codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base: checkpoint-600 (inference files)
Browse files- .gitattributes +1 -0
- checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/chat_template.jinja +54 -0
- checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/config.json +61 -0
- checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/pytorch_model.bin +3 -0
- checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/thought_projector.pt +3 -0
- checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/tokenizer.json +3 -0
- checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/tokenizer_config.json +30 -0
- checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/trainer_state.json +1474 -0
.gitattributes
CHANGED
|
@@ -163,3 +163,4 @@ checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-600/tokenizer.json filte
|
|
| 163 |
checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-900/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 164 |
checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/wandb/offline-run-20260702_042011-gs3t81f2/run-gs3t81f2.wandb filter=lfs diff=lfs merge=lfs -text
|
| 165 |
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 163 |
checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/checkpoint-900/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 164 |
checkpoints/codi_recon_delta_1.5b_ls2_rw0.03/wandb/offline-run-20260702_042011-gs3t81f2/run-gs3t81f2.wandb filter=lfs diff=lfs merge=lfs -text
|
| 165 |
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 166 |
+
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/chat_template.jinja
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 4 |
+
{{- messages[0]['content'] }}
|
| 5 |
+
{%- else %}
|
| 6 |
+
{{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
|
| 7 |
+
{%- endif %}
|
| 8 |
+
{{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 9 |
+
{%- for tool in tools %}
|
| 10 |
+
{{- "\n" }}
|
| 11 |
+
{{- tool | tojson }}
|
| 12 |
+
{%- endfor %}
|
| 13 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 14 |
+
{%- else %}
|
| 15 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 16 |
+
{{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
|
| 17 |
+
{%- else %}
|
| 18 |
+
{{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
|
| 19 |
+
{%- endif %}
|
| 20 |
+
{%- endif %}
|
| 21 |
+
{%- for message in messages %}
|
| 22 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
|
| 23 |
+
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
|
| 24 |
+
{%- elif message.role == "assistant" %}
|
| 25 |
+
{{- '<|im_start|>' + message.role }}
|
| 26 |
+
{%- if message.content %}
|
| 27 |
+
{{- '\n' + message.content }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{%- for tool_call in message.tool_calls %}
|
| 30 |
+
{%- if tool_call.function is defined %}
|
| 31 |
+
{%- set tool_call = tool_call.function %}
|
| 32 |
+
{%- endif %}
|
| 33 |
+
{{- '\n<tool_call>\n{"name": "' }}
|
| 34 |
+
{{- tool_call.name }}
|
| 35 |
+
{{- '", "arguments": ' }}
|
| 36 |
+
{{- tool_call.arguments | tojson }}
|
| 37 |
+
{{- '}\n</tool_call>' }}
|
| 38 |
+
{%- endfor %}
|
| 39 |
+
{{- '<|im_end|>\n' }}
|
| 40 |
+
{%- elif message.role == "tool" %}
|
| 41 |
+
{%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
|
| 42 |
+
{{- '<|im_start|>user' }}
|
| 43 |
+
{%- endif %}
|
| 44 |
+
{{- '\n<tool_response>\n' }}
|
| 45 |
+
{{- message.content }}
|
| 46 |
+
{{- '\n</tool_response>' }}
|
| 47 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 48 |
+
{{- '<|im_end|>\n' }}
|
| 49 |
+
{%- endif %}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{%- if add_generation_prompt %}
|
| 53 |
+
{{- '<|im_start|>assistant\n' }}
|
| 54 |
+
{%- endif %}
|
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen2ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"bos_token_id": 151643,
|
| 7 |
+
"dtype": "bfloat16",
|
| 8 |
+
"eos_token_id": 151643,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 1536,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 8960,
|
| 13 |
+
"layer_types": [
|
| 14 |
+
"full_attention",
|
| 15 |
+
"full_attention",
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention"
|
| 42 |
+
],
|
| 43 |
+
"max_position_embeddings": 32768,
|
| 44 |
+
"max_window_layers": 28,
|
| 45 |
+
"model_type": "qwen2",
|
| 46 |
+
"num_attention_heads": 12,
|
| 47 |
+
"num_hidden_layers": 28,
|
| 48 |
+
"num_key_value_heads": 2,
|
| 49 |
+
"pad_token_id": null,
|
| 50 |
+
"rms_norm_eps": 1e-06,
|
| 51 |
+
"rope_parameters": {
|
| 52 |
+
"rope_theta": 1000000.0,
|
| 53 |
+
"rope_type": "default"
|
| 54 |
+
},
|
| 55 |
+
"sliding_window": null,
|
| 56 |
+
"tie_word_embeddings": true,
|
| 57 |
+
"transformers_version": "5.9.0",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"use_sliding_window": false,
|
| 60 |
+
"vocab_size": 151676
|
| 61 |
+
}
|
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e70b3199bfba48d1f74e6ea35e172a0f839387a1eb0997abf1bdc9847a3111f9
|
| 3 |
+
size 3096275515
|
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/thought_projector.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c12feb87bb584ed5ebcf6bc41cc60630fe3b294bba4b9d574d58e134f0c63c56
|
| 3 |
+
size 9445953
|
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:279b924ef57508c2dc50060607fe7a473a569fd1f6054bdbbdb96e5675edd62e
|
| 3 |
+
size 11424000
|
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/tokenizer_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|endoftext|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|im_start|>",
|
| 10 |
+
"<|im_end|>",
|
| 11 |
+
"<|object_ref_start|>",
|
| 12 |
+
"<|object_ref_end|>",
|
| 13 |
+
"<|box_start|>",
|
| 14 |
+
"<|box_end|>",
|
| 15 |
+
"<|quad_start|>",
|
| 16 |
+
"<|quad_end|>",
|
| 17 |
+
"<|vision_start|>",
|
| 18 |
+
"<|vision_end|>",
|
| 19 |
+
"<|vision_pad|>",
|
| 20 |
+
"<|image_pad|>",
|
| 21 |
+
"<|video_pad|>"
|
| 22 |
+
],
|
| 23 |
+
"is_local": true,
|
| 24 |
+
"local_files_only": true,
|
| 25 |
+
"model_max_length": 32768,
|
| 26 |
+
"pad_token": "<|endoftext|>",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null
|
| 30 |
+
}
|
checkpoints/codi_recon_localfull_1.5b_ls1_rw0.1_len128_unknown_base/checkpoint-600/trainer_state.json
ADDED
|
@@ -0,0 +1,1474 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.17301038062283736,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0014417531718569781,
|
| 14 |
+
"grad_norm": 704.0,
|
| 15 |
+
"kd_loss": 0.2430853694677353,
|
| 16 |
+
"learning_rate": 8.88888888888889e-07,
|
| 17 |
+
"loss": 0.825105857849121,
|
| 18 |
+
"recon_loss": 2.0671944618225098,
|
| 19 |
+
"recon_trunc": 0.0,
|
| 20 |
+
"step": 5,
|
| 21 |
+
"student_loss": 0.16177205741405487,
|
| 22 |
+
"teacher_loss": 0.018314069136977196
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.0028835063437139563,
|
| 26 |
+
"grad_norm": 65.0,
|
| 27 |
+
"kd_loss": 0.21926775574684143,
|
| 28 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 29 |
+
"loss": 0.8297138214111328,
|
| 30 |
+
"recon_loss": 3.5016913414001465,
|
| 31 |
+
"recon_trunc": 0.0,
|
| 32 |
+
"step": 10,
|
| 33 |
+
"student_loss": 0.214988574385643,
|
| 34 |
+
"teacher_loss": 0.004123217426240444
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.004325259515570935,
|
| 38 |
+
"grad_norm": 2640.0,
|
| 39 |
+
"kd_loss": 0.24922692775726318,
|
| 40 |
+
"learning_rate": 3.1111111111111116e-06,
|
| 41 |
+
"loss": 0.8162508964538574,
|
| 42 |
+
"recon_loss": 4.021753311157227,
|
| 43 |
+
"recon_trunc": 0.0,
|
| 44 |
+
"step": 15,
|
| 45 |
+
"student_loss": 0.23302094638347626,
|
| 46 |
+
