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 sha256:2384781ce133a10f54529c93b98ec59b50111126e7512c331470af1f2c880f09
|
| 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": 0.2195727233491584,
|
| 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": 0.0005394907207596029,
|
| 14 |
+
"grad_norm": 369.69301600900053,
|
| 15 |
+
"learning_rate": 2.910112359550562e-06,
|
| 16 |
+
"loss": 1.8730627059936524,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0010789814415192059,
|
| 21 |
+
"grad_norm": 31.93478853689241,
|
| 22 |
+
"learning_rate": 3.921348314606742e-06,
|
| 23 |
+
"loss": 1.262525463104248,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0016184721622788088,
|
| 28 |
+
"grad_norm": 141.7556466471986,
|
| 29 |
+
"learning_rate": 4.932584269662922e-06,
|
| 30 |
+
"loss": 1.0669395446777343,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0021579628830384117,
|
| 35 |
+
"grad_norm": 127.98270782023548,
|
| 36 |
+
"learning_rate": 5.943820224719102e-06,
|
| 37 |
+
"loss": 1.1076365470886231,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.002697453603798015,
|
| 42 |
+
"grad_norm": 101.43748401214339,
|
| 43 |
+
"learning_rate": 6.955056179775282e-06,
|
| 44 |
+
"loss": 1.0109930038452148,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0032369443245576176,
|
| 49 |
+
"grad_norm": 899.0086783565027,
|
| 50 |
+
"learning_rate": 7.966292134831462e-06,
|
| 51 |
+
"loss": 0.9515115737915039,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0037764350453172208,
|
| 56 |
+
"grad_norm": 15.557933246692217,
|
| 57 |
+
"learning_rate": 8.977528089887641e-06,
|
| 58 |
+
"loss": 0.9290966987609863,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.0043159257660768235,
|
| 63 |
+
"grad_norm": 256.064896120912,
|
| 64 |
+
"learning_rate": 9.988764044943822e-06,
|
| 65 |
+
"loss": 1.1323322296142577,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.004855416486836427,
|
| 70 |
+
"grad_norm": 20.848525026966247,
|
| 71 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 72 |
+
"loss": 1.088106918334961,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.00539490720759603,
|
| 77 |
+
"grad_norm": 25.60593783327153,
|
| 78 |
+
"learning_rate": 1.2011235955056182e-05,
|
| 79 |
+
"loss": 0.9955007553100585,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.005934397928355632,
|
| 84 |
+
"grad_norm": 11.664176937579581,
|
| 85 |
+
"learning_rate": 1.3022471910112362e-05,
|
| 86 |
+
"loss": 0.942236328125,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.006473888649115235,
|
| 91 |
+
"grad_norm": 7.9248397322900646,
|
| 92 |
+
"learning_rate": 1.403370786516854e-05,
|
| 93 |
+
"loss": 0.9238415718078613,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.007013379369874838,
|
| 98 |
+
"grad_norm": 26.682291952362966,
|
| 99 |
+
"learning_rate": 1.504494382022472e-05,
|
| 100 |
+
"loss": 0.9361964225769043,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.0075528700906344415,
|
| 105 |
+
"grad_norm": 3.6691592434398466,
|
| 106 |
+
"learning_rate": 1.60561797752809e-05,
|
| 107 |
+
"loss": 0.927226448059082,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.008092360811394044,
|
| 112 |
+
"grad_norm": 15.191132047351614,
|
| 113 |
+
"learning_rate": 1.706741573033708e-05,
|
| 114 |
+
"loss": 0.9375665664672852,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.008631851532153647,
|
| 119 |
+
"grad_norm": 7.4209329049996455,
|
| 120 |
+
"learning_rate": 1.8078651685393256e-05,
|
| 121 |
+
"loss": 0.9221304893493653,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.00917134225291325,
|
| 126 |
+
"grad_norm": 5.050234356436225,
|
| 127 |
+
"learning_rate": 1.908988764044944e-05,
|
| 128 |
+
"loss": 0.9327095031738282,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.009710832973672853,
|
| 133 |
+
"grad_norm": 2.532425838265676,
|
| 134 |
+
"learning_rate": 2e-05,
|
| 135 |
+
"loss": 0.90814208984375,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.010250323694432456,
|
| 140 |
+
"grad_norm": 2.8569985447993664,
|
| 141 |
+
"learning_rate": 2e-05,
|
| 142 |
+
"loss": 0.930181884765625,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.01078981441519206,
|
| 147 |
+
"grad_norm": 3.7217733328209746,
|
| 148 |
+
"learning_rate": 2e-05,
|
| 149 |
+
"loss": 0.9137080192565918,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.011329305135951661,
|
| 154 |
+
"grad_norm": 1.6856605125936686,
|
| 155 |
+
"learning_rate": 2e-05,
|
| 156 |
+
"loss": 0.9173044204711914,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.011868795856711264,
|
| 161 |
+
"grad_norm": 4.042279257566151,
|
| 162 |
+
"learning_rate": 2e-05,
|
| 163 |
+
"loss": 0.9044102668762207,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.012408286577470867,
|
| 168 |
+
"grad_norm": 7.693180928481417,
|
| 169 |
+
"learning_rate": 2e-05,
|
| 170 |
+
"loss": 0.9013534545898437,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.01294777729823047,
|
| 175 |
+
"grad_norm": 3.5295771823615647,
|
| 176 |
+
"learning_rate": 2e-05,
|
| 177 |
+
"loss": 0.9148492813110352,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.013487268018990074,
|
| 182 |
+
"grad_norm": 7.839029508109661,
|
| 183 |
+
"learning_rate": 2e-05,
|
| 184 |
+
"loss": 0.9837350845336914,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.014026758739749677,
|
| 189 |
+
"grad_norm": 3.670914514568096,
|
| 190 |
+
"learning_rate": 2e-05,
|
| 191 |
+
"loss": 0.9146605491638183,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.01456624946050928,
|
| 196 |
+
"grad_norm": 3.5577694379027,
|
| 197 |
+
"learning_rate": 2e-05,
|
| 198 |
+
"loss": 0.9401106834411621,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.015105740181268883,
|
| 203 |
+
"grad_norm": 2.453202809896575,
|
| 204 |
+
"learning_rate": 2e-05,
|
| 205 |
+
"loss": 0.9477849960327148,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.015645230902028486,
|
| 210 |
+
"grad_norm": 1.9729075743285769,
|
| 211 |
+
"learning_rate": 2e-05,
|
| 212 |
+
"loss": 0.9066492080688476,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.016184721622788088,
|
| 217 |
+
"grad_norm": 2.8099192390223795,
|
| 218 |
+
"learning_rate": 2e-05,
|
| 219 |
+
"loss": 0.9063838958740235,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.016724212343547692,
|
| 224 |
+
"grad_norm": 4.236623113420927,
|
| 225 |
+
"learning_rate": 2e-05,
|
| 226 |
+
"loss": 0.8637477874755859,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.017263703064307294,
|
| 231 |
+
"grad_norm": 4.167628129961564,
|
| 232 |
+
"learning_rate": 2e-05,
|
| 233 |
+
"loss": 0.9061297416687012,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.017803193785066895,
|
| 238 |
+
"grad_norm": 6.935834245033307,
|
| 239 |
+
"learning_rate": 2e-05,
|
| 240 |
+
"loss": 0.8871530532836914,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.0183426845058265,
|
| 245 |
+
"grad_norm": 3.2423751981314384,
|
| 246 |
+
"learning_rate": 2e-05,
|
| 247 |
+
"loss": 0.9223550796508789,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0188821752265861,
|
| 252 |
+
"grad_norm": 30.606793216440163,
|
| 253 |
+
"learning_rate": 2e-05,
|
| 254 |
+
"loss": 3.54920654296875,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.019421665947345707,
|
| 259 |
+
"grad_norm": 4.550939608135107,
|
| 260 |
+
"learning_rate": 2e-05,
|
| 261 |
+
"loss": 0.9992364883422852,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.019961156668105308,
|
| 266 |
+
"grad_norm": 10.355788989674835,
|
| 267 |
+
"learning_rate": 2e-05,
|
| 268 |
+
"loss": 0.9463441848754883,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.020500647388864913,
|
| 273 |
+
"grad_norm": 9.470092673566741,
|
| 274 |
+
"learning_rate": 2e-05,
|
| 275 |
+
"loss": 0.9165935516357422,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.021040138109624514,
|
| 280 |
+
"grad_norm": 16.04366173130251,
|
| 281 |
+
"learning_rate": 2e-05,
|
| 282 |
+
"loss": 0.9199161529541016,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.02157962883038412,
|
| 287 |
+
"grad_norm": 16.276111063417574,
|
| 288 |
+
"learning_rate": 2e-05,
|
| 289 |
+
"loss": 0.8900386810302734,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.02211911955114372,
|
| 294 |
+
"grad_norm": 14.653812444903842,
|
| 295 |
+
"learning_rate": 2e-05,
|
| 296 |
+
"loss": 0.8979652404785157,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.022658610271903322,
|
| 301 |
+
"grad_norm": 2.6977548589169094,
|
| 302 |
+
"learning_rate": 2e-05,
|
| 303 |
+
"loss": 0.9128728866577148,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.023198100992662927,
|
| 308 |
+
"grad_norm": 23.582887550971233,
|
| 309 |
+
"learning_rate": 2e-05,
|
| 310 |
+
"loss": 0.8891826629638672,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.023737591713422528,
|
| 315 |
+
"grad_norm": 2.9436146384915443,
|
| 316 |
+
"learning_rate": 2e-05,
|
| 317 |
+
"loss": 0.9091741561889648,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.024277082434182133,
|
| 322 |
+
"grad_norm": 3.8178614247451996,
|
| 323 |
+
"learning_rate": 2e-05,
|
| 324 |
+
"loss": 0.9071731567382812,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.024816573154941735,
|
| 329 |
+
"grad_norm": 9.900017833351434,
|
| 330 |
+
"learning_rate": 2e-05,
|
| 331 |
+
"loss": 0.8803417205810546,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.02535606387570134,
|
| 336 |
+
"grad_norm": 11.377800042408293,
|
| 337 |
+
"learning_rate": 2e-05,
|
| 338 |
+
"loss": 0.9077526092529297,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.02589555459646094,
|
| 343 |
+
"grad_norm": 4.425640907224202,
|
| 344 |
+
"learning_rate": 2e-05,
|
| 345 |
+
"loss": 0.8951313018798828,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.026435045317220542,
|
| 350 |
+
"grad_norm": 5.543696039922662,
|
| 351 |
+
"learning_rate": 2e-05,
|
| 352 |
+
"loss": 0.9170653343200683,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.026974536037980147,
|
| 357 |
+
"grad_norm": 6.637675684831988,
|
| 358 |
+
"learning_rate": 2e-05,
|
| 359 |
+
