721156644821167,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.09963797135808798,
|
| 2184 |
+
"grad_norm": 1.8852007896443763,
|
| 2185 |
+
"learning_rate": 2e-05,
|
| 2186 |
+
"loss": 0.6588726997375488,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.09995835068721366,
|
| 2191 |
+
"grad_norm": 1.9309188732059843,
|
| 2192 |
+
"learning_rate": 2e-05,
|
| 2193 |
+
"loss": 0.6430281639099121,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.10027873001633934,
|
| 2198 |
+
"grad_norm": 1.8247916600609566,
|
| 2199 |
+
"learning_rate": 2e-05,
|
| 2200 |
+
"loss": 0.662308645248413,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.10059910934546502,
|
| 2205 |
+
"grad_norm": 20.234568515276454,
|
| 2206 |
+
"learning_rate": 2e-05,
|
| 2207 |
+
"loss": 0.6764077186584473,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.10091948867459072,
|
| 2212 |
+
"grad_norm": 2.270401562255825,
|
| 2213 |
+
"learning_rate": 2e-05,
|
| 2214 |
+
"loss": 0.6858000755310059,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.1012398680037164,
|
| 2219 |
+
"grad_norm": 3.3124667080997154,
|
| 2220 |
+
"learning_rate": 2e-05,
|
| 2221 |
+
"loss": 0.7174823760986329,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.10156024733284208,
|
| 2226 |
+
"grad_norm": 1.21628975030768,
|
| 2227 |
+
"learning_rate": 2e-05,
|
| 2228 |
+
"loss": 0.7047437191009521,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.10188062666196777,
|
| 2233 |
+
"grad_norm": 1.8787293361467063,
|
| 2234 |
+
"learning_rate": 2e-05,
|
| 2235 |
+
"loss": 0.6899389266967774,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.10220100599109345,
|
| 2240 |
+
"grad_norm": 3.435196857519197,
|
| 2241 |
+
"learning_rate": 2e-05,
|
| 2242 |
+
"loss": 0.678690242767334,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.10252138532021914,
|
| 2247 |
+
"grad_norm": 2.4128135351683486,
|
| 2248 |
+
"learning_rate": 2e-05,
|
| 2249 |
+
"loss": 0.6727965354919434,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.10284176464934483,
|
| 2254 |
+
"grad_norm": 5.463858532009493,
|
| 2255 |
+
"learning_rate": 2e-05,
|
| 2256 |
+
"loss": 0.7546576499938965,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.10316214397847051,
|
| 2261 |
+
"grad_norm": 1.1458273204458618,
|
| 2262 |
+
"learning_rate": 2e-05,
|
| 2263 |
+
"loss": 0.6437710762023926,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.10348252330759619,
|
| 2268 |
+
"grad_norm": 1.3601974957658387,
|
| 2269 |
+
"learning_rate": 2e-05,
|
| 2270 |
+
"loss": 0.6784330368041992,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.10380290263672187,
|
| 2275 |
+
"grad_norm": 1.4840886563319795,
|
| 2276 |
+
"learning_rate": 2e-05,
|
| 2277 |
+
"loss": 0.6686971664428711,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.10412328196584757,
|
| 2282 |
+
"grad_norm": 1.3597430285112961,
|
| 2283 |
+
"learning_rate": 2e-05,
|
| 2284 |
+
"loss": 0.6590595245361328,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.10444366129497325,
|
| 2289 |
+
"grad_norm": 2.7391179546585467,
|
| 2290 |
+
"learning_rate": 2e-05,
|
| 2291 |
+
"loss": 0.6681632995605469,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.10476404062409893,
|
| 2296 |
+
"grad_norm": 2.9881385443242867,
|
| 2297 |
+
"learning_rate": 2e-05,
|
| 2298 |
+
"loss": 0.6919444084167481,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.10508441995322462,