"teacher_loss": 0.10642654448747635
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.0057670126874279125,
|
| 50 |
+
"grad_norm": 106.0,
|
| 51 |
+
"kd_loss": 0.2604961395263672,
|
| 52 |
+
"learning_rate": 4.222222222222223e-06,
|
| 53 |
+
"loss": 0.7747313499450683,
|
| 54 |
+
"recon_loss": 2.7793593406677246,
|
| 55 |
+
"recon_trunc": 0.0,
|
| 56 |
+
"step": 20,
|
| 57 |
+
"student_loss": 0.18348067998886108,
|
| 58 |
+
"teacher_loss": 0.01916802115738392
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.00720876585928489,
|
| 62 |
+
"grad_norm": 34.75,
|
| 63 |
+
"kd_loss": 0.2377234250307083,
|
| 64 |
+
"learning_rate": 5.333333333333334e-06,
|
| 65 |
+
"loss": 0.7149038314819336,
|
| 66 |
+
"recon_loss": 1.7850466966629028,
|
| 67 |
+
"recon_trunc": 0.0,
|
| 68 |
+
"step": 25,
|
| 69 |
+
"student_loss": 0.1838301569223404,
|
| 70 |
+
"teacher_loss": 0.05608585476875305
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.00865051903114187,
|
| 74 |
+
"grad_norm": 43.0,
|
| 75 |
+
"kd_loss": 0.24326685070991516,
|
| 76 |
+
"learning_rate": 6.444444444444445e-06,
|
| 77 |
+
"loss": 0.6312935829162598,
|
| 78 |
+
"recon_loss": 1.2304421663284302,
|
| 79 |
+
"recon_trunc": 0.290909081697464,
|
| 80 |
+
"step": 30,
|
| 81 |
+
"student_loss": 0.1560046672821045,
|
| 82 |
+
"teacher_loss": 0.042699508368968964
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.010092272202998846,
|
| 86 |
+
"grad_norm": 288.0,
|
| 87 |
+
"kd_loss": 0.14811451733112335,
|
| 88 |
+
"learning_rate": 7.555555555555556e-06,
|
| 89 |
+
"loss": 0.5599963188171386,
|
| 90 |
+
"recon_loss": 0.8689510226249695,
|
| 91 |
+
"recon_trunc": 0.0,
|
| 92 |
+
"step": 35,
|
| 93 |
+
"student_loss": 0.1670655906200409,
|
| 94 |
+
"teacher_loss": 0.12141425907611847
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.011534025374855825,
|
| 98 |
+
"grad_norm": 14.0,
|
| 99 |
+
"kd_loss": 0.1193324401974678,
|
| 100 |
+
"learning_rate": 8.666666666666668e-06,
|
| 101 |
+
"loss": 0.4744464874267578,
|
| 102 |
+
"recon_loss": 1.6873657703399658,
|
| 103 |
+
"recon_trunc": 0.0,
|
| 104 |
+
"step": 40,
|
| 105 |
+
"student_loss": 0.01387017872184515,
|
| 106 |
+
"teacher_loss": 0.004245229996740818
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.012975778546712802,
|
| 110 |
+
"grad_norm": 8.5,
|
| 111 |
+
"kd_loss": 0.12177280336618423,
|
| 112 |
+
"learning_rate": 9.777777777777779e-06,
|
| 113 |
+
"loss": 0.46063594818115233,
|
| 114 |
+
"recon_loss": 2.2260711193084717,
|
| 115 |
+
"recon_trunc": 0.0,
|
| 116 |
+
"step": 45,
|
| 117 |
+
"student_loss": 0.011669134721159935,
|
| 118 |
+
"teacher_loss": 0.004297277890145779
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.01441753171856978,
|
| 122 |
+
"grad_norm": 15.1875,
|
| 123 |
+
"kd_loss": 0.1361657828092575,
|
| 124 |
+
"learning_rate": 9.999813520519569e-06,
|
| 125 |
+
"loss": 0.4066462993621826,
|
| 126 |
+
"recon_loss": 1.7938929796218872,
|
| 127 |
+
"recon_trunc": 0.0,
|
| 128 |
+
"step": 50,
|
| 129 |
+
"student_loss": 0.10370920598506927,
|
| 130 |
+
"teacher_loss": 0.06052231043577194
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.015859284890426758,
|
| 134 |
+
"grad_norm": 256.0,
|
| 135 |
+
"kd_loss": 0.07108579576015472,
|
| 136 |
+
"learning_rate": 9.999055971469864e-06,
|
| 137 |
+
"loss": 0.38227524757385256,
|
| 138 |
+
"recon_loss": 2.648643970489502,
|
| 139 |
+
"recon_trunc": 0.0,
|
| 140 |
+
"step": 55,
|
| 141 |
+
"student_loss": 0.018915466964244843,
|
| 142 |
+
"teacher_loss": 0.015859190374612808
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.01730103806228374,
|
| 146 |
+
"grad_norm": 498.0,
|
| 147 |
+
"kd_loss": 0.059754595160484314,
|
| 148 |
+
"learning_rate": 9.997715786107228e-06,
|
| 149 |
+
"loss": 0.4097599983215332,
|
| 150 |
+
"recon_loss": 6.485691070556641,
|
| 151 |
+
"recon_trunc": 0.0,
|
| 152 |
+
"step": 60,
|
| 153 |
+
"student_loss": 0.015150920487940311,
|
| 154 |
+
"teacher_loss": 0.009373589418828487
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.018742791234140715,
|
| 158 |
+
"grad_norm": 16.625,
|
| 159 |
+
"kd_loss": 0.1098899319767952,
|
| 160 |
+
"learning_rate": 9.995793120629282e-06,
|
| 161 |
+
"loss": 0.3952847719192505,
|
| 162 |
+
"recon_loss": 1.4629768133163452,
|
| 163 |
+
"recon_trunc": 0.0,
|
| 164 |
+
"step": 65,
|
| 165 |
+
"student_loss": 0.03704734891653061,
|
| 166 |
+
"teacher_loss": 0.014628535136580467
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.020184544405997693,
|
| 170 |
+
"grad_norm": 5.0625,
|
| 171 |
+
"kd_loss": 0.06907039880752563,
|
| 172 |
+
"learning_rate": 9.993288199121283e-06,
|
| 173 |
+
"loss": 0.3567195892333984,
|
| 174 |
+
"recon_loss": 1.3470016717910767,
|
| 175 |
+
"recon_trunc": 0.0,
|
| 176 |
+
"step": 70,
|
| 177 |
+
"student_loss": 0.010167889297008514,
|
| 178 |
+
"teacher_loss": 0.005847702734172344
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.02162629757785467,
|
| 182 |
+
"grad_norm": 4.21875,
|
| 183 |
+
"kd_loss": 0.09832968562841415,
|
| 184 |
+
"learning_rate": 9.990201313529995e-06,
|
| 185 |
+
"loss": 0.3472292900085449,
|
| 186 |
+
"recon_loss": 1.8702830076217651,
|
| 187 |
+
"recon_trunc": 0.0,
|
| 188 |
+
"step": 75,
|
| 189 |
+
"student_loss": 0.030455006286501884,
|
| 190 |
+
"teacher_loss": 0.01826225221157074
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.02306805074971165,
|
| 194 |
+
"grad_norm": 4.125,
|
| 195 |
+
"kd_loss": 0.04121251404285431,
|
| 196 |
+
"learning_rate": 9.986532823629675e-06,
|
| 197 |
+
"loss": 0.3332221984863281,
|
| 198 |
+
"recon_loss": 3.242809772491455,
|
| 199 |
+
"recon_trunc": 0.0,
|
| 200 |
+
"step": 80,
|
| 201 |
+
"student_loss": 0.0050420514307916164,
|
| 202 |
+
"teacher_loss": 0.003938032314181328
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.024509803921568627,
|
| 206 |
+
"grad_norm": 3.28125,
|
| 207 |
+
"kd_loss": 0.033844396471977234,
|
| 208 |
+
"learning_rate": 9.982283156980133e-06,
|
| 209 |
+
"loss": 0.29645652770996095,
|
| 210 |
+
"recon_loss": 1.9862295389175415,
|
| 211 |
+
"recon_trunc": 0.0,
|
| 212 |
+
"step": 85,
|
| 213 |
+
"student_loss": 0.01399241853505373,
|
| 214 |
+
"teacher_loss": 0.016125138849020004
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.025951557093425604,
|
| 218 |
+
"grad_norm": 3.453125,
|
| 219 |
+
"kd_loss": 0.0408337339758873,
|
| 220 |
+
"learning_rate": 9.977452808876907e-06,
|
| 221 |
+
"loss": 0.2894145011901855,
|
| 222 |
+
"recon_loss": 2.489513874053955,
|
| 223 |
+
"recon_trunc": 0.0,
|
| 224 |
+
"step": 90,
|
| 225 |
+
"student_loss": 0.030976219102740288,
|
| 226 |
+
"teacher_loss": 0.014120617881417274
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.027393310265282585,
|
| 230 |
+
"grad_norm": 3.3125,
|
| 231 |
+
"kd_loss": 0.06854432076215744,
|
| 232 |
+
"learning_rate": 9.972042342293531e-06,
|
| 233 |
+
"loss": 0.27684826850891114,
|
| 234 |
+
"recon_loss": 1.0546698570251465,
|
| 235 |
+
"recon_trunc": 0.0,
|
| 236 |
+
"step": 95,
|
| 237 |
+
"student_loss": 0.029786942526698112,
|
| 238 |
+
"teacher_loss": 0.020987942814826965
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.02883506343713956,
|
| 242 |
+
"grad_norm": 3.25,
|
| 243 |
+
"kd_loss": 0.022570092231035233,
|
| 244 |
+
"learning_rate": 9.966052387815923e-06,
|
| 245 |
+
"loss": 0.27300400733947755,
|
| 246 |
+
"recon_loss": 2.099252462387085,
|
| 247 |
+
"recon_trunc": 0.0,
|
| 248 |
+
"step": 100,
|
| 249 |
+
"student_loss": 0.016591031104326248,
|
| 250 |
+
"teacher_loss": 0.007802738808095455
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.03027681660899654,
|
| 254 |
+