"loss": 0.9328924179077148,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.02751402675873975,
|
| 364 |
+
"grad_norm": 4.541587002846055,
|
| 365 |
+
"learning_rate": 2e-05,
|
| 366 |
+
"loss": 0.9184404373168945,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.028053517479499353,
|
| 371 |
+
"grad_norm": 4.834955555833834,
|
| 372 |
+
"learning_rate": 2e-05,
|
| 373 |
+
"loss": 0.94840087890625,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.028593008200258955,
|
| 378 |
+
"grad_norm": 2.7476196554801118,
|
| 379 |
+
"learning_rate": 2e-05,
|
| 380 |
+
"loss": 0.9112020492553711,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.02913249892101856,
|
| 385 |
+
"grad_norm": 3.729222437612153,
|
| 386 |
+
"learning_rate": 2e-05,
|
| 387 |
+
"loss": 0.8715265274047852,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.02967198964177816,
|
| 392 |
+
"grad_norm": 1.8986444153489899,
|
| 393 |
+
"learning_rate": 2e-05,
|
| 394 |
+
"loss": 0.8817804336547852,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.030211480362537766,
|
| 399 |
+
"grad_norm": 19.652863197507482,
|
| 400 |
+
"learning_rate": 2e-05,
|
| 401 |
+
"loss": 0.9034578323364257,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.030750971083297367,
|
| 406 |
+
"grad_norm": 4.893935787998432,
|
| 407 |
+
"learning_rate": 2e-05,
|
| 408 |
+
"loss": 0.9201476097106933,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.03129046180405697,
|
| 413 |
+
"grad_norm": 2.121315677688142,
|
| 414 |
+
"learning_rate": 2e-05,
|
| 415 |
+
"loss": 0.9087504386901856,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.031829952524816574,
|
| 420 |
+
"grad_norm": 1.6805303205822455,
|
| 421 |
+
"learning_rate": 2e-05,
|
| 422 |
+
"loss": 0.8938014030456543,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.032369443245576175,
|
| 427 |
+
"grad_norm": 2.091016561123043,
|
| 428 |
+
"learning_rate": 2e-05,
|
| 429 |
+
"loss": 0.8701451301574707,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.03290893396633578,
|
| 434 |
+
"grad_norm": 4.4795444497420345,
|
| 435 |
+
"learning_rate": 2e-05,
|
| 436 |
+
"loss": 0.8988707542419434,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.033448424687095385,
|
| 441 |
+
"grad_norm": 1.40755956780777,
|
| 442 |
+
"learning_rate": 2e-05,
|
| 443 |
+
"loss": 0.8736795425415039,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.033987915407854986,
|
| 448 |
+
"grad_norm": 1.9677540233275401,
|
| 449 |
+
"learning_rate": 2e-05,
|
| 450 |
+
"loss": 0.8994430541992188,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.03452740612861459,
|
| 455 |
+
"grad_norm": 2.8537882081306605,
|
| 456 |
+
"learning_rate": 2e-05,
|
| 457 |
+
"loss": 0.8869138717651367,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.03506689684937419,
|
| 462 |
+
"grad_norm": 2.3406412447640133,
|
| 463 |
+
"learning_rate": 2e-05,
|
| 464 |
+
"loss": 0.8736772537231445,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.03560638757013379,
|
| 469 |
+
"grad_norm": 2.7002297840083647,
|
| 470 |
+
"learning_rate": 2e-05,
|
| 471 |
+
"loss": 0.8878070831298828,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.0361458782908934,
|
| 476 |
+
"grad_norm": 5.478051955992026,
|
| 477 |
+
"learning_rate": 2e-05,
|
| 478 |
+
"loss": 0.9007582664489746,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.036685369011653,
|
| 483 |
+
"grad_norm": 5.987775578117277,
|
| 484 |
+
"learning_rate": 2e-05,
|
| 485 |
+
"loss": 0.877619457244873,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.0372248597324126,
|
| 490 |
+
"grad_norm": 4.062183705280081,
|
| 491 |
+
"learning_rate": 2e-05,
|
| 492 |
+
"loss": 0.9615988731384277,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.0377643504531722,
|
| 497 |
+
"grad_norm": 1.7304116921547956,
|
| 498 |
+
"learning_rate": 2e-05,
|
| 499 |
+
"loss": 0.8999065399169922,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.03830384117393181,
|
| 504 |
+
"grad_norm": 10.224091687714415,
|
| 505 |
+
"learning_rate": 2e-05,
|
| 506 |
+
"loss": 0.8775262832641602,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.03884333189469141,
|
| 511 |
+
"grad_norm": 2.6973024084894885,
|
| 512 |
+
"learning_rate": 2e-05,
|
| 513 |
+
"loss": 0.88863525390625,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.039382822615451014,
|
| 518 |
+
"grad_norm": 3.0097373941657186,
|
| 519 |
+
"learning_rate": 2e-05,
|
| 520 |
+
"loss": 0.8921207427978516,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.039922313336210616,
|
| 525 |
+
"grad_norm": 2.4929780841655207,
|
| 526 |
+
"learning_rate": 2e-05,
|
| 527 |
+
"loss": 0.8870817184448242,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.04046180405697022,
|
| 532 |
+
"grad_norm": 2.680211035148592,
|
| 533 |
+
"learning_rate": 2e-05,
|
| 534 |
+
"loss": 0.8718931198120117,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.041001294777729826,
|
| 539 |
+
"grad_norm": 3.314725765838302,
|
| 540 |
+
"learning_rate": 2e-05,
|
| 541 |
+
"loss": 0.8696352005004883,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.04154078549848943,
|
| 546 |
+
"grad_norm": 2.8067410842808824,
|
| 547 |
+
"learning_rate": 2e-05,
|
| 548 |
+
"loss": 0.8808348655700684,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.04208027621924903,
|
| 553 |
+
"grad_norm": 4.568417784852867,
|
| 554 |
+
"learning_rate": 2e-05,
|
| 555 |
+
"loss": 0.8762812614440918,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.04261976694000863,
|
| 560 |
+
"grad_norm": 1.591466243191371,
|
| 561 |
+
"learning_rate": 2e-05,
|
| 562 |
+
"loss": 0.8897010803222656,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.04315925766076824,
|
| 567 |
+
"grad_norm": 1.7378624546635357,
|
| 568 |
+
"learning_rate": 2e-05,
|
| 569 |
+
"loss": 0.8871335983276367,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.04369874838152784,
|
| 574 |
+
"grad_norm": 5.282884941300159,
|
| 575 |
+
"learning_rate": 2e-05,
|
| 576 |
+
"loss": 0.8862749099731445,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.04423823910228744,
|
| 581 |
+
"grad_norm": 1.4282666163813993,
|
| 582 |
+
"learning_rate": 2e-05,
|
| 583 |
+
"loss": 0.9045479774475098,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.04477772982304704,
|
| 588 |
+
"grad_norm": 1.804678546954577,
|
| 589 |
+
"learning_rate": 2e-05,
|
| 590 |
+
"loss": 0.8669574737548829,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.045317220543806644,
|
| 595 |
+
"grad_norm": 2.7003745816690836,
|
| 596 |
+
"learning_rate": 2e-05,
|
| 597 |
+
"loss": 0.8664716720581055,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.04585671126456625,
|
| 602 |
+
"grad_norm": 2.8347718319860626,
|
| 603 |
+
"learning_rate": 2e-05,
|
| 604 |
+
"loss": 0.8740848541259766,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.046396201985325854,
|
| 609 |
+
"grad_norm": 2.3599301199842655,
|
| 610 |
+
"learning_rate": 2e-05,
|
| 611 |
+
"loss": 0.8661260604858398,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.046935692706085455,
|
| 616 |
+
"grad_norm": 2.675130769028597,
|
| 617 |
+
"learning_rate": 2e-05,
|
| 618 |
+
"loss": 0.8940586090087891,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.047475183426845056,
|
| 623 |
+
"grad_norm": 2.13219637159858,
|
| 624 |
+
"learning_rate": 2e-05,
|
| 625 |
+
"loss": 0.8656101226806641,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.04801467414760466,
|
| 630 |
+
"grad_norm": 5.746935578097955,
|
| 631 |
+
"learning_rate": 2e-05,
|
| 632 |
+
"loss": 0.9049406051635742,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.048554164868364266,
|
| 637 |
+
"grad_norm": 3.040442087390683,
|
| 638 |
+
"learning_rate": 2e-05,
|
| 639 |
+
"loss": 0.8673805236816406,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.04909365558912387,
|
| 644 |
+
"grad_norm": 1.8033864790060448,
|
| 645 |
+
"learning_rate": 2e-05,
|
| 646 |
+
"loss": 0.9076686859130859,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.04963314630988347,
|
| 651 |
+
"grad_norm": 3.7270371618478446,
|
| 652 |
+
"learning_rate": 2e-05,
|
| 653 |
+
"loss": 0.8856944084167481,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.05017263703064307,
|
| 658 |
+
"grad_norm": 18.34533608060895,
|
| 659 |
+
"learning_rate": 2e-05,
|
| 660 |
+
"loss": 0.8803435325622558,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.05071212775140268,
|
| 665 |
+
"grad_norm": 7.031326732375769,
|
| 666 |
+
"learning_rate": 2e-05,
|
| 667 |
+
"loss": 0.8922122955322266,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.05125161847216228,
|
| 672 |
+
"grad_norm": 1.271585738790212,
|
| 673 |
+
"learning_rate": 2e-05,
|
| 674 |
+
"loss": 0.8702299118041992,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.05179110919292188,
|
| 679 |
+
"grad_norm": 3.1087560610639984,
|
| 680 |
+
"learning_rate": 2e-05,
|
| 681 |
+
"loss": 0.869837474822998,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.05233059991368148,
|
| 686 |
+
"grad_norm": 1.336885754811761,
|
| 687 |
+
"learning_rate": 2e-05,
|
| 688 |
+
"loss": 0.8768095016479492,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.052870090634441085,
|
| 693 |
+
"grad_norm": 1.2937187665672654,
|
| 694 |
+
"learning_rate": 2e-05,
|
| 695 |
+
"loss": 0.8737648010253907,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.05340958135520069,
|
| 700 |
+
"grad_norm": 6.496337599049736,
|
| 701 |
+
"learning_rate": 2e-05,
|
| 702 |
+
"loss": 0.8852460861206055,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.053949072075960294,
|
| 707 |
+
"grad_norm": 2.314909026977299,
|
| 708 |
+
"learning_rate": 2e-05,
|
| 709 |
+
"loss": 0.9168342590332031,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.054488562796719896,
|
| 714 |
+
"grad_norm": 1.3797589526126182,
|
| 715 |
+
"learning_rate": 2e-05,
|
| 716 |
+
"loss": 0.9012920379638671,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.0550280535174795,
|
| 721 |
+
"grad_norm": 1.7456393044082175,
|
| 722 |
+
"learning_rate": 2e-05,
|
| 723 |
+
"loss": 0.8839190483093262,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.055567544238239105,
|
| 728 |
+
"grad_norm": 2.2368924316262913,
|
| 729 |