|
| 2303 |
+
"grad_norm": 1.7848898733758627,
|
| 2304 |
+
"learning_rate": 2e-05,
|
| 2305 |
+
"loss": 0.678253173828125,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.1054047992823503,
|
| 2310 |
+
"grad_norm": 1.0941482258819155,
|
| 2311 |
+
"learning_rate": 2e-05,
|
| 2312 |
+
"loss": 0.6677752494812011,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.10572517861147598,
|
| 2317 |
+
"grad_norm": 1.1439491512926567,
|
| 2318 |
+
"learning_rate": 2e-05,
|
| 2319 |
+
"loss": 0.6404883384704589,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.10604555794060168,
|
| 2324 |
+
"grad_norm": 1.6084438657234088,
|
| 2325 |
+
"learning_rate": 2e-05,
|
| 2326 |
+
"loss": 0.6679693222045898,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.10636593726972736,
|
| 2331 |
+
"grad_norm": 2.3221962830811265,
|
| 2332 |
+
"learning_rate": 2e-05,
|
| 2333 |
+
"loss": 0.6701303482055664,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.10668631659885304,
|
| 2338 |
+
"grad_norm": 1.1626176984318963,
|
| 2339 |
+
"learning_rate": 2e-05,
|
| 2340 |
+
"loss": 0.6760411262512207,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.10700669592797872,
|
| 2345 |
+
"grad_norm": 9.60833542052224,
|
| 2346 |
+
"learning_rate": 2e-05,
|
| 2347 |
+
"loss": 0.6958446502685547,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.10732707525710442,
|
| 2352 |
+
"grad_norm": 1.4570489896161745,
|
| 2353 |
+
"learning_rate": 2e-05,
|
| 2354 |
+
"loss": 0.7142772674560547,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.1076474545862301,
|
| 2359 |
+
"grad_norm": 1.589527243320569,
|
| 2360 |
+
"learning_rate": 2e-05,
|
| 2361 |
+
"loss": 0.6480647087097168,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.10796783391535578,
|
| 2366 |
+
"grad_norm": 3.7539533800821854,
|
| 2367 |
+
"learning_rate": 2e-05,
|
| 2368 |
+
"loss": 0.6926058769226074,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.10828821324448147,
|
| 2373 |
+
"grad_norm": 1.4343813974850903,
|
| 2374 |
+
"learning_rate": 2e-05,
|
| 2375 |
+
"loss": 0.6597871303558349,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.10860859257360715,
|
| 2380 |
+
"grad_norm": 1.700489983652249,
|
| 2381 |
+
"learning_rate": 2e-05,
|
| 2382 |
+
"loss": 0.6854336738586426,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.10892897190273283,
|
| 2387 |
+
"grad_norm": 2.262978455318886,
|
| 2388 |
+
"learning_rate": 2e-05,
|
| 2389 |
+
"loss": 0.6611383438110352,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.10924935123185853,
|
| 2394 |
+
"grad_norm": 3.7384144861916924,
|
| 2395 |
+
"learning_rate": 2e-05,
|
| 2396 |
+
"loss": 0.6642798900604248,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.10956973056098421,
|
| 2401 |
+
"grad_norm": 5.828313398251216,
|
| 2402 |
+
"learning_rate": 2e-05,
|
| 2403 |
+
"loss": 0.6781678676605225,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.10989010989010989,
|
| 2408 |
+
"grad_norm": 5.487121962637965,
|
| 2409 |
+
"learning_rate": 2e-05,
|
| 2410 |
+
"loss": 0.7087061882019043,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.11021048921923557,
|
| 2415 |
+
"grad_norm": 1.6357218209966338,
|
| 2416 |
+
"learning_rate": 2e-05,
|
| 2417 |
+
"loss": 0.6441525936126709,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.11053086854836126,
|
| 2422 |
+
"grad_norm": 1.9006368500272885,
|
| 2423 |
+