"grad_norm": 3.53125,
|
| 255 |
+
"kd_loss": 0.04893828183412552,
|
| 256 |
+
"learning_rate": 9.959483643568893e-06,
|
| 257 |
+
"loss": 0.2750415325164795,
|
| 258 |
+
"recon_loss": 0.9457756876945496,
|
| 259 |
+
"recon_trunc": 0.0,
|
| 260 |
+
"step": 105,
|
| 261 |
+
"student_loss": 0.05499906837940216,
|
| 262 |
+
"teacher_loss": 0.013182085007429123
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.031718569780853516,
|
| 266 |
+
"grad_norm": 2.75,
|
| 267 |
+
"kd_loss": 0.02219642885029316,
|
| 268 |
+
"learning_rate": 9.95233687513477e-06,
|
| 269 |
+
"loss": 0.26921510696411133,
|
| 270 |
+
"recon_loss": 3.6715874671936035,
|
| 271 |
+
"recon_trunc": 0.0,
|
| 272 |
+
"step": 110,
|
| 273 |
+
"student_loss": 0.01562587358057499,
|
| 274 |
+
"teacher_loss": 0.008027992211282253
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.03316032295271049,
|
| 278 |
+
"grad_norm": 2.734375,
|
| 279 |
+
"kd_loss": 0.017121493816375732,
|
| 280 |
+
"learning_rate": 9.944612915464183e-06,
|
| 281 |
+
"loss": 0.2387718915939331,
|
| 282 |
+
"recon_loss": 2.4717891216278076,
|
| 283 |
+
"recon_trunc": 0.0,
|
| 284 |
+
"step": 115,
|
| 285 |
+
"student_loss": 0.005411922931671143,
|
| 286 |
+
"teacher_loss": 0.031825024634599686
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.03460207612456748,
|
| 290 |
+
"grad_norm": 2.375,
|
| 291 |
+
"kd_loss": 0.04625450447201729,
|
| 292 |
+
"learning_rate": 9.936312664778978e-06,
|
| 293 |
+
"loss": 0.2392667293548584,
|
| 294 |
+
"recon_loss": 1.0688096284866333,
|
| 295 |
+
"recon_trunc": 0.0,
|
| 296 |
+
"step": 120,
|
| 297 |
+
"student_loss": 0.029523475095629692,
|
| 298 |
+
"teacher_loss": 0.009307589381933212
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.036043829296424454,
|
| 302 |
+
"grad_norm": 2.578125,
|
| 303 |
+
"kd_loss": 0.05075271427631378,
|
| 304 |
+
"learning_rate": 9.92743709046729e-06,
|
| 305 |
+
"loss": 0.2527806758880615,
|
| 306 |
+
"recon_loss": 1.899649739265442,
|
| 307 |
+
"recon_trunc": 0.0,
|
| 308 |
+
"step": 125,
|
| 309 |
+
"student_loss": 0.051832377910614014,
|
| 310 |
+
"teacher_loss": 0.015544950030744076
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.03748558246828143,
|
| 314 |
+
"grad_norm": 2.15625,
|
| 315 |
+
"kd_loss": 0.04277535527944565,
|
| 316 |
+
"learning_rate": 9.917987226970811e-06,
|
| 317 |
+
"loss": 0.2523399114608765,
|
| 318 |
+
"recon_loss": 2.138918161392212,
|
| 319 |
+
"recon_trunc": 0.0,
|
| 320 |
+
"step": 130,
|
| 321 |
+
"student_loss": 0.03786858543753624,
|
| 322 |
+
"teacher_loss": 0.019718820229172707
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.03892733564013841,
|
| 326 |
+
"grad_norm": 2.4375,
|
| 327 |
+
"kd_loss": 0.04147832468152046,
|
| 328 |
+
"learning_rate": 9.907964175664207e-06,
|
| 329 |
+
"loss": 0.2411898136138916,
|
| 330 |
+
"recon_loss": 1.1139979362487793,
|
| 331 |
+
"recon_trunc": 0.0,
|
| 332 |
+
"step": 135,
|
| 333 |
+
"student_loss": 0.012552146799862385,
|
| 334 |
+
"teacher_loss": 0.016261562705039978
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 0.040369088811995385,
|
| 338 |
+
"grad_norm": 2.453125,
|
| 339 |
+
"kd_loss": 0.03272290155291557,
|
| 340 |
+
"learning_rate": 9.897369104726772e-06,
|
| 341 |
+
"loss": 0.2410731077194214,
|
| 342 |
+
"recon_loss": 0.8702335357666016,
|
| 343 |
+
"recon_trunc": 0.0,
|
| 344 |
+
"step": 140,
|
| 345 |
+
"student_loss": 0.02782317064702511,
|
| 346 |
+
"teacher_loss": 0.032065339386463165
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.04181084198385236,
|
| 350 |
+
"grad_norm": 2.03125,
|
| 351 |
+
"kd_loss": 0.014335349202156067,
|
| 352 |
+
"learning_rate": 9.886203249006265e-06,
|
| 353 |
+
"loss": 0.2244941234588623,
|
| 354 |
+
"recon_loss": 2.9187490940093994,
|
| 355 |
+
"recon_trunc": 0.0,
|
| 356 |
+
"step": 145,
|
| 357 |
+
"student_loss": 0.0066062710247933865,
|
| 358 |
+
"teacher_loss": 0.006205227691680193
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 0.04325259515570934,
|
| 362 |
+
"grad_norm": 2.265625,
|
| 363 |
+
"kd_loss": 0.015548811294138432,
|
| 364 |
+
"learning_rate": 9.874467909874994e-06,
|
| 365 |
+
"loss": 0.23432111740112305,
|
| 366 |
+
"recon_loss": 2.134535789489746,
|
| 367 |
+
"recon_trunc": 0.0,
|
| 368 |
+
"step": 150,
|
| 369 |
+
"student_loss": 0.0055591329000890255,
|
| 370 |
+
"teacher_loss": 0.03213706240057945
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.04469434832756632,
|
| 374 |
+
"grad_norm": 4.34375,
|
| 375 |
+
"kd_loss": 0.05158035829663277,
|
| 376 |
+
"learning_rate": 9.862164455078141e-06,
|
| 377 |
+
"loss": 0.24995408058166504,
|
| 378 |
+
"recon_loss": 1.552893042564392,
|
| 379 |
+
"recon_trunc": 0.0,
|
| 380 |
+
"step": 155,
|
| 381 |
+
"student_loss": 0.032782189548015594,
|
| 382 |
+
"teacher_loss": 0.022193461656570435
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.0461361014994233,
|
| 386 |
+
"grad_norm": 2.5625,
|
| 387 |
+
"kd_loss": 0.021037286147475243,
|
| 388 |
+
"learning_rate": 9.849294318574353e-06,
|
| 389 |
+
"loss": 0.22788116931915284,
|
| 390 |
+
"recon_loss": 1.3483701944351196,
|
| 391 |
+
"recon_trunc": 0.0,
|
| 392 |
+
"step": 160,
|
| 393 |
+
"student_loss": 0.00251950160600245,
|
| 394 |
+
"teacher_loss": 0.0016650439938530326
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.04757785467128028,
|
| 398 |
+
"grad_norm": 3.484375,
|
| 399 |
+
"kd_loss": 0.019214345142245293,
|
| 400 |
+
"learning_rate": 9.835859000368615e-06,
|
| 401 |
+
"loss": 0.22725825309753417,
|
| 402 |
+
"recon_loss": 0.6318182349205017,
|
| 403 |
+
"recon_trunc": 0.0,
|
| 404 |
+
"step": 165,
|
| 405 |
+
"student_loss": 0.01272663101553917,
|
| 406 |
+
"teacher_loss": 0.004609123338013887
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.049019607843137254,
|
| 410 |
+
"grad_norm": 2.8125,
|
| 411 |
+
"kd_loss": 0.024736283347010612,
|
| 412 |
+
"learning_rate": 9.821860066337427e-06,
|
| 413 |
+
"loss": 0.22833685874938964,
|
| 414 |
+
"recon_loss": 1.4060359001159668,
|
| 415 |
+
"recon_trunc": 0.0,
|
| 416 |
+
"step": 170,
|
| 417 |
+
"student_loss": 0.009885711595416069,
|
| 418 |
+
"teacher_loss": 0.011500407010316849
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.05046136101499423,
|
| 422 |
+
"grad_norm": 2.5625,
|
| 423 |
+
"kd_loss": 0.03728032112121582,
|
| 424 |
+
"learning_rate": 9.807299148046301e-06,
|
| 425 |
+
"loss": 0.24652998447418212,
|
| 426 |
+
"recon_loss": 0.8636529445648193,
|
| 427 |
+
"recon_trunc": 0.0,
|
| 428 |
+
"step": 175,
|
| 429 |
+
"student_loss": 0.04721742123365402,
|
| 430 |
+
"teacher_loss": 0.033105362206697464
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.05190311418685121,
|
| 434 |
+
"grad_norm": 2.421875,
|
| 435 |
+
"kd_loss": 0.014192317612469196,
|
| 436 |
+
"learning_rate": 9.792177942559598e-06,
|
| 437 |
+
"loss": 0.225426983833313,
|
| 438 |
+
"recon_loss": 2.038878917694092,
|
| 439 |
+
"recon_trunc": 0.0,
|
| 440 |
+
"step": 180,
|
| 441 |
+
"student_loss": 0.0021350744646042585,
|
| 442 |
+
"teacher_loss": 0.001815832918509841
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.05334486735870819,
|
| 446 |
+
"grad_norm": 2.390625,
|
| 447 |
+
"kd_loss": 0.03198679909110069,
|
| 448 |
+
"learning_rate": 9.776498212242748e-06,
|
| 449 |
+
"loss": 0.22086071968078613,
|
| 450 |
+
"recon_loss": 0.8953023552894592,
|
| 451 |
+
"recon_trunc": 0.0,
|
| 452 |
+
"step": 185,
|
| 453 |
+
"student_loss": 0.043447788804769516,
|
| 454 |
+
"teacher_loss": 0.019889231771230698
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.05478662053056517,
|
| 458 |
+
"grad_norm": 2.546875,
|
| 459 |
+