+
"learning_rate": 2e-05,
|
| 730 |
+
"loss": 0.8841041564941406,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.05610703495899871,
|
| 735 |
+
"grad_norm": 2.3060976347623408,
|
| 736 |
+
"learning_rate": 2e-05,
|
| 737 |
+
"loss": 0.9180487632751465,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.05664652567975831,
|
| 742 |
+
"grad_norm": 2.2473033744197655,
|
| 743 |
+
"learning_rate": 2e-05,
|
| 744 |
+
"loss": 0.8983830451965332,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.05718601640051791,
|
| 749 |
+
"grad_norm": 3.33794913543599,
|
| 750 |
+
"learning_rate": 2e-05,
|
| 751 |
+
"loss": 0.9058303833007812,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.05772550712127751,
|
| 756 |
+
"grad_norm": 2.1863704996804874,
|
| 757 |
+
"learning_rate": 2e-05,
|
| 758 |
+
"loss": 0.8500079154968262,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.05826499784203712,
|
| 763 |
+
"grad_norm": 5.464104262184261,
|
| 764 |
+
"learning_rate": 2e-05,
|
| 765 |
+
"loss": 0.86676025390625,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.05880448856279672,
|
| 770 |
+
"grad_norm": 1.2580865472126652,
|
| 771 |
+
"learning_rate": 2e-05,
|
| 772 |
+
"loss": 0.8795917510986329,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.05934397928355632,
|
| 777 |
+
"grad_norm": 1.5392878698728925,
|
| 778 |
+
"learning_rate": 2e-05,
|
| 779 |
+
"loss": 0.855427360534668,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.059883470004315924,
|
| 784 |
+
"grad_norm": 1.6600456111757242,
|
| 785 |
+
"learning_rate": 2e-05,
|
| 786 |
+
"loss": 0.8580084800720215,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.06042296072507553,
|
| 791 |
+
"grad_norm": 1.9728666357091431,
|
| 792 |
+
"learning_rate": 2e-05,
|
| 793 |
+
"loss": 0.8592136383056641,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.060962451445835134,
|
| 798 |
+
"grad_norm": 2.622532245761096,
|
| 799 |
+
"learning_rate": 2e-05,
|
| 800 |
+
"loss": 0.8778785705566406,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.061501942166594735,
|
| 805 |
+
"grad_norm": 2.436300801764917,
|
| 806 |
+
"learning_rate": 2e-05,
|
| 807 |
+
"loss": 0.8558690071105957,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.062041432887354336,
|
| 812 |
+
"grad_norm": 2.072486307736441,
|
| 813 |
+
"learning_rate": 2e-05,
|
| 814 |
+
"loss": 0.8554096221923828,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.06258092360811394,
|
| 819 |
+
"grad_norm": 2.9578754600495105,
|
| 820 |
+
"learning_rate": 2e-05,
|
| 821 |
+
"loss": 0.8688982963562012,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.06312041432887354,
|
| 826 |
+
"grad_norm": 2.2185001887709106,
|
| 827 |
+
"learning_rate": 2e-05,
|
| 828 |
+
"loss": 0.8960395812988281,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.06365990504963315,
|
| 833 |
+
"grad_norm": 2.39317741084652,
|
| 834 |
+
"learning_rate": 2e-05,
|
| 835 |
+
"loss": 0.8809322357177735,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.06419939577039276,
|
| 840 |
+
"grad_norm": 7.053593954669792,
|
| 841 |
+
"learning_rate": 2e-05,
|
| 842 |
+
"loss": 0.8588828086853028,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.06473888649115235,
|
| 847 |
+
"grad_norm": 2.8316276030407392,
|
| 848 |
+
"learning_rate": 2e-05,
|
| 849 |
+
"loss": 0.863832664489746,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.06527837721191196,
|
| 854 |
+
"grad_norm": 55.263846840104094,
|
| 855 |
+
"learning_rate": 2e-05,
|
| 856 |
+
"loss": 0.8467828750610351,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.06581786793267155,
|
| 861 |
+
"grad_norm": 1.891195934618861,
|
| 862 |
+
"learning_rate": 2e-05,
|
| 863 |
+
"loss": 0.8810437202453614,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.06635735865343116,
|
| 868 |
+
"grad_norm": 4.262649131447584,
|
| 869 |
+
"learning_rate": 2e-05,
|
| 870 |
+
"loss": 0.8573444366455079,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.06689684937419077,
|
| 875 |
+
"grad_norm": 5.280398859967369,
|
| 876 |
+
"learning_rate": 2e-05,
|
| 877 |
+
"loss": 0.8775759696960449,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.06743634009495036,
|
| 882 |
+
"grad_norm": 13.161950494331434,
|
| 883 |
+
"learning_rate": 2e-05,
|
| 884 |
+
"loss": 0.8802067756652832,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.06797583081570997,
|
| 889 |
+
"grad_norm": 2.3840227633253708,
|
| 890 |
+
"learning_rate": 2e-05,
|
| 891 |
+
"loss": 0.8745565414428711,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.06851532153646957,
|
| 896 |
+
"grad_norm": 2.578881266139901,
|
| 897 |
+
"learning_rate": 2e-05,
|
| 898 |
+
"loss": 0.8929662704467773,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.06905481225722918,
|
| 903 |
+
"grad_norm": 1.3551093714665399,
|
| 904 |
+
"learning_rate": 2e-05,
|
| 905 |
+
"loss": 0.8856276512145996,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.06959430297798878,
|
| 910 |
+
"grad_norm": 1.421718866393892,
|
| 911 |
+
"learning_rate": 2e-05,
|
| 912 |
+
"loss": 0.8779384613037109,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.07013379369874838,
|
| 917 |
+
"grad_norm": 1.3412253586411367,
|
| 918 |
+
"learning_rate": 2e-05,
|
| 919 |
+
"loss": 0.8418590545654296,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.07067328441950799,
|
| 924 |
+
"grad_norm": 1.3164576570437523,
|
| 925 |
+
"learning_rate": 2e-05,
|
| 926 |
+
"loss": 0.8589597702026367,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.07121277514026758,
|
| 931 |
+
"grad_norm": 5.63391297375713,
|
| 932 |
+
"learning_rate": 2e-05,
|
| 933 |
+
"loss": 0.8794595718383789,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.07175226586102719,
|
| 938 |
+
"grad_norm": 1.3610004564150833,
|
| 939 |
+
"learning_rate": 2e-05,
|
| 940 |
+
"loss": 0.8666675567626954,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.0722917565817868,
|
| 945 |
+
"grad_norm": 1.7200672230946694,
|
| 946 |
+
"learning_rate": 2e-05,
|
| 947 |
+
"loss": 0.8707124710083007,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.07283124730254639,
|
| 952 |
+
"grad_norm": 1.3160601308789588,
|
| 953 |
+
"learning_rate": 2e-05,
|
| 954 |
+
"loss": 0.8457590103149414,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.073370738023306,
|
| 959 |
+
"grad_norm": 1.7778775450463142,
|
| 960 |
+
"learning_rate": 2e-05,
|
| 961 |
+
"loss": 0.8669673919677734,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.0739102287440656,
|
| 966 |
+
"grad_norm": 2.2742353103065485,
|
| 967 |
+
"learning_rate": 2e-05,
|
| 968 |
+
"loss": 0.920014762878418,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.0744497194648252,
|
| 973 |
+
"grad_norm": 4.185305052537614,
|
| 974 |
+
"learning_rate": 2e-05,
|
| 975 |
+
"loss": 0.9066905021667481,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.07498921018558481,
|
| 980 |
+
"grad_norm": 7.045183411349042,
|
| 981 |
+
"learning_rate": 2e-05,
|
| 982 |
+
"loss": 0.8720160484313965,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.0755287009063444,
|
| 987 |
+
"grad_norm": 2.377204425821727,
|
| 988 |
+
"learning_rate": 2e-05,
|
| 989 |
+
"loss": 0.9035792350769043,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.07606819162710401,
|
| 994 |
+
"grad_norm": 1.5514726466038995,
|
| 995 |
+
"learning_rate": 2e-05,
|
| 996 |
+
"loss": 0.8550067901611328,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.07660768234786362,
|
| 1001 |
+
"grad_norm": 3.2194246326766627,
|
| 1002 |
+
"learning_rate": 2e-05,
|
| 1003 |
+
"loss": 0.8613847732543946,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.07714717306862322,
|
| 1008 |
+
"grad_norm": 2.191413489389274,
|
| 1009 |
+
"learning_rate": 2e-05,
|
| 1010 |
+
"loss": 0.8483789443969727,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.07768666378938283,
|
| 1015 |
+
"grad_norm": 2.561083258333701,
|
| 1016 |
+
"learning_rate": 2e-05,
|
| 1017 |
+
"loss": 0.8561199188232422,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.07822615451014242,
|
| 1022 |
+
"grad_norm": 2.799486266864444,
|
| 1023 |
+
"learning_rate": 2e-05,
|
| 1024 |
+
"loss": 0.8749614715576172,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.07876564523090203,
|
| 1029 |
+
"grad_norm": 1.6625068228812598,
|
| 1030 |
+
"learning_rate": 2e-05,
|
| 1031 |
+
"loss": 0.8524051666259765,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.07930513595166164,
|
| 1036 |
+
"grad_norm": 1.5627273498445615,
|
| 1037 |
+
"learning_rate": 2e-05,
|
| 1038 |
+
"loss": 0.8613437652587891,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.07984462667242123,
|
| 1043 |
+
"grad_norm": 3.1187718040693815,
|
| 1044 |
+
"learning_rate": 2e-05,
|
| 1045 |
+
"loss": 0.8489472389221191,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.08038411739318084,
|
| 1050 |
+
"grad_norm": 1.999242227430214,
|
| 1051 |
+
"learning_rate": 2e-05,
|
| 1052 |
+
"loss": 0.8557394027709961,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.08092360811394043,
|
| 1057 |
+
"grad_norm": 2.1662789248263086,
|
| 1058 |
+
"learning_rate": 2e-05,
|
| 1059 |
+
"loss": 0.8410913467407226,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.08146309883470004,
|
| 1064 |
+
"grad_norm": 1.117593071600269,
|
| 1065 |
+
"learning_rate": 2e-05,
|
| 1066 |
+
"loss": 0.8719884872436523,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.08200258955545965,
|
| 1071 |
+
"grad_norm": 1.7282463420500256,
|
| 1072 |
+
"learning_rate": 2e-05,
|
| 1073 |
+
"loss": 0.8429347991943359,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.08254208027621925,
|
| 1078 |
+
"grad_norm": 1.10160222981424,
|
| 1079 |
+
"learning_rate": 2e-05,
|
| 1080 |
+
"loss": 0.8589319229125977,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.08308157099697885,
|
| 1085 |
+
"grad_norm": 1.657648568922712,
|
| 1086 |
+
"learning_rate": 2e-05,
|
| 1087 |
+
"loss": 0.8519350051879883,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.08362106171773845,
|
| 1092 |
+
"grad_norm": 1.2061011251185034,
|
| 1093 |
+
"learning_rate": 2e-05,
|
| 1094 |
+