"learning_rate": 2e-05,
|
| 2424 |
+
"loss": 0.6848069667816162,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.11085124787748694,
|
| 2429 |
+
"grad_norm": 1.4766899104425295,
|
| 2430 |
+
"learning_rate": 2e-05,
|
| 2431 |
+
"loss": 0.6893895149230957,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.11117162720661262,
|
| 2436 |
+
"grad_norm": 2.769884408658176,
|
| 2437 |
+
"learning_rate": 2e-05,
|
| 2438 |
+
"loss": 0.6294010162353516,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.11149200653573832,
|
| 2443 |
+
"grad_norm": 4.65526982930436,
|
| 2444 |
+
"learning_rate": 2e-05,
|
| 2445 |
+
"loss": 0.6821585655212402,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.111812385864864,
|
| 2450 |
+
"grad_norm": 1.7647620795155232,
|
| 2451 |
+
"learning_rate": 2e-05,
|
| 2452 |
+
"loss": 0.6744832515716552,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.11213276519398968,
|
| 2457 |
+
"grad_norm": 1.2028872660134693,
|
| 2458 |
+
"learning_rate": 2e-05,
|
| 2459 |
+
"loss": 0.663267707824707,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.11245314452311538,
|
| 2464 |
+
"grad_norm": 4.612458331540539,
|
| 2465 |
+
"learning_rate": 2e-05,
|
| 2466 |
+
"loss": 0.6622488021850585,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.11277352385224106,
|
| 2471 |
+
"grad_norm": 3.060996296931568,
|
| 2472 |
+
"learning_rate": 2e-05,
|
| 2473 |
+
"loss": 0.6657146453857422,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.11309390318136674,
|
| 2478 |
+
"grad_norm": 1.3030075500669922,
|
| 2479 |
+
"learning_rate": 2e-05,
|
| 2480 |
+
"loss": 0.6406706809997559,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.11341428251049242,
|
| 2485 |
+
"grad_norm": 2.019294690660488,
|
| 2486 |
+
"learning_rate": 2e-05,
|
| 2487 |
+
"loss": 0.6651981353759766,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.11373466183961811,
|
| 2492 |
+
"grad_norm": 1.2086238864300596,
|
| 2493 |
+
"learning_rate": 2e-05,
|
| 2494 |
+
"loss": 0.6543137550354003,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.11405504116874379,
|
| 2499 |
+
"grad_norm": 1.2655664277456855,
|
| 2500 |
+
"learning_rate": 2e-05,
|
| 2501 |
+
"loss": 0.6376056671142578,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.11437542049786947,
|
| 2506 |
+
"grad_norm": 2.0908487857785176,
|
| 2507 |
+
"learning_rate": 2e-05,
|
| 2508 |
+
"loss": 0.6656252861022949,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.11469579982699517,
|
| 2513 |
+
"grad_norm": 5.261221133161816,
|
| 2514 |
+
"learning_rate": 2e-05,
|
| 2515 |
+
"loss": 0.6583863258361816,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.11501617915612085,
|
| 2520 |
+
"grad_norm": 2.3910426503344664,
|
| 2521 |
+
"learning_rate": 2e-05,
|
| 2522 |
+
"loss": 0.6792299270629882,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.11533655848524653,
|
| 2527 |
+
"grad_norm": 1.211765741906558,
|
| 2528 |
+
"learning_rate": 2e-05,
|
| 2529 |
+
"loss": 0.7030490398406982,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.11565693781437221,
|
| 2534 |
+
"grad_norm": 1.9832630555170212,
|
| 2535 |
+
"learning_rate": 2e-05,
|
| 2536 |
+
"loss": 0.6824533462524414,
|
| 2537 |
+
"step": 3610
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.1159773171434979,
|
| 2541 |
+
"grad_norm": 2.252162260913484,
|
| 2542 |
+
"learning_rate": 2e-05,
|
| 2543 |
+
"loss": 0.6910503387451172,