"kd_loss": 0.046611037105321884,
|
| 460 |
+
"learning_rate": 9.76026178455684e-06,
|
| 461 |
+
"loss": 0.2095113515853882,
|
| 462 |
+
"recon_loss": 1.5233120918273926,
|
| 463 |
+
"recon_trunc": 0.0,
|
| 464 |
+
"step": 190,
|
| 465 |
+
"student_loss": 0.014042397029697895,
|
| 466 |
+
"teacher_loss": 0.0016930525889620185
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.056228373702422146,
|
| 470 |
+
"grad_norm": 2.515625,
|
| 471 |
+
"kd_loss": 0.018354102969169617,
|
| 472 |
+
"learning_rate": 9.743470551845625e-06,
|
| 473 |
+
"loss": 0.20453329086303712,
|
| 474 |
+
"recon_loss": 0.9134235978126526,
|
| 475 |
+
"recon_trunc": 0.0,
|
| 476 |
+
"step": 195,
|
| 477 |
+
"student_loss": 0.0058402433060109615,
|
| 478 |
+
"teacher_loss": 0.004270657431334257
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.05767012687427912,
|
| 482 |
+
"grad_norm": 2.546875,
|
| 483 |
+
"kd_loss": 0.029964027926325798,
|
| 484 |
+
"learning_rate": 9.726126471114985e-06,
|
| 485 |
+
"loss": 0.21229453086853028,
|
| 486 |
+
"recon_loss": 1.0529005527496338,
|
| 487 |
+
"recon_trunc": 0.0,
|
| 488 |
+
"step": 200,
|
| 489 |
+
"student_loss": 0.0065546901896595955,
|
| 490 |
+
"teacher_loss": 0.004668537527322769
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.0591118800461361,
|
| 494 |
+
"grad_norm": 2.328125,
|
| 495 |
+
"kd_loss": 0.016300605610013008,
|
| 496 |
+
"learning_rate": 9.708231563804828e-06,
|
| 497 |
+
"loss": 0.22150065898895263,
|
| 498 |
+
"recon_loss": 3.866762399673462,
|
| 499 |
+
"recon_trunc": 0.0,
|
| 500 |
+
"step": 205,
|
| 501 |
+
"student_loss": 0.011293341405689716,
|
| 502 |
+
"teacher_loss": 0.012114761397242546
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 0.06055363321799308,
|
| 506 |
+
"grad_norm": 3.46875,
|
| 507 |
+
"kd_loss": 0.014887100085616112,
|
| 508 |
+
"learning_rate": 9.689787915553497e-06,
|
| 509 |
+
"loss": 0.2299870491027832,
|
| 510 |
+
"recon_loss": 0.7259498834609985,
|
| 511 |
+
"recon_trunc": 0.0,
|
| 512 |
+
"step": 210,
|
| 513 |
+
"student_loss": 0.008276550099253654,
|
| 514 |
+
"teacher_loss": 0.008154639042913914
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.061995386389850055,
|
| 518 |
+
"grad_norm": 2.546875,
|
| 519 |
+
"kd_loss": 0.025979718193411827,
|
| 520 |
+
"learning_rate": 9.670797675954693e-06,
|
| 521 |
+
"loss": 0.21829309463500976,
|
| 522 |
+
"recon_loss": 0.9380886554718018,
|
| 523 |
+
"recon_trunc": 0.0,
|
| 524 |
+
"step": 215,
|
| 525 |
+
"student_loss": 0.018656907603144646,
|
| 526 |
+
"teacher_loss": 0.025025509297847748
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.06343713956170703,
|
| 530 |
+
"grad_norm": 2.359375,
|
| 531 |
+
"kd_loss": 0.02608303911983967,
|
| 532 |
+
"learning_rate": 9.651263058306932e-06,
|
| 533 |
+
"loss": 0.20896072387695314,
|
| 534 |
+
"recon_loss": 1.2348960638046265,
|
| 535 |
+
"recon_trunc": 0.0,
|
| 536 |
+
"step": 220,
|
| 537 |
+
"student_loss": 0.007329518906772137,
|
| 538 |
+
"teacher_loss": 0.016939902678132057
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.06487889273356401,
|
| 542 |
+
"grad_norm": 1.890625,
|
| 543 |
+
"kd_loss": 0.018794769421219826,
|
| 544 |
+
"learning_rate": 9.631186339355599e-06,
|
| 545 |
+
"loss": 0.19410723447799683,
|
| 546 |
+
"recon_loss": 0.9309818148612976,
|
| 547 |
+
"recon_trunc": 0.0,
|
| 548 |
+
"step": 225,
|
| 549 |
+
"student_loss": 0.0014466241700574756,
|
| 550 |
+
"teacher_loss": 0.001579243573360145
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.06632064590542099,
|
| 554 |
+
"grad_norm": 3.0625,
|
| 555 |
+
"kd_loss": 0.02463342435657978,
|
| 556 |
+
"learning_rate": 9.610569859027585e-06,
|
| 557 |
+
"loss": 0.21999239921569824,
|
| 558 |
+
"recon_loss": 1.6447029113769531,
|
| 559 |
+
"recon_trunc": 0.0,
|
| 560 |
+
"step": 230,
|
| 561 |
+
"student_loss": 0.05218880996108055,
|
| 562 |
+
"teacher_loss": 0.04487629234790802
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.06776239907727798,
|
| 566 |
+
"grad_norm": 2.546875,
|
| 567 |
+
"kd_loss": 0.025545483455061913,
|
| 568 |
+
"learning_rate": 9.589416020158577e-06,
|
| 569 |
+
"loss": 0.19575120210647584,
|
| 570 |
+
"recon_loss": 0.6338058710098267,
|
| 571 |
+
"recon_trunc": 0.0,
|
| 572 |
+
"step": 235,
|
| 573 |
+
"student_loss": 0.013113539665937424,
|
| 574 |
+
"teacher_loss": 0.01345077808946371
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 0.06920415224913495,
|
| 578 |
+
"grad_norm": 2.453125,
|
| 579 |
+
"kd_loss": 0.030363285914063454,
|
| 580 |
+
"learning_rate": 9.567727288213005e-06,
|
| 581 |
+
"loss": 0.20205872058868407,
|
| 582 |
+
"recon_loss": 0.9078859686851501,
|
| 583 |
+
"recon_trunc": 0.0,
|
| 584 |
+
"step": 240,
|
| 585 |
+
"student_loss": 0.012684637680649757,
|
| 586 |
+
"teacher_loss": 0.013201592490077019
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.07064590542099193,
|
| 590 |
+
"grad_norm": 2.28125,
|
| 591 |
+
"kd_loss": 0.030748743563890457,
|
| 592 |
+
"learning_rate": 9.545506190996695e-06,
|
| 593 |
+
"loss": 0.18676055669784547,
|
| 594 |
+
"recon_loss": 0.6405854225158691,
|
| 595 |
+
"recon_trunc": 0.0,
|
| 596 |
+
"step": 245,
|
| 597 |
+
"student_loss": 0.004571032710373402,
|
| 598 |
+
"teacher_loss": 0.008879352360963821
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.07208765859284891,
|
| 602 |
+
"grad_norm": 4.125,
|
| 603 |
+
"kd_loss": 0.01464357040822506,
|
| 604 |
+
"learning_rate": 9.52275531836226e-06,
|
| 605 |
+
"loss": 0.21321358680725097,
|
| 606 |
+
"recon_loss": 1.3516043424606323,
|
| 607 |
+
"recon_trunc": 0.0,
|
| 608 |
+
"step": 250,
|
| 609 |
+
"student_loss": 0.056694742292165756,
|
| 610 |
+
"teacher_loss": 0.0598812997341156
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.07352941176470588,
|
| 614 |
+
"grad_norm": 1.921875,
|
| 615 |
+
"kd_loss": 0.033974818885326385,
|
| 616 |
+
"learning_rate": 9.499477321907243e-06,
|
| 617 |
+
"loss": 0.1943159818649292,
|
| 618 |
+
"recon_loss": 0.6920376420021057,
|
| 619 |
+
"recon_trunc": 0.0,
|
| 620 |
+
"step": 255,
|
| 621 |
+
"student_loss": 0.027635732665657997,
|
| 622 |
+
"teacher_loss": 0.013332012109458447
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.07497116493656286,
|
| 626 |
+
"grad_norm": 7.15625,
|
| 627 |
+
"kd_loss": 0.026340534910559654,
|
| 628 |
+
"learning_rate": 9.475674914665084e-06,
|
| 629 |
+
"loss": 0.2081134557723999,
|
| 630 |
+
"recon_loss": 1.395743489265442,
|
| 631 |
+
"recon_trunc": 0.20000000298023224,
|
| 632 |
+
"step": 260,
|
| 633 |
+
"student_loss": 0.02220655232667923,
|
| 634 |
+
"teacher_loss": 0.00963656883686781
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.07641291810841984,
|
| 638 |
+
"grad_norm": 2.71875,
|
| 639 |
+
"kd_loss": 0.012761188670992851,
|
| 640 |
+
"learning_rate": 9.451350870788922e-06,
|
| 641 |
+
"loss": 0.19083361625671386,
|
| 642 |
+
"recon_loss": 2.5343592166900635,
|
| 643 |
+
"recon_trunc": 0.0,
|
| 644 |
+
"step": 265,
|
| 645 |
+
"student_loss": 0.0085363257676363,
|
| 646 |
+
"teacher_loss": 0.009733619168400764
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.07785467128027682,
|
| 650 |
+
"grad_norm": 1.9140625,
|
| 651 |
+
"kd_loss": 0.028379660099744797,
|
| 652 |
+
"learning_rate": 9.426508025228255e-06,
|
| 653 |
+
"loss": 0.2052919864654541,
|
| 654 |
+
"recon_loss": 0.8769041299819946,
|
| 655 |
+
"recon_trunc": 0.0,
|
| 656 |
+
"step": 270,
|
| 657 |
+
"student_loss": 0.04869980365037918,
|
| 658 |
+
"teacher_loss": 0.04083171859383583
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.07929642445213379,
|
| 662 |
+
"grad_norm": 4.46875,
|
| 663 |
+