"loss": 0.8548383712768555,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.08416055243849806,
|
| 1099 |
+
"grad_norm": 1.6503845091419074,
|
| 1100 |
+
"learning_rate": 2e-05,
|
| 1101 |
+
"loss": 0.8504886627197266,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.08470004315925767,
|
| 1106 |
+
"grad_norm": 1.650450748636652,
|
| 1107 |
+
"learning_rate": 2e-05,
|
| 1108 |
+
"loss": 0.8272921562194824,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.08523953388001726,
|
| 1113 |
+
"grad_norm": 1.2153538779996496,
|
| 1114 |
+
"learning_rate": 2e-05,
|
| 1115 |
+
"loss": 0.8586090087890625,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.08577902460077687,
|
| 1120 |
+
"grad_norm": 1.4119654167230689,
|
| 1121 |
+
"learning_rate": 2e-05,
|
| 1122 |
+
"loss": 0.8150126457214355,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.08631851532153648,
|
| 1127 |
+
"grad_norm": 2.0481812733972435,
|
| 1128 |
+
"learning_rate": 2e-05,
|
| 1129 |
+
"loss": 0.8605268478393555,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.08685800604229607,
|
| 1134 |
+
"grad_norm": 1.3493621359452488,
|
| 1135 |
+
"learning_rate": 2e-05,
|
| 1136 |
+
"loss": 0.8761652946472168,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.08739749676305568,
|
| 1141 |
+
"grad_norm": 1.937128776151837,
|
| 1142 |
+
"learning_rate": 2e-05,
|
| 1143 |
+
"loss": 0.8717695236206054,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.08793698748381527,
|
| 1148 |
+
"grad_norm": 1.2331164884742856,
|
| 1149 |
+
"learning_rate": 2e-05,
|
| 1150 |
+
"loss": 0.8695238113403321,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.08847647820457488,
|
| 1155 |
+
"grad_norm": 1.0766868462241113,
|
| 1156 |
+
"learning_rate": 2e-05,
|
| 1157 |
+
"loss": 0.8439886093139648,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.08901596892533449,
|
| 1162 |
+
"grad_norm": 4.825188821823706,
|
| 1163 |
+
"learning_rate": 2e-05,
|
| 1164 |
+
"loss": 0.8383602142333985,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.08955545964609408,
|
| 1169 |
+
"grad_norm": 1.0473009888263762,
|
| 1170 |
+
"learning_rate": 2e-05,
|
| 1171 |
+
"loss": 0.8446158409118653,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.0900949503668537,
|
| 1176 |
+
"grad_norm": 2.1550692634060407,
|
| 1177 |
+
"learning_rate": 2e-05,
|
| 1178 |
+
"loss": 0.8526595115661622,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.09063444108761329,
|
| 1183 |
+
"grad_norm": 1.1805712525240586,
|
| 1184 |
+
"learning_rate": 2e-05,
|
| 1185 |
+
"loss": 0.8445590019226075,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.0911739318083729,
|
| 1190 |
+
"grad_norm": 1.3154682778410502,
|
| 1191 |
+
"learning_rate": 2e-05,
|
| 1192 |
+
"loss": 0.8289753913879394,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.0917134225291325,
|
| 1197 |
+
"grad_norm": 1.7425671016352309,
|
| 1198 |
+
"learning_rate": 2e-05,
|
| 1199 |
+
"loss": 0.8550616264343261,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.0922529132498921,
|
| 1204 |
+
"grad_norm": 1.809326593705475,
|
| 1205 |
+
"learning_rate": 2e-05,
|
| 1206 |
+
"loss": 0.8592042922973633,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.09279240397065171,
|
| 1211 |
+
"grad_norm": 2.294726987915818,
|
| 1212 |
+
"learning_rate": 2e-05,
|
| 1213 |
+
"loss": 0.8464397430419922,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.0933318946914113,
|
| 1218 |
+
"grad_norm": 1.3341335707857838,
|
| 1219 |
+
"learning_rate": 2e-05,
|
| 1220 |
+
"loss": 0.8401545524597168,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.09387138541217091,
|
| 1225 |
+
"grad_norm": 1.7059856939535623,
|
| 1226 |
+
"learning_rate": 2e-05,
|
| 1227 |
+
"loss": 0.850313663482666,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.09441087613293052,
|
| 1232 |
+
"grad_norm": 1.1913305227905306,
|
| 1233 |
+
"learning_rate": 2e-05,
|
| 1234 |
+
"loss": 0.8246526718139648,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.09495036685369011,
|
| 1239 |
+
"grad_norm": 2.4712263811656063,
|
| 1240 |
+
"learning_rate": 2e-05,
|
| 1241 |
+
"loss": 0.8324945449829102,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.09548985757444972,
|
| 1246 |
+
"grad_norm": 1.2460535060628168,
|
| 1247 |
+
"learning_rate": 2e-05,
|
| 1248 |
+
"loss": 0.83499755859375,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.09602934829520932,
|
| 1253 |
+
"grad_norm": 0.9541283688175325,
|
| 1254 |
+
"learning_rate": 2e-05,
|
| 1255 |
+
"loss": 0.8438259124755859,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.09656883901596892,
|
| 1260 |
+
"grad_norm": 4.15270429876545,
|
| 1261 |
+
"learning_rate": 2e-05,
|
| 1262 |
+
"loss": 0.8374526977539063,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.09710832973672853,
|
| 1267 |
+
"grad_norm": 1.2843301784727401,
|
| 1268 |
+
"learning_rate": 2e-05,
|
| 1269 |
+
"loss": 0.855588722229004,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.09764782045748813,
|
| 1274 |
+
"grad_norm": 1.2852787584009888,
|
| 1275 |
+
"learning_rate": 2e-05,
|
| 1276 |
+
"loss": 0.8446294784545898,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.09818731117824774,
|
| 1281 |
+
"grad_norm": 1.1732513743084148,
|
| 1282 |
+
"learning_rate": 2e-05,
|
| 1283 |
+
"loss": 0.8550775527954102,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.09872680189900734,
|
| 1288 |
+
"grad_norm": 1.0484004988454474,
|
| 1289 |
+
"learning_rate": 2e-05,
|
| 1290 |
+
"loss": 0.8461496353149414,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.09926629261976694,
|
| 1295 |
+
"grad_norm": 1.1796374157652922,
|
| 1296 |
+
"learning_rate": 2e-05,
|
| 1297 |
+
"loss": 0.8495699882507324,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.09980578334052655,
|
| 1302 |
+
"grad_norm": 1.530009834654909,
|
| 1303 |
+
"learning_rate": 2e-05,
|
| 1304 |
+
"loss": 0.8529845237731933,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.10034527406128614,
|
| 1309 |
+
"grad_norm": 1.3821515513567662,
|
| 1310 |
+
"learning_rate": 2e-05,
|
| 1311 |
+
"loss": 0.8383827209472656,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.10088476478204575,
|
| 1316 |
+
"grad_norm": 1.433339075396923,
|
| 1317 |
+
"learning_rate": 2e-05,
|
| 1318 |
+
"loss": 0.8441427230834961,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.10142425550280536,
|
| 1323 |
+
"grad_norm": 1.4202627926244262,
|
| 1324 |
+
"learning_rate": 2e-05,
|
| 1325 |
+
"loss": 0.8355476379394531,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.10196374622356495,
|
| 1330 |
+
"grad_norm": 1.275529436709381,
|
| 1331 |
+
"learning_rate": 2e-05,
|
| 1332 |
+
"loss": 0.8475621223449707,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.10250323694432456,
|
| 1337 |
+
"grad_norm": 4.886608266675648,
|
| 1338 |
+
"learning_rate": 2e-05,
|
| 1339 |
+
"loss": 0.8222907066345215,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.10304272766508416,
|
| 1344 |
+
"grad_norm": 1.1032859405565207,
|
| 1345 |
+
"learning_rate": 2e-05,
|
| 1346 |
+
"loss": 0.8717263221740723,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.10358221838584376,
|
| 1351 |
+
"grad_norm": 1.4792708796316567,
|
| 1352 |
+
"learning_rate": 2e-05,
|
| 1353 |
+
"loss": 0.8422025680541992,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.10412170910660337,
|
| 1358 |
+
"grad_norm": 1.052829235330944,
|
| 1359 |
+
"learning_rate": 2e-05,
|
| 1360 |
+
"loss": 0.8275224685668945,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.10466119982736297,
|
| 1365 |
+
"grad_norm": 1.5243538008576782,
|
| 1366 |
+
"learning_rate": 2e-05,
|
| 1367 |
+
"loss": 0.8201045989990234,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.10520069054812257,
|
| 1372 |
+
"grad_norm": 1.8549508993672652,
|
| 1373 |
+
"learning_rate": 2e-05,
|
| 1374 |
+
"loss": 0.8507599830627441,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.10574018126888217,
|
| 1379 |
+
"grad_norm": 3.2676775863279293,
|
| 1380 |
+
"learning_rate": 2e-05,
|
| 1381 |
+
"loss": 0.8315736770629882,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.10627967198964178,
|
| 1386 |
+
"grad_norm": 3.170505632566329,
|
| 1387 |
+
"learning_rate": 2e-05,
|
| 1388 |
+
"loss": 0.8597011566162109,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.10681916271040139,
|
| 1393 |
+
"grad_norm": 1.225518275743383,
|
| 1394 |
+
"learning_rate": 2e-05,
|
| 1395 |
+
"loss": 0.8449475288391113,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.10735865343116098,
|
| 1400 |
+
"grad_norm": 1.7639666202458668,
|
| 1401 |
+
"learning_rate": 2e-05,
|
| 1402 |
+
"loss": 0.8267492294311524,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.10789814415192059,
|
| 1407 |
+
"grad_norm": 1.7820697092202835,
|
| 1408 |
+
"learning_rate": 2e-05,
|
| 1409 |
+
"loss": 0.8312843322753907,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.10843763487268018,
|
| 1414 |
+
"grad_norm": 4.015953726833389,
|
| 1415 |
+
"learning_rate": 2e-05,
|
| 1416 |
+
"loss": 0.8443691253662109,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.10897712559343979,
|
| 1421 |
+
"grad_norm": 1.999195946930485,
|
| 1422 |
+
"learning_rate": 2e-05,
|
| 1423 |
+
"loss": 0.8289900779724121,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.1095166163141994,
|
| 1428 |
+
"grad_norm": 1.8651265962731969,
|
| 1429 |
+
"learning_rate": 2e-05,
|
| 1430 |
+
"loss": 0.840304946899414,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.110056107034959,
|
| 1435 |
+
"grad_norm": 1.7034465477351663,
|
| 1436 |
+
"learning_rate": 2e-05,
|
| 1437 |
+
"loss": 0.8282983779907227,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.1105955977557186,
|
| 1442 |
+
"grad_norm": 1.292874343679415,
|
| 1443 |
+
"learning_rate": 2e-05,
|
| 1444 |
+
"loss": 0.8187969207763672,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.11113508847647821,
|
| 1449 |
+
"grad_norm": 9.900066428760711,
|
| 1450 |
+
"learning_rate": 2e-05,
|
| 1451 |
+
"loss": 0.8275846481323242,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.1116745791972378,
|
| 1456 |
+
"grad_norm": 1.4024064010510686,
|
| 1457 |
+
"learning_rate": 2e-05,
|
| 1458 |
+