|
| 2544 |
+
"step": 3620
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.11629769647262359,
|
| 2548 |
+
"grad_norm": 1.2137738587533493,
|
| 2549 |
+
"learning_rate": 2e-05,
|
| 2550 |
+
"loss": 0.6761364936828613,
|
| 2551 |
+
"step": 3630
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.11661807580174927,
|
| 2555 |
+
"grad_norm": 1.6480847954089972,
|
| 2556 |
+
"learning_rate": 2e-05,
|
| 2557 |
+
"loss": 0.6949734687805176,
|
| 2558 |
+
"step": 3640
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.11693845513087496,
|
| 2562 |
+
"grad_norm": 2.773368722100306,
|
| 2563 |
+
"learning_rate": 2e-05,
|
| 2564 |
+
"loss": 0.6389493942260742,
|
| 2565 |
+
"step": 3650
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.11725883446000064,
|
| 2569 |
+
"grad_norm": 2.5465741003121365,
|
| 2570 |
+
"learning_rate": 2e-05,
|
| 2571 |
+
"loss": 0.6798902034759522,
|
| 2572 |
+
"step": 3660
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.11757921378912632,
|
| 2576 |
+
"grad_norm": 3.003062238735499,
|
| 2577 |
+
"learning_rate": 2e-05,
|
| 2578 |
+
"loss": 0.6937604427337647,
|
| 2579 |
+
"step": 3670
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.11789959311825202,
|
| 2583 |
+
"grad_norm": 1.0771834972401186,
|
| 2584 |
+
"learning_rate": 2e-05,
|
| 2585 |
+
"loss": 0.6856004238128662,
|
| 2586 |
+
"step": 3680
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.1182199724473777,
|
| 2590 |
+
"grad_norm": 3.2223166693736465,
|
| 2591 |
+
"learning_rate": 2e-05,
|
| 2592 |
+
"loss": 0.7183493614196778,
|
| 2593 |
+
"step": 3690
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.11854035177650338,
|
| 2597 |
+
"grad_norm": 1.90983436518728,
|
| 2598 |
+
"learning_rate": 2e-05,
|
| 2599 |
+
"loss": 0.6716906547546386,
|
| 2600 |
+
"step": 3700
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.11886073110562906,
|
| 2604 |
+
"grad_norm": 2.0873401588953433,
|
| 2605 |
+
"learning_rate": 2e-05,
|
| 2606 |
+
"loss": 0.666513729095459,
|
| 2607 |
+
"step": 3710
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.11918111043475475,
|
| 2611 |
+
"grad_norm": 1.1975205039758825,
|
| 2612 |
+
"learning_rate": 2e-05,
|
| 2613 |
+
"loss": 0.6365289211273193,
|
| 2614 |
+
"step": 3720
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.11950148976388043,
|
| 2618 |
+
"grad_norm": 41.55973254495974,
|
| 2619 |
+
"learning_rate": 2e-05,
|
| 2620 |
+
"loss": 0.7242575645446777,
|
| 2621 |
+
"step": 3730
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.11982186909300611,
|
| 2625 |
+
"grad_norm": 1.7969703921280566,
|
| 2626 |
+
"learning_rate": 2e-05,
|
| 2627 |
+
"loss": 0.6333506584167481,
|
| 2628 |
+
"step": 3740
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.12014224842213181,
|
| 2632 |
+
"grad_norm": 1.1484967465503775,
|
| 2633 |
+
"learning_rate": 2e-05,
|
| 2634 |
+
"loss": 0.681650447845459,
|
| 2635 |
+
"step": 3750
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 0.12046262775125749,
|
| 2639 |
+
"grad_norm": 1.491743333469875,
|
| 2640 |
+
"learning_rate": 2e-05,
|
| 2641 |
+
"loss": 0.683896541595459,
|
| 2642 |
+
"step": 3760
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.12078300708038317,
|
| 2646 |
+
"grad_norm": 1.9896310599000153,
|
| 2647 |
+
"learning_rate": 2e-05,
|
| 2648 |
+
"loss": 0.6723162651062011,
|
| 2649 |
+
"step": 3770