"kd_loss": 0.013499215245246887,
|
| 664 |
+
"learning_rate": 9.401149273398547e-06,
|
| 665 |
+
"loss": 0.2169323205947876,
|
| 666 |
+
"recon_loss": 1.631927251815796,
|
| 667 |
+
"recon_trunc": 0.0,
|
| 668 |
+
"step": 275,
|
| 669 |
+
"student_loss": 0.07303008437156677,
|
| 670 |
+
"teacher_loss": 0.05521775409579277
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 0.08073817762399077,
|
| 674 |
+
"grad_norm": 3.328125,
|
| 675 |
+
"kd_loss": 0.025790857151150703,
|
| 676 |
+
"learning_rate": 9.37527757084375e-06,
|
| 677 |
+
"loss": 0.2107837200164795,
|
| 678 |
+
"recon_loss": 1.9519470930099487,
|
| 679 |
+
"recon_trunc": 0.0,
|
| 680 |
+
"step": 280,
|
| 681 |
+
"student_loss": 0.01667192578315735,
|
| 682 |
+
"teacher_loss": 0.008985173888504505
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.08217993079584775,
|
| 686 |
+
"grad_norm": 3.140625,
|
| 687 |
+
"kd_loss": 0.019596580415964127,
|
| 688 |
+
"learning_rate": 9.348895932891857e-06,
|
| 689 |
+
"loss": 0.18018556833267213,
|
| 690 |
+
"recon_loss": 1.4045679569244385,
|
| 691 |
+
"recon_trunc": 0.0,
|
| 692 |
+
"step": 285,
|
| 693 |
+
"student_loss": 0.044362958520650864,
|
| 694 |
+
"teacher_loss": 0.025971855968236923
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 0.08362168396770472,
|
| 698 |
+
"grad_norm": 4.0625,
|
| 699 |
+
"kd_loss": 0.021219631657004356,
|
| 700 |
+
"learning_rate": 9.322007434303449e-06,
|
| 701 |
+
"loss": 0.18771085739135743,
|
| 702 |
+
"recon_loss": 0.9686179161071777,
|
| 703 |
+
"recon_trunc": 0.0,
|
| 704 |
+
"step": 290,
|
| 705 |
+
"student_loss": 0.027421345934271812,
|
| 706 |
+
"teacher_loss": 0.025595808401703835
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.0850634371395617,
|
| 710 |
+
"grad_norm": 2.84375,
|
| 711 |
+
"kd_loss": 0.02482793666422367,
|
| 712 |
+
"learning_rate": 9.29461520891335e-06,
|
| 713 |
+
"loss": 0.20411481857299804,
|
| 714 |
+
"recon_loss": 0.8968526124954224,
|
| 715 |
+
"recon_trunc": 0.0,
|
| 716 |
+
"step": 295,
|
| 717 |
+
"student_loss": 0.002501523122191429,
|
| 718 |
+
"teacher_loss": 0.0010767965577542782
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.08650519031141868,
|
| 722 |
+
"grad_norm": 2.390625,
|
| 723 |
+
"kd_loss": 0.02157168835401535,
|
| 724 |
+
"learning_rate": 9.266722449265371e-06,
|
| 725 |
+
"loss": 0.18465697765350342,
|
| 726 |
+
"recon_loss": 1.2718404531478882,
|
| 727 |
+
"recon_trunc": 0.0,
|
| 728 |
+
"step": 300,
|
| 729 |
+
"student_loss": 0.07626867294311523,
|
| 730 |
+
"teacher_loss": 0.0482635572552681
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.08794694348327567,
|
| 734 |
+
"grad_norm": 2.375,
|
| 735 |
+
"kd_loss": 0.02058529667556286,
|
| 736 |
+
"learning_rate": 9.238332406240224e-06,
|
| 737 |
+
"loss": 0.18665231466293336,
|
| 738 |
+
"recon_loss": 0.9268073439598083,
|
| 739 |
+
"recon_trunc": 0.0,
|
| 740 |
+
"step": 305,
|
| 741 |
+
"student_loss": 0.0018058717250823975,
|
| 742 |
+
"teacher_loss": 0.0018525687046349049
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 0.08938869665513265,
|
| 746 |
+
"grad_norm": 2.5625,
|
| 747 |
+
"kd_loss": 0.03731205314397812,
|
| 748 |
+
"learning_rate": 9.209448388676636e-06,
|
| 749 |
+
"loss": 0.19349218606948854,
|
| 750 |
+
"recon_loss": 0.9608579874038696,
|
| 751 |
+
"recon_trunc": 0.0,
|
| 752 |
+
"step": 310,
|
| 753 |
+
"student_loss": 0.008257206529378891,
|
| 754 |
+
"teacher_loss": 0.0030505042523145676
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.09083044982698962,
|
| 758 |
+
"grad_norm": 2.609375,
|
| 759 |
+
"kd_loss": 0.028524678200483322,
|
| 760 |
+
"learning_rate": 9.180073762985701e-06,
|
| 761 |
+
"loss": 0.19262093305587769,
|
| 762 |
+
"recon_loss": 0.6517610549926758,
|
| 763 |
+
"recon_trunc": 0.0,
|
| 764 |
+
"step": 315,
|
| 765 |
+
"student_loss": 0.022836260497570038,
|
| 766 |
+
"teacher_loss": 0.01476720068603754
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.0922722029988466,
|
| 770 |
+
"grad_norm": 2.640625,
|
| 771 |
+
"kd_loss": 0.02896862104535103,
|
| 772 |
+
"learning_rate": 9.15021195275854e-06,
|
| 773 |
+
"loss": 0.1821727752685547,
|
| 774 |
+
"recon_loss": 0.7273185849189758,
|
| 775 |
+
"recon_trunc": 0.0,
|
| 776 |
+
"step": 320,
|
| 777 |
+
"student_loss": 0.03774423152208328,
|
| 778 |
+
"teacher_loss": 0.011754393577575684
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.09371395617070358,
|
| 782 |
+
"grad_norm": 2.828125,
|
| 783 |
+
"kd_loss": 0.022670011967420578,
|
| 784 |
+
"learning_rate": 9.119866438367263e-06,
|
| 785 |
+
"loss": 0.19521080255508422,
|
| 786 |
+
"recon_loss": 1.036689281463623,
|
| 787 |
+
"recon_trunc": 0.0,
|
| 788 |
+
"step": 325,
|
| 789 |
+
"student_loss": 0.05931982025504112,
|
| 790 |
+
"teacher_loss": 0.007148706819862127
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 0.09515570934256055,
|
| 794 |
+
"grad_norm": 2.640625,
|
| 795 |
+
"kd_loss": 0.012179434299468994,
|
| 796 |
+
"learning_rate": 9.089040756559352e-06,
|
| 797 |
+
"loss": 0.1976340651512146,
|
| 798 |
+
"recon_loss": 0.6355065107345581,
|
| 799 |
+
"recon_trunc": 0.0,
|
| 800 |
+
"step": 330,
|
| 801 |
+
"student_loss": 0.001890555489808321,
|
| 802 |
+
"teacher_loss": 0.0018264790996909142
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.09659746251441753,
|
| 806 |
+
"grad_norm": 2.84375,
|
| 807 |
+
"kd_loss": 0.02101888321340084,
|
| 808 |
+
"learning_rate": 9.057738500045453e-06,
|
| 809 |
+
"loss": 0.18308634757995607,
|
| 810 |
+
"recon_loss": 0.7007512450218201,
|
| 811 |
+
"recon_trunc": 0.0,
|
| 812 |
+
"step": 335,
|
| 813 |
+
"student_loss": 0.01875852793455124,
|
| 814 |
+
"teacher_loss": 0.013294468633830547
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.09803921568627451,
|
| 818 |
+
"grad_norm": 2.015625,
|
| 819 |
+
"kd_loss": 0.028847362846136093,
|
| 820 |
+
"learning_rate": 9.025963317080641e-06,
|
| 821 |
+
"loss": 0.18786072731018066,
|
| 822 |
+
"recon_loss": 0.6635362505912781,
|
| 823 |
+
"recon_trunc": 0.0,
|
| 824 |
+
"step": 340,
|
| 825 |
+
"student_loss": 0.014487682841718197,
|
| 826 |
+
"teacher_loss": 0.004602581728249788
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.09948096885813149,
|
| 830 |
+
"grad_norm": 3.015625,
|
| 831 |
+
"kd_loss": 0.019740767776966095,
|
| 832 |
+
"learning_rate": 8.993718911039221e-06,
|
| 833 |
+
"loss": 0.18705078363418579,
|
| 834 |
+
"recon_loss": 1.7711620330810547,
|
| 835 |
+
"recon_trunc": 0.0,
|
| 836 |
+
"step": 345,
|
| 837 |
+
"student_loss": 0.0453316755592823,
|
| 838 |
+
"teacher_loss": 0.09916361421346664
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 0.10092272202998846,
|
| 842 |
+
"grad_norm": 3.109375,
|
| 843 |
+
"kd_loss": 0.028285956010222435,
|
| 844 |
+
"learning_rate": 8.96100903998311e-06,
|
| 845 |
+
"loss": 0.18774936199188233,
|
| 846 |
+
"recon_loss": 1.4156874418258667,
|
| 847 |
+
"recon_trunc": 0.0,
|
| 848 |
+
"step": 350,
|
| 849 |
+
"student_loss": 0.02220643311738968,
|
| 850 |
+
"teacher_loss": 0.005953202955424786
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.10236447520184544,
|
| 854 |
+
"grad_norm": 4.03125,
|
| 855 |
+
"kd_loss": 0.02711368165910244,
|
| 856 |
+
"learning_rate": 8.927837516223824e-06,
|
| 857 |
+
"loss": 0.18523555994033813,
|
| 858 |
+
"recon_loss": 1.2790436744689941,
|
| 859 |
+
"recon_trunc": 0.0,
|
| 860 |
+
"step": 355,
|
| 861 |
+
"student_loss": 0.05644513666629791,
|
| 862 |
+
"teacher_loss": 0.039577875286340714
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.10380622837370242,
|
| 866 |
+
"grad_norm": 2.15625,
|
| 867 |
+
"kd_loss": 0.02200433798134327,
|