"loss": 0.8306047439575195,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.11221406991799741,
|
| 1463 |
+
"grad_norm": 2.400290796313511,
|
| 1464 |
+
"learning_rate": 2e-05,
|
| 1465 |
+
"loss": 0.8676346778869629,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.11275356063875701,
|
| 1470 |
+
"grad_norm": 3.514349150867208,
|
| 1471 |
+
"learning_rate": 2e-05,
|
| 1472 |
+
"loss": 0.8274039268493653,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.11329305135951662,
|
| 1477 |
+
"grad_norm": 5.088709612171823,
|
| 1478 |
+
"learning_rate": 2e-05,
|
| 1479 |
+
"loss": 0.8288514137268066,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.11383254208027623,
|
| 1484 |
+
"grad_norm": 2.098725303443767,
|
| 1485 |
+
"learning_rate": 2e-05,
|
| 1486 |
+
"loss": 0.8381128311157227,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.11437203280103582,
|
| 1491 |
+
"grad_norm": 1.3305492466588589,
|
| 1492 |
+
"learning_rate": 2e-05,
|
| 1493 |
+
"loss": 0.8685014724731446,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.11491152352179543,
|
| 1498 |
+
"grad_norm": 2.7110556064904956,
|
| 1499 |
+
"learning_rate": 2e-05,
|
| 1500 |
+
"loss": 0.8481921195983887,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.11545101424255502,
|
| 1505 |
+
"grad_norm": 1.10367668589633,
|
| 1506 |
+
"learning_rate": 2e-05,
|
| 1507 |
+
"loss": 0.8573419570922851,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.11599050496331463,
|
| 1512 |
+
"grad_norm": 2.138417581209623,
|
| 1513 |
+
"learning_rate": 2e-05,
|
| 1514 |
+
"loss": 0.8302045822143554,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.11652999568407424,
|
| 1519 |
+
"grad_norm": 1.1112744471980331,
|
| 1520 |
+
"learning_rate": 2e-05,
|
| 1521 |
+
"loss": 0.8130320549011231,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.11706948640483383,
|
| 1526 |
+
"grad_norm": 1.6334189344204764,
|
| 1527 |
+
"learning_rate": 2e-05,
|
| 1528 |
+
"loss": 0.8478602409362793,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.11760897712559344,
|
| 1533 |
+
"grad_norm": 2.7583960750005816,
|
| 1534 |
+
"learning_rate": 2e-05,
|
| 1535 |
+
"loss": 0.8246549606323242,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.11814846784635304,
|
| 1540 |
+
"grad_norm": 1.9058948199400891,
|
| 1541 |
+
"learning_rate": 2e-05,
|
| 1542 |
+
"loss": 0.8250743865966796,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.11868795856711264,
|
| 1547 |
+
"grad_norm": 1.586316165202703,
|
| 1548 |
+
"learning_rate": 2e-05,
|
| 1549 |
+
"loss": 0.8306868553161622,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.11922744928787225,
|
| 1554 |
+
"grad_norm": 3.6562916703913606,
|
| 1555 |
+
"learning_rate": 2e-05,
|
| 1556 |
+
"loss": 0.8404013633728027,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.11976694000863185,
|
| 1561 |
+
"grad_norm": 1.914570415791936,
|
| 1562 |
+
"learning_rate": 2e-05,
|
| 1563 |
+
"loss": 0.8350730895996094,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.12030643072939146,
|
| 1568 |
+
"grad_norm": 1.4049455715505264,
|
| 1569 |
+
"learning_rate": 2e-05,
|
| 1570 |
+
"loss": 0.8396036148071289,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.12084592145015106,
|
| 1575 |
+
"grad_norm": 1.1836748795523662,
|
| 1576 |
+
"learning_rate": 2e-05,
|
| 1577 |
+
"loss": 0.8288493156433105,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.12138541217091066,
|
| 1582 |
+
"grad_norm": 2.5165436971018735,
|
| 1583 |
+
"learning_rate": 2e-05,
|
| 1584 |
+
"loss": 0.8394759178161622,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.12192490289167027,
|
| 1589 |
+
"grad_norm": 1.8023654123652022,
|
| 1590 |
+
"learning_rate": 2e-05,
|
| 1591 |
+
"loss": 0.8561839103698731,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.12246439361242986,
|
| 1596 |
+
"grad_norm": 2.197758047068318,
|
| 1597 |
+
"learning_rate": 2e-05,
|
| 1598 |
+
"loss": 0.8342147827148437,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.12300388433318947,
|
| 1603 |
+
"grad_norm": 1.3512470884395074,
|
| 1604 |
+
"learning_rate": 2e-05,
|
| 1605 |
+
"loss": 0.8416089057922364,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.12354337505394908,
|
| 1610 |
+
"grad_norm": 1.158221169858105,
|
| 1611 |
+
"learning_rate": 2e-05,
|
| 1612 |
+
"loss": 0.815174674987793,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.12408286577470867,
|
| 1617 |
+
"grad_norm": 2.1931315632737336,
|
| 1618 |
+
"learning_rate": 2e-05,
|
| 1619 |
+
"loss": 0.8198381423950195,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.12462235649546828,
|
| 1624 |
+
"grad_norm": 1.2020698197548285,
|
| 1625 |
+
"learning_rate": 2e-05,
|
| 1626 |
+
"loss": 0.8212334632873535,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.1251618472162279,
|
| 1631 |
+
"grad_norm": 1.2630839020259081,
|
| 1632 |
+
"learning_rate": 2e-05,
|
| 1633 |
+
"loss": 0.8234346389770508,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.12570133793698748,
|
| 1638 |
+
"grad_norm": 1.05835478339789,
|
| 1639 |
+
"learning_rate": 2e-05,
|
| 1640 |
+
"loss": 0.8266730308532715,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.12624082865774708,
|
| 1645 |
+
"grad_norm": 1.153807197212215,
|
| 1646 |
+
"learning_rate": 2e-05,
|
| 1647 |
+
"loss": 0.8297053337097168,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.1267803193785067,
|
| 1652 |
+
"grad_norm": 1.081701471063699,
|
| 1653 |
+
"learning_rate": 2e-05,
|
| 1654 |
+
"loss": 0.850483512878418,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.1273198100992663,
|
| 1659 |
+
"grad_norm": 1.1032276472226967,
|
| 1660 |
+
"learning_rate": 2e-05,
|
| 1661 |
+
"loss": 0.8351642608642578,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.1278593008200259,
|
| 1666 |
+
"grad_norm": 0.9373186180851497,
|
| 1667 |
+
"learning_rate": 2e-05,
|
| 1668 |
+
"loss": 0.850758171081543,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.1283987915407855,
|
| 1673 |
+
"grad_norm": 0.9185447727528806,
|
| 1674 |
+
"learning_rate": 2e-05,
|
| 1675 |
+
"loss": 0.8138019561767578,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.1289382822615451,
|
| 1680 |
+
"grad_norm": 1.4547050261067835,
|
| 1681 |
+
"learning_rate": 2e-05,
|
| 1682 |
+
"loss": 0.8109475135803222,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.1294777729823047,
|
| 1687 |
+
"grad_norm": 0.8962680388570466,
|
| 1688 |
+
"learning_rate": 2e-05,
|
| 1689 |
+
"loss": 0.8358553886413574,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.1300172637030643,
|
| 1694 |
+
"grad_norm": 0.9967276931315883,
|
| 1695 |
+
"learning_rate": 2e-05,
|
| 1696 |
+
"loss": 0.8311929702758789,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.13055675442382392,
|
| 1701 |
+
"grad_norm": 1.171890528912994,
|
| 1702 |
+
"learning_rate": 2e-05,
|
| 1703 |
+
"loss": 0.8504758834838867,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.1310962451445835,
|
| 1708 |
+
"grad_norm": 0.9664067450616619,
|
| 1709 |
+
"learning_rate": 2e-05,
|
| 1710 |
+
"loss": 0.8218025207519531,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.1316357358653431,
|
| 1715 |
+
"grad_norm": 1.0236162825206507,
|
| 1716 |
+
"learning_rate": 2e-05,
|
| 1717 |
+
"loss": 0.8196083068847656,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.13217522658610273,
|
| 1722 |
+
"grad_norm": 1.9012584301013733,
|
| 1723 |
+
"learning_rate": 2e-05,
|
| 1724 |
+
"loss": 0.823670768737793,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.13271471730686232,
|
| 1729 |
+
"grad_norm": 1.1023811611206806,
|
| 1730 |
+
"learning_rate": 2e-05,
|
| 1731 |
+
"loss": 0.8224227905273438,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.13325420802762192,
|
| 1736 |
+
"grad_norm": 9.134961938530486,
|
| 1737 |
+
"learning_rate": 2e-05,
|
| 1738 |
+
"loss": 0.8160662651062012,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.13379369874838154,
|
| 1743 |
+
"grad_norm": 4.065113504605792,
|
| 1744 |
+
"learning_rate": 2e-05,
|
| 1745 |
+
"loss": 0.842041015625,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.13433318946914113,
|
| 1750 |
+
"grad_norm": 6.789485151507618,
|
| 1751 |
+
"learning_rate": 2e-05,
|
| 1752 |
+
"loss": 0.8334197998046875,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.13487268018990073,
|
| 1757 |
+
"grad_norm": 3.755408787739404,
|
| 1758 |
+
"learning_rate": 2e-05,
|
| 1759 |
+
"loss": 0.8373711585998536,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.13541217091066032,
|
| 1764 |
+
"grad_norm": 4.1766718888496115,
|
| 1765 |
+
"learning_rate": 2e-05,
|
| 1766 |
+
"loss": 0.826263427734375,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.13595166163141995,
|
| 1771 |
+
"grad_norm": 1.4503676530546985,
|
| 1772 |
+
"learning_rate": 2e-05,
|
| 1773 |
+
"loss": 0.8411909103393554,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.13649115235217954,
|
| 1778 |
+
"grad_norm": 2.9419252466356443,
|
| 1779 |
+
"learning_rate": 2e-05,
|
| 1780 |
+
"loss": 0.8234809875488281,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.13703064307293913,
|
| 1785 |
+
"grad_norm": 3.2828905803244144,
|
| 1786 |
+
"learning_rate": 2e-05,
|
| 1787 |
+
"loss": 0.8695075035095214,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.13757013379369876,
|
| 1792 |
+
"grad_norm": 1.096460041177136,
|
| 1793 |
+
"learning_rate": 2e-05,
|
| 1794 |
+
"loss": 0.8468042373657226,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.13810962451445835,
|
| 1799 |
+
"grad_norm": 1.0597785888179652,
|
| 1800 |
+
"learning_rate": 2e-05,
|
| 1801 |
+
"loss": 0.8341262817382813,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.13864911523521795,
|
| 1806 |
+
"grad_norm": 9.087223552942303,
|
| 1807 |
+
"learning_rate": 2e-05,
|
| 1808 |
+
"loss": 0.8169255256652832,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.13918860595597757,
|
| 1813 |
+
"grad_norm": 7.72894258287119,
|
| 1814 |
+
"learning_rate": 2e-05,
|
| 1815 |
+
"loss": 0.8361001014709473,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.13972809667673716,
|
| 1820 |
+
"grad_norm": 1.84749896259604,
|
| 1821 |
+
"learning_rate": 2e-05,
|