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 0.12110338640950886,
|
| 2653 |
+
"grad_norm": 4.355204703369626,
|
| 2654 |
+
"learning_rate": 2e-05,
|
| 2655 |
+
"loss": 0.6825237274169922,
|
| 2656 |
+
"step": 3780
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.12142376573863455,
|
| 2660 |
+
"grad_norm": 2.6337758244589446,
|
| 2661 |
+
"learning_rate": 2e-05,
|
| 2662 |
+
"loss": 0.6493152618408203,
|
| 2663 |
+
"step": 3790
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.12174414506776023,
|
| 2667 |
+
"grad_norm": 1.4311522270790977,
|
| 2668 |
+
"learning_rate": 2e-05,
|
| 2669 |
+
"loss": 0.6465260982513428,
|
| 2670 |
+
"step": 3800
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.1220645243968859,
|
| 2674 |
+
"grad_norm": 1.2584234760746995,
|
| 2675 |
+
"learning_rate": 2e-05,
|
| 2676 |
+
"loss": 0.6119470119476318,
|
| 2677 |
+
"step": 3810
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 0.1223849037260116,
|
| 2681 |
+
"grad_norm": 1.8362919442209469,
|
| 2682 |
+
"learning_rate": 2e-05,
|
| 2683 |
+
"loss": 0.6505841732025146,
|
| 2684 |
+
"step": 3820
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.12270528305513728,
|
| 2688 |
+
"grad_norm": 3.2911678224166363,
|
| 2689 |
+
"learning_rate": 2e-05,
|
| 2690 |
+
"loss": 0.7023921012878418,
|
| 2691 |
+
"step": 3830
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 0.12302566238426296,
|
| 2695 |
+
"grad_norm": 1.8619496153981334,
|
| 2696 |
+
"learning_rate": 2e-05,
|
| 2697 |
+
"loss": 0.676816749572754,
|
| 2698 |
+
"step": 3840
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 0.12334604171338866,
|
| 2702 |
+
"grad_norm": 1.2607793375322882,
|
| 2703 |
+
"learning_rate": 2e-05,
|
| 2704 |
+
"loss": 0.6600100994110107,
|
| 2705 |
+
"step": 3850
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.12366642104251434,
|
| 2709 |
+
"grad_norm": 1.3500576986027626,
|
| 2710 |
+
"learning_rate": 2e-05,
|
| 2711 |
+
"loss": 0.6765620708465576,
|
| 2712 |
+
"step": 3860
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 0.12398680037164002,
|
| 2716 |
+
"grad_norm": 1.3401379911166065,
|
| 2717 |
+
"learning_rate": 2e-05,
|
| 2718 |
+
"loss": 0.6644556045532226,
|
| 2719 |
+
"step": 3870
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 0.12430717970076571,
|
| 2723 |
+
"grad_norm": 1.3436995912683865,
|
| 2724 |
+
"learning_rate": 2e-05,
|
| 2725 |
+
"loss": 0.6688062191009522,
|
| 2726 |
+
"step": 3880
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.1246275590298914,
|
| 2730 |
+
"grad_norm": 2.1554623740628323,
|
| 2731 |
+
"learning_rate": 2e-05,
|
| 2732 |
+
"loss": 0.6625113487243652,
|
| 2733 |
+
"step": 3890
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 0.12494793835901707,
|
| 2737 |
+
"grad_norm": 2.5490411787256693,
|
| 2738 |
+
"learning_rate": 2e-05,
|
| 2739 |
+
"loss": 0.6453998565673829,
|
| 2740 |
+
"step": 3900
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 0.12526831768814276,
|
| 2744 |
+
"grad_norm": 2.1422672897977475,
|
| 2745 |
+
"learning_rate": 2e-05,
|
| 2746 |
+
"loss": 0.6868985652923584,
|
| 2747 |
+
"step": 3910
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 0.12558869701726844,
|
| 2751 |
+
"grad_norm": 8.647039761010506,
|
| 2752 |
+
"learning_rate": 2e-05,
|
| 2753 |
+
"loss": 0.6461306571960449,
|
| 2754 |
+
"step": 3920
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 0.12590907634639414,
|