| 868 |
+
"learning_rate": 8.89420820587817e-06,
|
| 869 |
+
"loss": 0.1873791217803955,
|
| 870 |
+
"recon_loss": 0.9904807806015015,
|
| 871 |
+
"recon_trunc": 0.0555555559694767,
|
| 872 |
+
"step": 360,
|
| 873 |
+
"student_loss": 0.004485330078750849,
|
| 874 |
+
"teacher_loss": 0.004191179294139147
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.1052479815455594,
|
| 878 |
+
"grad_norm": 2.390625,
|
| 879 |
+
"kd_loss": 0.02350509539246559,
|
| 880 |
+
"learning_rate": 8.86012502841764e-06,
|
| 881 |
+
"loss": 0.17797458171844482,
|
| 882 |
+
"recon_loss": 0.7470078468322754,
|
| 883 |
+
"recon_trunc": 0.0,
|
| 884 |
+
"step": 365,
|
| 885 |
+
"student_loss": 0.027660565450787544,
|
| 886 |
+
"teacher_loss": 0.06122901663184166
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.10668973471741638,
|
| 890 |
+
"grad_norm": 1.9921875,
|
| 891 |
+
"kd_loss": 0.014148482121527195,
|
| 892 |
+
"learning_rate": 8.825591956211614e-06,
|
| 893 |
+
"loss": 0.17095459699630738,
|
| 894 |
+
"recon_loss": 1.0832712650299072,
|
| 895 |
+
"recon_trunc": 0.0,
|
| 896 |
+
"step": 370,
|
| 897 |
+
"student_loss": 0.08812573552131653,
|
| 898 |
+
"teacher_loss": 0.010357359424233437
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.10813148788927336,
|
| 902 |
+
"grad_norm": 2.484375,
|
| 903 |
+
"kd_loss": 0.03929801285266876,
|
| 904 |
+
"learning_rate": 8.790613014064371e-06,
|
| 905 |
+
"loss": 0.16936860084533692,
|
| 906 |
+
"recon_loss": 1.6295180320739746,
|
| 907 |
+
"recon_trunc": 0.0,
|
| 908 |
+
"step": 375,
|
| 909 |
+
"student_loss": 0.011690870858728886,
|
| 910 |
+
"teacher_loss": 0.09399953484535217
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 0.10957324106113034,
|
| 914 |
+
"grad_norm": 3.125,
|
| 915 |
+
"kd_loss": 0.018021779134869576,
|
| 916 |
+
"learning_rate": 8.755192278746e-06,
|
| 917 |
+
"loss": 0.17092232704162597,
|
| 918 |
+
"recon_loss": 1.3323853015899658,
|
| 919 |
+
"recon_trunc": 0.0,
|
| 920 |
+
"step": 380,
|
| 921 |
+
"student_loss": 0.010468835942447186,
|
| 922 |
+
"teacher_loss": 0.009244673885405064
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.11101499423298732,
|
| 926 |
+
"grad_norm": 2.015625,
|
| 927 |
+
"kd_loss": 0.025558089837431908,
|
| 928 |
+
"learning_rate": 8.719333878517274e-06,
|
| 929 |
+
"loss": 0.15772898197174073,
|
| 930 |
+
"recon_loss": 0.369345486164093,
|
| 931 |
+
"recon_trunc": 0.0,
|
| 932 |
+
"step": 385,
|
| 933 |
+
"student_loss": 0.00318582053296268,
|
| 934 |
+
"teacher_loss": 0.0008088735630735755
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.11245674740484429,
|
| 938 |
+
"grad_norm": 2.1875,
|
| 939 |
+
"kd_loss": 0.01489198487251997,
|
| 940 |
+
"learning_rate": 8.683041992648474e-06,
|
| 941 |
+
"loss": 0.17867419719696045,
|
| 942 |
+
"recon_loss": 1.965988039970398,
|
| 943 |
+
"recon_trunc": 0.0,
|
| 944 |
+
"step": 390,
|
| 945 |
+
"student_loss": 0.05299150571227074,
|
| 946 |
+
"teacher_loss": 0.036044854670763016
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.11389850057670127,
|
| 950 |
+
"grad_norm": 3.5625,
|
| 951 |
+
"kd_loss": 0.015575695782899857,
|
| 952 |
+
"learning_rate": 8.646320850932326e-06,
|
| 953 |
+
"loss": 0.16602530479431152,
|
| 954 |
+
"recon_loss": 1.0856443643569946,
|
| 955 |
+
"recon_trunc": 0.0,
|
| 956 |
+
"step": 395,
|
| 957 |
+
"student_loss": 0.012944542802870274,
|
| 958 |
+
"teacher_loss": 0.009949937462806702
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 0.11534025374855825,
|
| 962 |
+
"grad_norm": 2.328125,
|
| 963 |
+
"kd_loss": 0.025076374411582947,
|
| 964 |
+
"learning_rate": 8.609174733191012e-06,
|
| 965 |
+
"loss": 0.19546109437942505,
|
| 966 |
+
"recon_loss": 0.8048983216285706,
|
| 967 |
+
"recon_trunc": 0.0,
|
| 968 |
+
"step": 400,
|
| 969 |
+
"student_loss": 0.019003333523869514,
|
| 970 |
+
"teacher_loss": 0.0027343458496034145
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.11678200692041522,
|
| 974 |
+
"grad_norm": 2.0,
|
| 975 |
+
"kd_loss": 0.011262590065598488,
|
| 976 |
+
"learning_rate": 8.571607968777349e-06,
|
| 977 |
+
"loss": 0.17050211429595946,
|
| 978 |
+
"recon_loss": 0.7520943284034729,
|
| 979 |
+
"recon_trunc": 0.0,
|
| 980 |
+
"step": 405,
|
| 981 |
+
"student_loss": 0.020238369703292847,
|
| 982 |
+
"teacher_loss": 0.016718050464987755
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.1182237600922722,
|
| 986 |
+
"grad_norm": 2.390625,
|
| 987 |
+
"kd_loss": 0.028635602444410324,
|
| 988 |
+
"learning_rate": 8.533624936070225e-06,
|
| 989 |
+
"loss": 0.15760048627853393,
|
| 990 |
+
"recon_loss": 0.9762012958526611,
|
| 991 |
+
"recon_trunc": 0.0,
|
| 992 |
+
"step": 410,
|
| 993 |
+
"student_loss": 0.10509705543518066,
|
| 994 |
+
"teacher_loss": 0.033958207815885544
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 0.11966551326412918,
|
| 998 |
+
"grad_norm": 3.015625,
|
| 999 |
+
"kd_loss": 0.010353137739002705,
|
| 1000 |
+
"learning_rate": 8.495230061964289e-06,
|
| 1001 |
+
"loss": 0.16627098321914674,
|
| 1002 |
+
"recon_loss": 1.469043254852295,
|
| 1003 |
+
"recon_trunc": 0.0,
|
| 1004 |
+
"step": 415,
|
| 1005 |
+
"student_loss": 0.0025488450191915035,
|
| 1006 |
+
"teacher_loss": 0.001737672952003777
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.12110726643598616,
|
| 1010 |
+
"grad_norm": 2.109375,
|
| 1011 |
+
"kd_loss": 0.018990106880664825,
|
| 1012 |
+
"learning_rate": 8.456427821354001e-06,
|
| 1013 |
+
"loss": 0.16616129875183105,
|
| 1014 |
+
"recon_loss": 0.9399528503417969,
|
| 1015 |
+
"recon_trunc": 0.0,
|
| 1016 |
+
"step": 420,
|
| 1017 |
+
"student_loss": 0.014011203311383724,
|
| 1018 |
+
"teacher_loss": 0.029460012912750244
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.12254901960784313,
|
| 1022 |
+
"grad_norm": 3.15625,
|
| 1023 |
+
"kd_loss": 0.010911962017416954,
|
| 1024 |
+
"learning_rate": 8.417222736612092e-06,
|
| 1025 |
+
"loss": 0.19283196926116944,
|
| 1026 |
+
"recon_loss": 1.1884320974349976,
|
| 1027 |
+
"recon_trunc": 0.0,
|
| 1028 |
+
"step": 425,
|
| 1029 |
+
"student_loss": 0.0011641188757494092,
|
| 1030 |
+
"teacher_loss": 0.0010047216201201081
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.12399077277970011,
|
| 1034 |
+
"grad_norm": 2.3125,
|
| 1035 |
+
"kd_loss": 0.017093488946557045,
|
| 1036 |
+
"learning_rate": 8.377619377062483e-06,
|
| 1037 |
+
"loss": 0.18180520534515382,
|
| 1038 |
+
"recon_loss": 1.1236519813537598,
|
| 1039 |
+
"recon_trunc": 0.0,
|
| 1040 |
+
"step": 430,
|
| 1041 |
+
"student_loss": 0.036160510033369064,
|
| 1042 |
+
"teacher_loss": 0.027615487575531006
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 0.1254325259515571,
|
| 1046 |
+
"grad_norm": 2.59375,
|
| 1047 |
+
"kd_loss": 0.01262976136058569,
|
| 1048 |
+
"learning_rate": 8.337622358447728e-06,
|
| 1049 |
+
"loss": 0.16819295883178711,
|
| 1050 |
+
"recon_loss": 1.2611421346664429,
|
| 1051 |
+
"recon_trunc": 0.0,
|
| 1052 |
+
"step": 435,
|
| 1053 |
+
"student_loss": 0.006829291116446257,
|
| 1054 |
+
"teacher_loss": 0.00490315118804574
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 0.12687427912341406,
|
| 1058 |
+
"grad_norm": 3.953125,
|
| 1059 |
+
"kd_loss": 0.011193808168172836,
|
| 1060 |
+
"learning_rate": 8.29723634239106e-06,
|
| 1061 |
+
"loss": 0.1792707085609436,
|
| 1062 |
+
"recon_loss": 1.015066146850586,
|
| 1063 |
+
"recon_trunc": 0.0,
|
| 1064 |
+
"step": 440,
|
| 1065 |
+
"student_loss": 0.014378616586327553,
|
| 1066 |
+
"teacher_loss": 0.007203614804893732
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.12831603229527105,
|
| 1070 |
+
"grad_norm": 5.90625,
|