| 1822 |
+
"loss": 0.8118861198425293,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.14026758739749676,
|
| 1827 |
+
"grad_norm": 1.499923768048581,
|
| 1828 |
+
"learning_rate": 2e-05,
|
| 1829 |
+
"loss": 0.8059223175048829,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.14080707811825638,
|
| 1834 |
+
"grad_norm": 2.469774702348604,
|
| 1835 |
+
"learning_rate": 2e-05,
|
| 1836 |
+
"loss": 0.8464885711669922,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.14134656883901597,
|
| 1841 |
+
"grad_norm": 1.4562619868061146,
|
| 1842 |
+
"learning_rate": 2e-05,
|
| 1843 |
+
"loss": 0.855031681060791,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.14188605955977557,
|
| 1848 |
+
"grad_norm": 1.1580412885044675,
|
| 1849 |
+
"learning_rate": 2e-05,
|
| 1850 |
+
"loss": 0.8250855445861817,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.14242555028053516,
|
| 1855 |
+
"grad_norm": 5.85834983160985,
|
| 1856 |
+
"learning_rate": 2e-05,
|
| 1857 |
+
"loss": 0.8170978546142578,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.14296504100129478,
|
| 1862 |
+
"grad_norm": 2.1118214513392153,
|
| 1863 |
+
"learning_rate": 2e-05,
|
| 1864 |
+
"loss": 0.8098548889160156,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.14350453172205438,
|
| 1869 |
+
"grad_norm": 14.992058801195402,
|
| 1870 |
+
"learning_rate": 2e-05,
|
| 1871 |
+
"loss": 0.8217266082763672,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.14404402244281397,
|
| 1876 |
+
"grad_norm": 2.3085553913502035,
|
| 1877 |
+
"learning_rate": 2e-05,
|
| 1878 |
+
"loss": 0.8170116424560547,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.1445835131635736,
|
| 1883 |
+
"grad_norm": 17.707266771089483,
|
| 1884 |
+
"learning_rate": 2e-05,
|
| 1885 |
+
"loss": 0.8280685424804688,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.1451230038843332,
|
| 1890 |
+
"grad_norm": 3.1378615826812766,
|
| 1891 |
+
"learning_rate": 2e-05,
|
| 1892 |
+
"loss": 0.8434804916381836,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.14566249460509278,
|
| 1897 |
+
"grad_norm": 2.5515862288748354,
|
| 1898 |
+
"learning_rate": 2e-05,
|
| 1899 |
+
"loss": 0.8259258270263672,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.1462019853258524,
|
| 1904 |
+
"grad_norm": 3.9230866849646424,
|
| 1905 |
+
"learning_rate": 2e-05,
|
| 1906 |
+
"loss": 0.8185581207275391,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.146741476046612,
|
| 1911 |
+
"grad_norm": 3.2217451837996975,
|
| 1912 |
+
"learning_rate": 2e-05,
|
| 1913 |
+
"loss": 0.8379020690917969,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.1472809667673716,
|
| 1918 |
+
"grad_norm": 1.557738526634509,
|
| 1919 |
+
"learning_rate": 2e-05,
|
| 1920 |
+
"loss": 0.8445243835449219,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.1478204574881312,
|
| 1925 |
+
"grad_norm": 1.5481180648756823,
|
| 1926 |
+
"learning_rate": 2e-05,
|
| 1927 |
+
"loss": 0.8328718185424805,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.1483599482088908,
|
| 1932 |
+
"grad_norm": 1.4388832631294857,
|
| 1933 |
+
"learning_rate": 2e-05,
|
| 1934 |
+
"loss": 0.8253545761108398,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.1488994389296504,
|
| 1939 |
+
"grad_norm": 1.0539255503501914,
|
| 1940 |
+
"learning_rate": 2e-05,
|
| 1941 |
+
"loss": 0.8445590972900391,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.14943892965041,
|
| 1946 |
+
"grad_norm": 4.95376827487241,
|
| 1947 |
+
"learning_rate": 2e-05,
|
| 1948 |
+
"loss": 0.8173467636108398,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.14997842037116962,
|
| 1953 |
+
"grad_norm": 4.600755881351036,
|
| 1954 |
+
"learning_rate": 2e-05,
|
| 1955 |
+
"loss": 0.8473451614379883,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.15051791109192922,
|
| 1960 |
+
"grad_norm": 1.3528051911908148,
|
| 1961 |
+
"learning_rate": 2e-05,
|
| 1962 |
+
"loss": 0.8311235427856445,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.1510574018126888,
|
| 1967 |
+
"grad_norm": 2.5244296910788786,
|
| 1968 |
+
"learning_rate": 2e-05,
|
| 1969 |
+
"loss": 0.8462516784667968,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.15159689253344844,
|
| 1974 |
+
"grad_norm": 1.3547145890263894,
|
| 1975 |
+
"learning_rate": 2e-05,
|
| 1976 |
+
"loss": 0.8145876884460449,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.15213638325420803,
|
| 1981 |
+
"grad_norm": 4.904666690029179,
|
| 1982 |
+
"learning_rate": 2e-05,
|
| 1983 |
+
"loss": 0.8427519798278809,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.15267587397496762,
|
| 1988 |
+
"grad_norm": 1.609216240729035,
|
| 1989 |
+
"learning_rate": 2e-05,
|
| 1990 |
+
"loss": 0.8435686111450196,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.15321536469572725,
|
| 1995 |
+
"grad_norm": 1.4281030100315293,
|
| 1996 |
+
"learning_rate": 2e-05,
|
| 1997 |
+
"loss": 0.8191211700439454,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.15375485541648684,
|
| 2002 |
+
"grad_norm": 1.7798898314789362,
|
| 2003 |
+
"learning_rate": 2e-05,
|
| 2004 |
+
"loss": 0.8122947692871094,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.15429434613724644,
|
| 2009 |
+
"grad_norm": 2.830774437932546,
|
| 2010 |
+
"learning_rate": 2e-05,
|
| 2011 |
+
"loss": 0.8318849563598633,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.15483383685800603,
|
| 2016 |
+
"grad_norm": 5.397744742205086,
|
| 2017 |
+
"learning_rate": 2e-05,
|
| 2018 |
+
"loss": 0.8305788993835449,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.15537332757876565,
|
| 2023 |
+
"grad_norm": 2.898600936491205,
|
| 2024 |
+
"learning_rate": 2e-05,
|
| 2025 |
+
"loss": 0.8408110618591309,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.15591281829952525,
|
| 2030 |
+
"grad_norm": 8.773413677216482,
|
| 2031 |
+
"learning_rate": 2e-05,
|
| 2032 |
+
"loss": 0.8318193435668946,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.15645230902028484,
|
| 2037 |
+
"grad_norm": 2.3331580039530095,
|
| 2038 |
+
"learning_rate": 2e-05,
|
| 2039 |
+
"loss": 0.8389785766601563,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.15699179974104446,
|
| 2044 |
+
"grad_norm": 2.568113164269533,
|
| 2045 |
+
"learning_rate": 2e-05,
|
| 2046 |
+
"loss": 0.8040918350219727,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.15753129046180406,
|
| 2051 |
+
"grad_norm": 1.4668399503612397,
|
| 2052 |
+
"learning_rate": 2e-05,
|
| 2053 |
+
"loss": 0.8159090042114258,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.15807078118256365,
|
| 2058 |
+
"grad_norm": 4.911971870615249,
|
| 2059 |
+
"learning_rate": 2e-05,
|
| 2060 |
+
"loss": 0.8376889228820801,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.15861027190332327,
|
| 2065 |
+
"grad_norm": 2.981636311233402,
|
| 2066 |
+
"learning_rate": 2e-05,
|
| 2067 |
+
"loss": 0.8562882423400879,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.15914976262408287,
|
| 2072 |
+
"grad_norm": 7.625355317159627,
|
| 2073 |
+
"learning_rate": 2e-05,
|
| 2074 |
+
"loss": 0.8109912872314453,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.15968925334484246,
|
| 2079 |
+
"grad_norm": 2.1473747341723555,
|
| 2080 |
+
"learning_rate": 2e-05,
|
| 2081 |
+
"loss": 0.852902603149414,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.16022874406560206,
|
| 2086 |
+
"grad_norm": 4.307544898725762,
|
| 2087 |
+
"learning_rate": 2e-05,
|
| 2088 |
+
"loss": 0.8281949043273926,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.16076823478636168,
|
| 2093 |
+
"grad_norm": 5.656390266767483,
|
| 2094 |
+
"learning_rate": 2e-05,
|
| 2095 |
+
"loss": 0.7931592464447021,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.16130772550712127,
|
| 2100 |
+
"grad_norm": 2.595677289608645,
|
| 2101 |
+
"learning_rate": 2e-05,
|
| 2102 |
+
"loss": 0.8464061737060546,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.16184721622788087,
|
| 2107 |
+
"grad_norm": 2.5286497672098376,
|
| 2108 |
+
"learning_rate": 2e-05,
|
| 2109 |
+
"loss": 0.8217041015625,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.1623867069486405,
|
| 2114 |
+
"grad_norm": 1.6815165067247406,
|
| 2115 |
+
"learning_rate": 2e-05,
|
| 2116 |
+
"loss": 0.8131906509399414,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.16292619766940009,
|
| 2121 |
+
"grad_norm": 4.470085181162879,
|
| 2122 |
+
"learning_rate": 2e-05,
|
| 2123 |
+
"loss": 0.8476617813110352,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.16346568839015968,
|
| 2128 |
+
"grad_norm": 57.757712059800895,
|
| 2129 |
+
"learning_rate": 2e-05,
|
| 2130 |
+
"loss": 0.8235321044921875,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.1640051791109193,
|
| 2135 |
+
"grad_norm": 1.3196900653412662,
|
| 2136 |
+
"learning_rate": 2e-05,
|
| 2137 |
+
"loss": 0.8411979675292969,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.1645446698316789,
|
| 2142 |
+
"grad_norm": 1.3565437997414518,
|
| 2143 |
+
"learning_rate": 2e-05,
|
| 2144 |
+
"loss": 0.8321975708007813,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.1650841605524385,
|
| 2149 |
+
"grad_norm": 1.9250529621406256,
|
| 2150 |
+
"learning_rate": 2e-05,
|
| 2151 |
+
"loss": 0.8183302879333496,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.1656236512731981,
|
| 2156 |
+
"grad_norm": 1.6665163841705182,
|
| 2157 |
+
"learning_rate": 2e-05,
|
| 2158 |
+
"loss": 0.8212745666503907,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.1661631419939577,
|
| 2163 |
+
"grad_norm": 0.9624218525439471,
|
| 2164 |
+
"learning_rate": 2e-05,
|
| 2165 |
+
"loss": 0.8360950469970703,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.1667026327147173,
|
| 2170 |
+
"grad_norm": 1.6154798724669914,
|
| 2171 |
+
"learning_rate": 2e-05,
|
| 2172 |
+
"loss": 0.8246768951416016,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.1672421234354769,
|
| 2177 |
+
"grad_norm": 1.4309300679224686,
|
| 2178 |
+
"learning_rate": 2e-05,
|
| 2179 |
+
"loss": 0.8143646240234375,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.16778161415623652,
|
| 2184 |
+
"grad_norm": 1.1954152653837478,
|
| 2185 |
+