| 2758 |
+
"grad_norm": 4.353384006867587,
|
| 2759 |
+
"learning_rate": 2e-05,
|
| 2760 |
+
"loss": 0.6968627452850342,
|
| 2761 |
+
"step": 3930
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 0.12622945567551982,
|
| 2765 |
+
"grad_norm": 5.312771003446082,
|
| 2766 |
+
"learning_rate": 2e-05,
|
| 2767 |
+
"loss": 0.6628563880920411,
|
| 2768 |
+
"step": 3940
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 0.1265498350046455,
|
| 2772 |
+
"grad_norm": 1.2483064513173903,
|
| 2773 |
+
"learning_rate": 2e-05,
|
| 2774 |
+
"loss": 0.6820049285888672,
|
| 2775 |
+
"step": 3950
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 0.1268702143337712,
|
| 2779 |
+
"grad_norm": 1.7052048413362177,
|
| 2780 |
+
"learning_rate": 2e-05,
|
| 2781 |
+
"loss": 0.6507192134857178,
|
| 2782 |
+
"step": 3960
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 0.12719059366289687,
|
| 2786 |
+
"grad_norm": 1.442220593830901,
|
| 2787 |
+
"learning_rate": 2e-05,
|
| 2788 |
+
"loss": 0.6692718505859375,
|
| 2789 |
+
"step": 3970
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 0.12751097299202255,
|
| 2793 |
+
"grad_norm": 21.674983962515157,
|
| 2794 |
+
"learning_rate": 2e-05,
|
| 2795 |
+
"loss": 0.6392804622650147,
|
| 2796 |
+
"step": 3980
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 0.12783135232114823,
|
| 2800 |
+
"grad_norm": 9.728263472908212,
|
| 2801 |
+
"learning_rate": 2e-05,
|
| 2802 |
+
"loss": 0.656775426864624,
|
| 2803 |
+
"step": 3990
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 0.12815173165027394,
|
| 2807 |
+
"grad_norm": 2.039616161760123,
|
| 2808 |
+
"learning_rate": 2e-05,
|
| 2809 |
+
"loss": 0.7095576286315918,
|
| 2810 |
+
"step": 4000
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 0.12847211097939962,
|
| 2814 |
+
"grad_norm": 4.0175728810676805,
|
| 2815 |
+
"learning_rate": 2e-05,
|
| 2816 |
+
"loss": 0.6841886043548584,
|
| 2817 |
+
"step": 4010
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 0.1287924903085253,
|
| 2821 |
+
"grad_norm": 19.73418939091534,
|
| 2822 |
+
"learning_rate": 2e-05,
|
| 2823 |
+
"loss": 0.6358445644378662,
|
| 2824 |
+
"step": 4020
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 0.12911286963765098,
|
| 2828 |
+
"grad_norm": 1.739996971425279,
|
| 2829 |
+
"learning_rate": 2e-05,
|
| 2830 |
+
"loss": 0.6840127944946289,
|
| 2831 |
+
"step": 4030
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 0.12943324896677666,
|
| 2835 |
+
"grad_norm": 1.450557707510894,
|
| 2836 |
+
"learning_rate": 2e-05,
|
| 2837 |
+
"loss": 0.6738371849060059,
|
| 2838 |
+
"step": 4040
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 0.12975362829590234,
|
| 2842 |
+
"grad_norm": 2.0404367072041,
|
| 2843 |
+
"learning_rate": 2e-05,
|
| 2844 |
+
"loss": 0.6749392509460449,
|
| 2845 |
+
"step": 4050
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 0.13007400762502802,
|
| 2849 |
+
"grad_norm": 1.6413128097895142,
|
| 2850 |
+
"learning_rate": 2e-05,
|
| 2851 |
+
"loss": 0.6790598869323731,
|
| 2852 |
+
"step": 4060
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 0.13039438695415373,
|
| 2856 |
+
"grad_norm": 1.2445737299020967,
|
| 2857 |
+
"learning_rate": 2e-05,
|
| 2858 |
+
"loss": 0.7365617752075195,
|
| 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:e12d9e9836f90b7e3bf6e6beadff81e2cd76af2fbd6aa526cb1396f85ef63153
|
| 3 |
+
size 7313
|