| 1071 |
+
"kd_loss": 0.024301575496792793,
|
| 1072 |
+
"learning_rate": 8.256466035853077e-06,
|
| 1073 |
+
"loss": 0.16281299591064452,
|
| 1074 |
+
"recon_loss": 2.0686748027801514,
|
| 1075 |
+
"recon_trunc": 0.0,
|
| 1076 |
+
"step": 445,
|
| 1077 |
+
"student_loss": 0.03150418400764465,
|
| 1078 |
+
"teacher_loss": 0.007875236682593822
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 0.12975778546712802,
|
| 1082 |
+
"grad_norm": 2.234375,
|
| 1083 |
+
"kd_loss": 0.021582748740911484,
|
| 1084 |
+
"learning_rate": 8.215316190583153e-06,
|
| 1085 |
+
"loss": 0.16349010467529296,
|
| 1086 |
+
"recon_loss": 0.8926537036895752,
|
| 1087 |
+
"recon_trunc": 0.0,
|
| 1088 |
+
"step": 450,
|
| 1089 |
+
"student_loss": 0.02079097554087639,
|
| 1090 |
+
"teacher_loss": 0.01684233918786049
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 0.131199538638985,
|
| 1094 |
+
"grad_norm": 1.9609375,
|
| 1095 |
+
"kd_loss": 0.008602065965533257,
|
| 1096 |
+
"learning_rate": 8.173791602565622e-06,
|
| 1097 |
+
"loss": 0.17787095308303832,
|
| 1098 |
+
"recon_loss": 0.7228834629058838,
|
| 1099 |
+
"recon_trunc": 0.0,
|
| 1100 |
+
"step": 455,
|
| 1101 |
+
"student_loss": 0.0013961649965494871,
|
| 1102 |
+
"teacher_loss": 0.0011512988712638617
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.13264129181084197,
|
| 1106 |
+
"grad_norm": 2.5625,
|
| 1107 |
+
"kd_loss": 0.018057696521282196,
|
| 1108 |
+
"learning_rate": 8.13189711146081e-06,
|
| 1109 |
+
"loss": 0.180067777633667,
|
| 1110 |
+
"recon_loss": 1.1073039770126343,
|
| 1111 |
+
"recon_trunc": 0.0,
|
| 1112 |
+
"step": 460,
|
| 1113 |
+
"student_loss": 0.005878835450857878,
|
| 1114 |
+
"teacher_loss": 0.010166267864406109
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.13408304498269896,
|
| 1118 |
+
"grad_norm": 2.671875,
|
| 1119 |
+
"kd_loss": 0.013382851146161556,
|
| 1120 |
+
"learning_rate": 8.08963760004098e-06,
|
| 1121 |
+
"loss": 0.15073424577713013,
|
| 1122 |
+
"recon_loss": 0.23623330891132355,
|
| 1123 |
+
"recon_trunc": 0.0,
|
| 1124 |
+
"step": 465,
|
| 1125 |
+
"student_loss": 0.0007799513987265527,
|
| 1126 |
+
"teacher_loss": 0.0006509500090032816
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 0.13552479815455595,
|
| 1130 |
+
"grad_norm": 2.359375,
|
| 1131 |
+
"kd_loss": 0.011669280007481575,
|
| 1132 |
+
"learning_rate": 8.047017993621242e-06,
|
| 1133 |
+
"loss": 0.17336150407791137,
|
| 1134 |
+
"recon_loss": 2.228370428085327,
|
| 1135 |
+
"recon_trunc": 0.0,
|
| 1136 |
+
"step": 470,
|
| 1137 |
+
"student_loss": 0.007680488284677267,
|
| 1138 |
+
"teacher_loss": 0.005909646395593882
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 0.13696655132641292,
|
| 1142 |
+
"grad_norm": 2.25,
|
| 1143 |
+
"kd_loss": 0.010834953747689724,
|
| 1144 |
+
"learning_rate": 8.004043259485519e-06,
|
| 1145 |
+
"loss": 0.1676584243774414,
|
| 1146 |
+
"recon_loss": 3.427149534225464,
|
| 1147 |
+
"recon_trunc": 0.0,
|
| 1148 |
+
"step": 475,
|
| 1149 |
+
"student_loss": 0.002225591568276286,
|
| 1150 |
+
"teacher_loss": 0.002081170678138733
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.1384083044982699,
|
| 1154 |
+
"grad_norm": 2.109375,
|
| 1155 |
+
"kd_loss": 0.008530965074896812,
|
| 1156 |
+
"learning_rate": 7.960718406307602e-06,
|
| 1157 |
+
"loss": 0.1668088436126709,
|
| 1158 |
+
"recon_loss": 2.6840450763702393,
|
| 1159 |
+
"recon_trunc": 0.0,
|
| 1160 |
+
"step": 480,
|
| 1161 |
+
"student_loss": 0.001257926574908197,
|
| 1162 |
+
"teacher_loss": 0.0011718194000422955
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 0.13985005767012687,
|
| 1166 |
+
"grad_norm": 1.984375,
|
| 1167 |
+
"kd_loss": 0.02287101000547409,
|
| 1168 |
+
"learning_rate": 7.917048483567408e-06,
|
| 1169 |
+
"loss": 0.16856667995452881,
|
| 1170 |
+
"recon_loss": 1.235361933708191,
|
| 1171 |
+
"recon_trunc": 0.0,
|
| 1172 |
+
"step": 485,
|
| 1173 |
+
"student_loss": 0.005858918186277151,
|
| 1174 |
+
"teacher_loss": 0.0503169484436512
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 0.14129181084198386,
|
| 1178 |
+
"grad_norm": 2.0625,
|
| 1179 |
+
"kd_loss": 0.02468051016330719,
|
| 1180 |
+
"learning_rate": 7.873038580962453e-06,
|
| 1181 |
+
"loss": 0.18247668743133544,
|
| 1182 |
+
"recon_loss": 1.525710105895996,
|
| 1183 |
+
"recon_trunc": 0.0,
|
| 1184 |
+
"step": 490,
|
| 1185 |
+
"student_loss": 0.0058046127669513226,
|
| 1186 |
+
"teacher_loss": 0.006677514407783747
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.14273356401384082,
|
| 1190 |
+
"grad_norm": 1.640625,
|
| 1191 |
+
"kd_loss": 0.02074977196753025,
|
| 1192 |
+
"learning_rate": 7.828693827814662e-06,
|
| 1193 |
+
"loss": 0.15302610397338867,
|
| 1194 |
+
"recon_loss": 0.8546621799468994,
|
| 1195 |
+
"recon_trunc": 0.0,
|
| 1196 |
+
"step": 495,
|
| 1197 |
+
"student_loss": 0.04865972697734833,
|
| 1198 |
+
"teacher_loss": 0.029363585636019707
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.14417531718569782,
|
| 1202 |
+
"grad_norm": 2.296875,
|
| 1203 |
+
"kd_loss": 0.014111906290054321,
|
| 1204 |
+
"learning_rate": 7.784019392472539e-06,
|
| 1205 |
+
"loss": 0.18049139976501466,
|
| 1206 |
+
"recon_loss": 0.6205062866210938,
|
| 1207 |
+
"recon_trunc": 0.0,
|
| 1208 |
+
"step": 500,
|
| 1209 |
+
"student_loss": 0.0009538677986711264,
|
| 1210 |
+
"teacher_loss": 0.040066931396722794
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.14561707035755478,
|
| 1214 |
+
"grad_norm": 2.359375,
|
| 1215 |
+
"kd_loss": 0.011222281493246555,
|
| 1216 |
+
"learning_rate": 7.739020481708816e-06,
|
| 1217 |
+
"loss": 0.1553753137588501,
|
| 1218 |
+
"recon_loss": 0.8017247319221497,
|
| 1219 |
+
"recon_trunc": 0.0,
|
| 1220 |
+
"step": 505,
|
| 1221 |
+
"student_loss": 0.001245547435246408,
|
| 1222 |
+
"teacher_loss": 0.0014529024483636022
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 0.14705882352941177,
|
| 1226 |
+
"grad_norm": 2.265625,
|
| 1227 |
+
"kd_loss": 0.011529708281159401,
|
| 1228 |
+
"learning_rate": 7.693702340113587e-06,
|
| 1229 |
+
"loss": 0.1459495782852173,
|
| 1230 |
+
"recon_loss": 0.7228213548660278,
|
| 1231 |
+
"recon_trunc": 0.0,
|
| 1232 |
+
"step": 510,
|
| 1233 |
+
"student_loss": 0.0009306384599767625,
|
| 1234 |
+
"teacher_loss": 0.000919950776733458
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.14850057670126873,
|
| 1238 |
+
"grad_norm": 2.359375,
|
| 1239 |
+
"kd_loss": 0.014361941255629063,
|
| 1240 |
+
"learning_rate": 7.648070249483074e-06,
|
| 1241 |
+
"loss": 0.18740689754486084,
|
| 1242 |
+
"recon_loss": 1.7365156412124634,
|
| 1243 |
+
"recon_trunc": 0.0,
|
| 1244 |
+
"step": 515,
|
| 1245 |
+
"student_loss": 0.0016227519372478127,
|
| 1246 |
+
"teacher_loss": 0.0017847054405137897
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 0.14994232987312572,
|
| 1250 |
+
"grad_norm": 2.015625,
|
| 1251 |
+
"kd_loss": 0.014994297176599503,
|
| 1252 |
+
"learning_rate": 7.602129528204023e-06,
|
| 1253 |
+
"loss": 0.18152790069580077,
|
| 1254 |
+
"recon_loss": 1.1670464277267456,
|
| 1255 |
+
"recon_trunc": 0.0,
|
| 1256 |
+
"step": 520,
|
| 1257 |
+
"student_loss": 0.002644772408530116,
|
| 1258 |
+
"teacher_loss": 0.013961577787995338
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 0.1513840830449827,
|
| 1262 |
+
"grad_norm": 2.125,
|
| 1263 |
+
"kd_loss": 0.020380739122629166,
|
| 1264 |
+
"learning_rate": 7.555885530633864e-06,
|
| 1265 |
+
"loss": 0.16728808879852294,
|
| 1266 |
+
"recon_loss": 0.4622202515602112,
|
| 1267 |
+
"recon_trunc": 0.0,
|
| 1268 |
+
"step": 525,
|
| 1269 |
+
"student_loss": 0.01667250692844391,
|
| 1270 |