"learning_rate": 2e-05,
|
| 2186 |
+
"loss": 0.8255207061767578,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.1683211048769961,
|
| 2191 |
+
"grad_norm": 3.874535720645589,
|
| 2192 |
+
"learning_rate": 2e-05,
|
| 2193 |
+
"loss": 0.8268468856811524,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.1688605955977557,
|
| 2198 |
+
"grad_norm": 1.666585578869163,
|
| 2199 |
+
"learning_rate": 2e-05,
|
| 2200 |
+
"loss": 0.8370718955993652,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.16940008631851533,
|
| 2205 |
+
"grad_norm": 2.1122716764285214,
|
| 2206 |
+
"learning_rate": 2e-05,
|
| 2207 |
+
"loss": 0.8222600936889648,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.16993957703927492,
|
| 2212 |
+
"grad_norm": 1.626502422759658,
|
| 2213 |
+
"learning_rate": 2e-05,
|
| 2214 |
+
"loss": 0.8307378768920899,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.17047906776003452,
|
| 2219 |
+
"grad_norm": 1.249709631021305,
|
| 2220 |
+
"learning_rate": 2e-05,
|
| 2221 |
+
"loss": 0.8274057388305665,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.17101855848079414,
|
| 2226 |
+
"grad_norm": 1.2755622837169793,
|
| 2227 |
+
"learning_rate": 2e-05,
|
| 2228 |
+
"loss": 0.8418668746948242,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.17155804920155374,
|
| 2233 |
+
"grad_norm": 1.1471931685322956,
|
| 2234 |
+
"learning_rate": 2e-05,
|
| 2235 |
+
"loss": 0.8037077903747558,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.17209753992231333,
|
| 2240 |
+
"grad_norm": 2.731067427188893,
|
| 2241 |
+
"learning_rate": 2e-05,
|
| 2242 |
+
"loss": 0.7964674949645996,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.17263703064307295,
|
| 2247 |
+
"grad_norm": 1.4738003450064086,
|
| 2248 |
+
"learning_rate": 2e-05,
|
| 2249 |
+
"loss": 0.8007509231567382,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.17317652136383255,
|
| 2254 |
+
"grad_norm": 1.310820117633026,
|
| 2255 |
+
"learning_rate": 2e-05,
|
| 2256 |
+
"loss": 0.8201332092285156,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.17371601208459214,
|
| 2261 |
+
"grad_norm": 1.3355379129188545,
|
| 2262 |
+
"learning_rate": 2e-05,
|
| 2263 |
+
"loss": 0.8006715774536133,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.17425550280535174,
|
| 2268 |
+
"grad_norm": 2.5523015565267078,
|
| 2269 |
+
"learning_rate": 2e-05,
|
| 2270 |
+
"loss": 0.8170543670654297,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.17479499352611136,
|
| 2275 |
+
"grad_norm": 7.648994384982638,
|
| 2276 |
+
"learning_rate": 2e-05,
|
| 2277 |
+
"loss": 0.7966285705566406,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.17533448424687095,
|
| 2282 |
+
"grad_norm": 2.2291376008212502,
|
| 2283 |
+
"learning_rate": 2e-05,
|
| 2284 |
+
"loss": 0.8066202163696289,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.17587397496763055,
|
| 2289 |
+
"grad_norm": 1.9013943798842645,
|
| 2290 |
+
"learning_rate": 2e-05,
|
| 2291 |
+
"loss": 0.8028726577758789,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.17641346568839017,
|
| 2296 |
+
"grad_norm": 1.2094542057997177,
|
| 2297 |
+
"learning_rate": 2e-05,
|
| 2298 |
+
"loss": 0.8179764747619629,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.17695295640914976,
|
| 2303 |
+
"grad_norm": 2.7367851475598,
|
| 2304 |
+
"learning_rate": 2e-05,
|
| 2305 |
+
"loss": 0.8103551864624023,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.17749244712990936,
|
| 2310 |
+
"grad_norm": 2.9058412897859656,
|
| 2311 |
+
"learning_rate": 2e-05,
|
| 2312 |
+
"loss": 0.8264286041259765,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.17803193785066898,
|
| 2317 |
+
"grad_norm": 3.226004030131601,
|
| 2318 |
+
"learning_rate": 2e-05,
|
| 2319 |
+
"loss": 0.8190610885620118,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.17857142857142858,
|
| 2324 |
+
"grad_norm": 1.8721988330858943,
|
| 2325 |
+
"learning_rate": 2e-05,
|
| 2326 |
+
"loss": 0.8306256294250488,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.17911091929218817,
|
| 2331 |
+
"grad_norm": 1.3899444238570022,
|
| 2332 |
+
"learning_rate": 2e-05,
|
| 2333 |
+
"loss": 0.8446199417114257,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.17965041001294776,
|
| 2338 |
+
"grad_norm": 1.0929088079648228,
|
| 2339 |
+
"learning_rate": 2e-05,
|
| 2340 |
+
"loss": 0.7976521968841552,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.1801899007337074,
|
| 2345 |
+
"grad_norm": 1.1793561049774377,
|
| 2346 |
+
"learning_rate": 2e-05,
|
| 2347 |
+
"loss": 0.8168546676635742,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.18072939145446698,
|
| 2352 |
+
"grad_norm": 15.176731004709248,
|
| 2353 |
+
"learning_rate": 2e-05,
|
| 2354 |
+
"loss": 0.8124425888061524,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.18126888217522658,
|
| 2359 |
+
"grad_norm": 1.0594125222809312,
|
| 2360 |
+
"learning_rate": 2e-05,
|
| 2361 |
+
"loss": 0.806308364868164,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.1818083728959862,
|
| 2366 |
+
"grad_norm": 1.9453105377863469,
|
| 2367 |
+
"learning_rate": 2e-05,
|
| 2368 |
+
"loss": 0.8045181274414063,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.1823478636167458,
|
| 2373 |
+
"grad_norm": 0.9815060321000388,
|
| 2374 |
+
"learning_rate": 2e-05,
|
| 2375 |
+
"loss": 0.8289067268371582,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.1828873543375054,
|
| 2380 |
+
"grad_norm": 1.107626410337324,
|
| 2381 |
+
"learning_rate": 2e-05,
|
| 2382 |
+
"loss": 0.8187108993530273,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.183426845058265,
|
| 2387 |
+
"grad_norm": 1.3713782627751003,
|
| 2388 |
+
"learning_rate": 2e-05,
|
| 2389 |
+
"loss": 0.8212367057800293,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.1839663357790246,
|
| 2394 |
+
"grad_norm": 1.1934434763982462,
|
| 2395 |
+
"learning_rate": 2e-05,
|
| 2396 |
+
"loss": 0.8170612335205079,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.1845058264997842,
|
| 2401 |
+
"grad_norm": 1.0151425442795299,
|
| 2402 |
+
"learning_rate": 2e-05,
|
| 2403 |
+
"loss": 0.8014205932617188,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.18504531722054382,
|
| 2408 |
+
"grad_norm": 0.9339054423212217,
|
| 2409 |
+
"learning_rate": 2e-05,
|
| 2410 |
+
"loss": 0.8339118003845215,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.18558480794130341,
|
| 2415 |
+
"grad_norm": 0.967368753176617,
|
| 2416 |
+
"learning_rate": 2e-05,
|
| 2417 |
+
"loss": 0.8132877349853516,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.186124298662063,
|
| 2422 |
+
"grad_norm": 2.162294345081955,
|
| 2423 |
+
"learning_rate": 2e-05,
|
| 2424 |
+
"loss": 0.8348134994506836,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.1866637893828226,
|
| 2429 |
+
"grad_norm": 2.687504589015093,
|
| 2430 |
+
"learning_rate": 2e-05,
|
| 2431 |
+
"loss": 0.8514577865600585,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.18720328010358223,
|
| 2436 |
+
"grad_norm": 1.3616314045359637,
|
| 2437 |
+
"learning_rate": 2e-05,
|
| 2438 |
+
"loss": 0.8360185623168945,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.18774277082434182,
|
| 2443 |
+
"grad_norm": 0.7945412849733318,
|
| 2444 |
+
"learning_rate": 2e-05,
|
| 2445 |
+
"loss": 0.8060104370117187,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.18828226154510141,
|
| 2450 |
+
"grad_norm": 1.2865284566965982,
|
| 2451 |
+
"learning_rate": 2e-05,
|
| 2452 |
+
"loss": 0.8192827224731445,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.18882175226586104,
|
| 2457 |
+
"grad_norm": 1.0470430689868928,
|
| 2458 |
+
"learning_rate": 2e-05,
|
| 2459 |
+
"loss": 0.8104745864868164,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.18936124298662063,
|
| 2464 |
+
"grad_norm": 1.0033888590154632,
|
| 2465 |
+
"learning_rate": 2e-05,
|
| 2466 |
+
"loss": 0.8070037841796875,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.18990073370738023,
|
| 2471 |
+
"grad_norm": 1.4237596335672749,
|
| 2472 |
+
"learning_rate": 2e-05,
|
| 2473 |
+
"loss": 0.7981664657592773,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.19044022442813985,
|
| 2478 |
+
"grad_norm": 0.9746666528478075,
|
| 2479 |
+
"learning_rate": 2e-05,
|
| 2480 |
+
"loss": 0.8370217323303223,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.19097971514889944,
|
| 2485 |
+
"grad_norm": 2.370032426413775,
|
| 2486 |
+
"learning_rate": 2e-05,
|
| 2487 |
+
"loss": 0.8089248657226562,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.19151920586965904,
|
| 2492 |
+
"grad_norm": 1.4096737349425104,
|
| 2493 |
+
"learning_rate": 2e-05,
|
| 2494 |
+
"loss": 0.8232223510742187,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.19205869659041863,
|
| 2499 |
+
"grad_norm": 1.5464845999590675,
|
| 2500 |
+
"learning_rate": 2e-05,
|
| 2501 |
+
"loss": 0.8215899467468262,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.19259818731117825,
|
| 2506 |
+
"grad_norm": 1.8902013609436816,
|
| 2507 |
+
"learning_rate": 2e-05,
|
| 2508 |
+
"loss": 0.8071002960205078,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.19313767803193785,
|
| 2513 |
+
"grad_norm": 1.0834961800168135,
|
| 2514 |
+
"learning_rate": 2e-05,
|
| 2515 |
+
"loss": 0.818499755859375,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.19367716875269744,
|
| 2520 |
+
"grad_norm": 1.469175436322324,
|
| 2521 |
+
"learning_rate": 2e-05,
|
| 2522 |
+
"loss": 0.8098075866699219,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.19421665947345707,
|
| 2527 |
+
"grad_norm": 1.4457828669467623,
|
| 2528 |
+
"learning_rate": 2e-05,
|
| 2529 |
+
"loss": 0.8190594673156738,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.19475615019421666,
|
| 2534 |
+
"grad_norm": 1.2339738565051197,
|
| 2535 |
+
"learning_rate": 2e-05,
|
| 2536 |
+
"loss": 0.8001772880554199,
|
| 2537 |
+
"step": 3610
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.19529564091497625,
|
| 2541 |
+
"grad_norm": 1.0129969586529555,
|
| 2542 |
+
"learning_rate": 2e-05,
|
| 2543 |
+
"loss": 0.8169227600097656,
|
| 2544 |
+
"step": 3620
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.19583513163573588,
|
| 2548 |
+
"grad_norm": 1.8405886663060382,
|
| 2549 |
+
"learning_rate": 2e-05,
|
| 2550 |
+
"loss": 0.8161848068237305,
|
| 2551 |
+
"step": 3630
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.19637462235649547,
|
| 2555 |
+
"grad_norm": 0.8630501803188849,
|
| 2556 |
+
"learning_rate": 2e-05,
|
| 2557 |
+
"loss": 0.7952068328857422,
|
| 2558 |
+
"step": 3640
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.19691411307725507,
|
| 2562 |
+
"grad_norm": 0.9252474483323625,
|
| 2563 |
+
"learning_rate": 2e-05,
|
| 2564 |
+
"loss": 0.836590576171875,
|
| 2565 |
+
"step": 3650
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.1974536037980147,
|
| 2569 |
+
"grad_norm": 0.9338702892336649,
|
| 2570 |
+
"learning_rate": 2e-05,
|
| 2571 |
+
"loss": 0.8275209426879883,
|
| 2572 |
+
"step": 3660
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.19799309451877428,
|
| 2576 |
+
"grad_norm": 0.9675019978413074,
|
| 2577 |
+
"learning_rate": 2e-05,
|
| 2578 |
+
"loss": 0.8117290496826172,
|
| 2579 |
+
"step": 3670
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.19853258523953388,
|
| 2583 |
+
"grad_norm": 0.8692908353972926,
|
| 2584 |
+
"learning_rate": 2e-05,
|
| 2585 |
+
"loss": 0.8203290939331055,
|
| 2586 |
+
"step": 3680
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.19907207596029347,
|
| 2590 |
+
"grad_norm": 0.9432619722907122,
|
| 2591 |
+
"learning_rate": 2e-05,
|
| 2592 |
+
"loss": 0.8062999725341797,
|
| 2593 |
+
"step": 3690
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.1996115666810531,
|
| 2597 |
+
"grad_norm": 0.9465547041865168,
|
| 2598 |
+
"learning_rate": 2e-05,
|
| 2599 |
+
"loss": 0.8359698295593262,
|
| 2600 |
+
"step": 3700
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.2001510574018127,
|
| 2604 |
+
"grad_norm": 0.8544475554924795,
|
| 2605 |
+
"learning_rate": 2e-05,
|
| 2606 |
+
"loss": 0.8113945960998535,
|
| 2607 |
+
"step": 3710
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.20069054812257228,
|
| 2611 |
+
"grad_norm": 1.0261458270031636,
|
| 2612 |
+
"learning_rate": 2e-05,
|
| 2613 |
+
"loss": 0.8012717247009278,
|
| 2614 |
+
"step": 3720
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.2012300388433319,
|
| 2618 |
+
"grad_norm": 2.3510637262715686,
|
| 2619 |
+
"learning_rate": 2e-05,
|
| 2620 |
+
"loss": 0.7971127510070801,
|
| 2621 |
+
"step": 3730
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.2017695295640915,
|
| 2625 |
+
"grad_norm": 1.0401878649419207,
|
| 2626 |
+
"learning_rate": 2e-05,
|
| 2627 |
+
"loss": 0.8233554840087891,
|
| 2628 |
+
"step": 3740
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.2023090202848511,
|
| 2632 |
+
"grad_norm": 1.6871599880351058,
|
| 2633 |
+
"learning_rate": 2e-05,
|
| 2634 |
+
"loss": 0.8121973037719726,
|
| 2635 |
+
"step": 3750
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 0.20284851100561072,
|
| 2639 |
+
"grad_norm": 0.9834662903562297,
|
| 2640 |
+
"learning_rate": 2e-05,
|
| 2641 |
+
"loss": 0.799620532989502,
|
| 2642 |
+
"step": 3760
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.2033880017263703,
|
| 2646 |
+
"grad_norm": 13.856106429547337,
|
| 2647 |
+
"learning_rate": 2e-05,
|
| 2648 |
+
"loss": 0.8100821495056152,
|
| 2649 |
+
"step": 3770
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 0.2039274924471299,
|
| 2653 |
+
"grad_norm": 1.7971124760415595,
|
| 2654 |
+
"learning_rate": 2e-05,
|
| 2655 |
+
"loss": 0.822230339050293,
|
| 2656 |
+
"step": 3780
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.2044669831678895,
|
| 2660 |
+
"grad_norm": 1.898826077210772,
|
| 2661 |
+
"learning_rate": 2e-05,
|
| 2662 |
+
"loss": 0.8280195236206055,
|
| 2663 |
+
"step": 3790
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.20500647388864912,
|
| 2667 |
+
"grad_norm": 1.5690048084284152,
|
| 2668 |
+
"learning_rate": 2e-05,
|
| 2669 |
+
"loss": 0.7768063545227051,
|
| 2670 |
+
"step": 3800
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.20554596460940872,
|
| 2674 |
+
"grad_norm": 1.2924271342168896,
|
| 2675 |
+
"learning_rate": 2e-05,
|
| 2676 |
+
"loss": 0.8139206886291503,
|
| 2677 |
+
"step": 3810
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 0.2060854553301683,
|
| 2681 |
+
"grad_norm": 1.713906617227459,
|
| 2682 |
+
"learning_rate": 2e-05,
|
| 2683 |
+
"loss": 0.8099102020263672,
|
| 2684 |
+
"step": 3820
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.20662494605092793,
|
| 2688 |
+
"grad_norm": 1.0000768599712642,
|
| 2689 |
+
"learning_rate": 2e-05,
|
| 2690 |
+
"loss": 0.8023160934448242,
|
| 2691 |
+
"step": 3830
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 0.20716443677168753,
|
| 2695 |
+
"grad_norm": 1.8923446682002336,
|
| 2696 |
+
"learning_rate": 2e-05,
|
| 2697 |
+
"loss": 0.8316279411315918,
|
| 2698 |
+
"step": 3840
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 0.20770392749244712,
|
| 2702 |
+
"grad_norm": 0.9410427812888499,
|
| 2703 |
+
"learning_rate": 2e-05,
|
| 2704 |
+
"loss": 0.8073904037475585,
|
| 2705 |
+
"step": 3850
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.20824341821320674,
|
| 2709 |
+
"grad_norm": 1.8310165996432775,
|
| 2710 |
+
"learning_rate": 2e-05,
|
| 2711 |
+
"loss": 0.8441598892211915,
|
| 2712 |
+
"step": 3860
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 0.20878290893396634,
|
| 2716 |
+
"grad_norm": 3.2539518984867444,
|
| 2717 |
+
"learning_rate": 2e-05,
|
| 2718 |
+
"loss": 0.8111666679382324,
|
| 2719 |
+
"step": 3870
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 0.20932239965472593,
|
| 2723 |
+
"grad_norm": 1.0747584520027538,
|
| 2724 |
+
"learning_rate": 2e-05,
|
| 2725 |
+
"loss": 0.8289793968200684,
|
| 2726 |
+
"step": 3880
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.20986189037548555,
|
| 2730 |
+
"grad_norm": 1.1928486588388492,
|
| 2731 |
+
"learning_rate": 2e-05,
|
| 2732 |
+
"loss": 0.7862348556518555,
|
| 2733 |
+
"step": 3890
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 0.21040138109624515,
|
| 2737 |
+
"grad_norm": 1.7434197055178486,
|
| 2738 |
+
"learning_rate": 2e-05,
|
| 2739 |
+
"loss": 0.8227115631103515,
|
| 2740 |
+
"step": 3900
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 0.21094087181700474,
|
| 2744 |
+
"grad_norm": 1.0267098850458918,
|
| 2745 |
+
"learning_rate": 2e-05,
|
| 2746 |
+
"loss": 0.8255733489990235,
|
| 2747 |
+
"step": 3910
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 0.21148036253776434,
|
| 2751 |
+
"grad_norm": 1.2265343976727108,
|
| 2752 |
+
"learning_rate": 2e-05,
|
| 2753 |
+
"loss": 0.7926362514495849,
|
| 2754 |
+
"step": 3920
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 0.21201985325852396,
|
| 2758 |
+
"grad_norm": 0.9373764058253546,
|
| 2759 |
+
"learning_rate": 2e-05,
|
| 2760 |
+
"loss": 0.7910985469818115,
|
| 2761 |
+
"step": 3930
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 0.21255934397928355,
|
| 2765 |
+
"grad_norm": 1.194071164302341,
|
| 2766 |
+
"learning_rate": 2e-05,
|
| 2767 |
+
"loss": 0.8127195358276367,
|
| 2768 |
+
"step": 3940
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 0.21309883470004315,
|
| 2772 |
+
"grad_norm": 1.260475572232143,
|
| 2773 |
+
"learning_rate": 2e-05,
|
| 2774 |
+
"loss": 0.8285701751708985,
|
| 2775 |
+
"step": 3950
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 0.21363832542080277,
|
| 2779 |
+
"grad_norm": 0.9898466932738494,
|
| 2780 |
+
"learning_rate": 2e-05,
|
| 2781 |
+
"loss": 0.816091251373291,
|
| 2782 |
+
"step": 3960
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 0.21417781614156237,
|
| 2786 |
+
"grad_norm": 1.0090063218881065,
|
| 2787 |
+
"learning_rate": 2e-05,
|
| 2788 |
+
"loss": 0.7883634090423584,
|
| 2789 |
+
"step": 3970
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 0.21471730686232196,
|
| 2793 |
+
"grad_norm": 1.6131569146168403,
|
| 2794 |
+
"learning_rate": 2e-05,
|
| 2795 |
+
"loss": 0.8127913475036621,
|
| 2796 |
+
"step": 3980
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 0.21525679758308158,
|
| 2800 |
+
"grad_norm": 1.1595573594793847,
|
| 2801 |
+
"learning_rate": 2e-05,
|
| 2802 |
+
"loss": 0.8038475036621093,
|
| 2803 |
+
"step": 3990
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 0.21579628830384118,
|
| 2807 |
+
"grad_norm": 1.9569762958682257,
|
| 2808 |
+
"learning_rate": 2e-05,
|
| 2809 |
+
"loss": 0.8159636497497559,
|
| 2810 |
+
"step": 4000
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 0.21633577902460077,
|
| 2814 |
+
"grad_norm": 2.335005979386267,
|
| 2815 |
+
"learning_rate": 2e-05,
|
| 2816 |
+
"loss": 0.8046688079833985,
|
| 2817 |
+
"step": 4010
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 0.21687526974536037,
|
| 2821 |
+
"grad_norm": 1.0399747803957953,
|
| 2822 |
+
"learning_rate": 2e-05,
|
| 2823 |
+
"loss": 0.8062277793884277,
|
| 2824 |
+
"step": 4020
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 0.21741476046612,
|
| 2828 |
+
"grad_norm": 2.0436577055413636,
|
| 2829 |
+
"learning_rate": 2e-05,
|
| 2830 |
+
"loss": 0.7829396724700928,
|
| 2831 |
+
"step": 4030
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 0.21795425118687958,
|
| 2835 |
+
"grad_norm": 3.7324962194464963,
|
| 2836 |
+
"learning_rate": 2e-05,
|
| 2837 |
+
"loss": 0.779371976852417,
|
| 2838 |
+
"step": 4040
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 0.21849374190763918,
|
| 2842 |
+
"grad_norm": 1.1382565851810948,
|
| 2843 |
+
"learning_rate": 2e-05,
|
| 2844 |
+
"loss": 0.7965741634368897,
|
| 2845 |
+
"step": 4050
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 0.2190332326283988,
|
| 2849 |
+
"grad_norm": 1.8123028132941512,
|
| 2850 |
+
"learning_rate": 2e-05,
|
| 2851 |
+
"loss": 0.7883219718933105,
|
| 2852 |
+
"step": 4060
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 0.2195727233491584,
|
| 2856 |
+
"grad_norm": 1.0506410293373154,
|
| 2857 |
+
"learning_rate": 2e-05,
|
| 2858 |
+
"loss": 0.797592306137085,
|
| 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:cda2ce1a3d48dbaae97584542b63fb1b2940959f4e24db1a7c83985cbc8f2f9d
|
| 3 |
+
size 7313
|