+
"teacher_loss": 0.0030309504363685846
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.15282583621683968,
|
| 1274 |
+
"grad_norm": 2.765625,
|
| 1275 |
+
"kd_loss": 0.014126142486929893,
|
| 1276 |
+
"learning_rate": 7.509343646476647e-06,
|
| 1277 |
+
"loss": 0.16154258251190184,
|
| 1278 |
+
"recon_loss": 0.5091199278831482,
|
| 1279 |
+
"recon_trunc": 0.0,
|
| 1280 |
+
"step": 530,
|
| 1281 |
+
"student_loss": 0.0013975589536130428,
|
| 1282 |
+
"teacher_loss": 0.00434094900265336
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 0.15426758938869667,
|
| 1286 |
+
"grad_norm": 2.5625,
|
| 1287 |
+
"kd_loss": 0.015880489721894264,
|
| 1288 |
+
"learning_rate": 7.462509300154892e-06,
|
| 1289 |
+
"loss": 0.15149027109146118,
|
| 1290 |
+
"recon_loss": 0.7252497673034668,
|
| 1291 |
+
"recon_trunc": 0.0,
|
| 1292 |
+
"step": 535,
|
| 1293 |
+
"student_loss": 0.004540898371487856,
|
| 1294 |
+
"teacher_loss": 0.005663472227752209
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 0.15570934256055363,
|
| 1298 |
+
"grad_norm": 2.640625,
|
| 1299 |
+
"kd_loss": 0.01861300878226757,
|
| 1300 |
+
"learning_rate": 7.415387950177369e-06,
|
| 1301 |
+
"loss": 0.15698952674865724,
|
| 1302 |
+
"recon_loss": 0.9284359216690063,
|
| 1303 |
+
"recon_trunc": 0.0,
|
| 1304 |
+
"step": 540,
|
| 1305 |
+
"student_loss": 0.0019844318740069866,
|
| 1306 |
+
"teacher_loss": 0.0008210762753151357
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.15715109573241062,
|
| 1310 |
+
"grad_norm": 2.15625,
|
| 1311 |
+
"kd_loss": 0.027400236576795578,
|
| 1312 |
+
"learning_rate": 7.367985088502914e-06,
|
| 1313 |
+
"loss": 0.14488091468811035,
|
| 1314 |
+
"recon_loss": 1.2230253219604492,
|
| 1315 |
+
"recon_trunc": 0.0,
|
| 1316 |
+
"step": 545,
|
| 1317 |
+
"student_loss": 0.02683446928858757,
|
| 1318 |
+
"teacher_loss": 0.008221668191254139
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.15859284890426759,
|
| 1322 |
+
"grad_norm": 2.328125,
|
| 1323 |
+
"kd_loss": 0.0119510842487216,
|
| 1324 |
+
"learning_rate": 7.320306239900343e-06,
|
| 1325 |
+
"loss": 0.17230284214019775,
|
| 1326 |
+
"recon_loss": 0.9870997667312622,
|
| 1327 |
+
"recon_trunc": 0.0,
|
| 1328 |
+
"step": 550,
|
| 1329 |
+
"student_loss": 0.0011140481801703572,
|
| 1330 |
+
"teacher_loss": 0.000988529296591878
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.16003460207612458,
|
| 1334 |
+
"grad_norm": 1.8046875,
|
| 1335 |
+
"kd_loss": 0.02639094367623329,
|
| 1336 |
+
"learning_rate": 7.27235696130455e-06,
|
| 1337 |
+
"loss": 0.14926526546478272,
|
| 1338 |
+
"recon_loss": 0.7281983494758606,
|
| 1339 |
+
"recon_trunc": 0.0,
|
| 1340 |
+
"step": 555,
|
| 1341 |
+
"student_loss": 0.019614629447460175,
|
| 1342 |
+
"teacher_loss": 0.007916383445262909
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 0.16147635524798154,
|
| 1346 |
+
"grad_norm": 1.6796875,
|
| 1347 |
+
"kd_loss": 0.026238463819026947,
|
| 1348 |
+
"learning_rate": 7.2241428411688415e-06,
|
| 1349 |
+
"loss": 0.1573400616645813,
|
| 1350 |
+
"recon_loss": 0.6570865511894226,
|
| 1351 |
+
"recon_trunc": 0.0,
|
| 1352 |
+
"step": 560,
|
| 1353 |
+
"student_loss": 0.03498203307390213,
|
| 1354 |
+
"teacher_loss": 0.022231703624129295
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.16291810841983853,
|
| 1358 |
+
"grad_norm": 1.8984375,
|
| 1359 |
+
"kd_loss": 0.022579427808523178,
|
| 1360 |
+
"learning_rate": 7.1756694988136165e-06,
|
| 1361 |
+
"loss": 0.16935594081878663,
|
| 1362 |
+
"recon_loss": 0.9929645657539368,
|
| 1363 |
+
"recon_trunc": 0.0,
|
| 1364 |
+
"step": 565,
|
| 1365 |
+
"student_loss": 0.005478309467434883,
|
| 1366 |
+
"teacher_loss": 0.012976857833564281
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.1643598615916955,
|
| 1370 |
+
"grad_norm": 2.5625,
|
| 1371 |
+
"kd_loss": 0.029618442058563232,
|
| 1372 |
+
"learning_rate": 7.126942583771433e-06,
|
| 1373 |
+
"loss": 0.16574451923370362,
|
| 1374 |
+
"recon_loss": 1.726365089416504,
|
| 1375 |
+
"recon_trunc": 0.0,
|
| 1376 |
+
"step": 570,
|
| 1377 |
+
"student_loss": 0.06722798943519592,
|
| 1378 |
+
"teacher_loss": 0.03769288584589958
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 0.16580161476355249,
|
| 1382 |
+
"grad_norm": 7.40625,
|
| 1383 |
+
"kd_loss": 0.010942453518509865,
|
| 1384 |
+
"learning_rate": 7.077967775128558e-06,
|
| 1385 |
+
"loss": 0.15607452392578125,
|
| 1386 |
+
"recon_loss": 1.8715800046920776,
|
| 1387 |
+
"recon_trunc": 0.0,
|
| 1388 |
+
"step": 575,
|
| 1389 |
+
"student_loss": 0.013373637571930885,
|
| 1390 |
+
"teacher_loss": 0.009487644769251347
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 0.16724336793540945,
|
| 1394 |
+
"grad_norm": 2.421875,
|
| 1395 |
+
"kd_loss": 0.019244417548179626,
|
| 1396 |
+
"learning_rate": 7.028750780863078e-06,
|
| 1397 |
+
"loss": 0.16174238920211792,
|
| 1398 |
+
"recon_loss": 0.5514757037162781,
|
| 1399 |
+
"recon_trunc": 0.0,
|
| 1400 |
+
"step": 580,
|
| 1401 |
+
"student_loss": 0.014346003532409668,
|
| 1402 |
+
"teacher_loss": 0.014511034823954105
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.16868512110726644,
|
| 1406 |
+
"grad_norm": 1.96875,
|
| 1407 |
+
"kd_loss": 0.01612640917301178,
|
| 1408 |
+
"learning_rate": 6.979297337179632e-06,
|
| 1409 |
+
"loss": 0.15640640258789062,
|
| 1410 |
+
"recon_loss": 0.6919020414352417,
|
| 1411 |
+
"recon_trunc": 0.0,
|
| 1412 |
+
"step": 585,
|
| 1413 |
+
"student_loss": 0.03018530085682869,
|
| 1414 |
+
"teacher_loss": 0.025109009817242622
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 0.1701268742791234,
|
| 1418 |
+
"grad_norm": 2.453125,
|
| 1419 |
+
"kd_loss": 0.023658068850636482,
|
| 1420 |
+
"learning_rate": 6.929613207840875e-06,
|
| 1421 |
+
"loss": 0.17652963399887084,
|
| 1422 |
+
"recon_loss": 0.7898370623588562,
|
| 1423 |
+
"recon_trunc": 0.0,
|
| 1424 |
+
"step": 590,
|
| 1425 |
+
"student_loss": 0.019267184659838676,
|
| 1426 |
+
"teacher_loss": 0.014148333109915257
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 0.1715686274509804,
|
| 1430 |
+
"grad_norm": 1.7265625,
|
| 1431 |
+
"kd_loss": 0.02613808400928974,
|
| 1432 |
+
"learning_rate": 6.8797041834956955e-06,
|
| 1433 |
+
"loss": 0.15170495510101317,
|
| 1434 |
+
"recon_loss": 0.5493659377098083,
|
| 1435 |
+
"recon_trunc": 0.0,
|
| 1436 |
+
"step": 595,
|
| 1437 |
+
"student_loss": 0.03503395989537239,
|
| 1438 |
+
"teacher_loss": 0.028825486078858376
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.17301038062283736,
|
| 1442 |
+
"grad_norm": 1.7890625,
|
| 1443 |
+
"kd_loss": 0.027726082131266594,
|
| 1444 |
+
"learning_rate": 6.829576081004339e-06,
|
| 1445 |
+
"loss": 0.15570831298828125,
|
| 1446 |
+
"recon_loss": 0.8309513926506042,
|
| 1447 |
+
"recon_trunc": 0.0,
|
| 1448 |
+
"step": 600,
|
| 1449 |
+
"student_loss": 0.029871823266148567,
|
| 1450 |
+
"teacher_loss": 0.020271865651011467
|
| 1451 |
+
}
|
| 1452 |
+
],
|
| 1453 |
+
"logging_steps": 5,
|
| 1454 |
+
"max_steps": 1500,
|
| 1455 |
+
"num_input_tokens_seen": 0,
|
| 1456 |
+
"num_train_epochs": 1,
|
| 1457 |
+
"save_steps": 300,
|
| 1458 |
+
"stateful_callbacks": {
|
| 1459 |
+
"TrainerControl": {
|
| 1460 |
+
"args": {
|
| 1461 |
+
"should_epoch_stop": false,
|
| 1462 |
+
"should_evaluate": false,
|
| 1463 |
+
"should_log": false,
|
| 1464 |
+
"should_save": true,
|
| 1465 |
+
"should_training_stop": false
|
| 1466 |
+
},
|
| 1467 |
+
"attributes": {}
|
| 1468 |
+
}
|
| 1469 |
+
},
|
| 1470 |
+
"total_flos": 0.0,
|
| 1471 |
+
"train_batch_size": 1,
|
| 1472 |
+
"trial_name": null,
|
| 1473 |
+
"trial_params": null
|
| 1474 |
+
}
|