Upload folder using huggingface_hub
Browse files- checkpoint-3000/adapter_config.json +55 -0
- checkpoint-3000/adapter_model.safetensors +3 -0
- checkpoint-3000/chat_template.jinja +15 -0
- checkpoint-3000/optimizer.pt +3 -0
- checkpoint-3000/rng_state.pth +3 -0
- checkpoint-3000/scaler.pt +3 -0
- checkpoint-3000/scheduler.pt +3 -0
- checkpoint-3000/tokenizer.json +0 -0
- checkpoint-3000/tokenizer_config.json +0 -0
- checkpoint-3000/trainer_state.json +511 -0
- checkpoint-3000/training_args.bin +3 -0
checkpoint-3000/adapter_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": {
|
| 6 |
+
"base_model_class": "LlamaForCausalLM",
|
| 7 |
+
"parent_library": "transformers.models.llama.modeling_llama",
|
| 8 |
+
"unsloth_fixed": true
|
| 9 |
+
},
|
| 10 |
+
"base_model_name_or_path": "/tmp/colab_artifacts_v2/resized_model",
|
| 11 |
+
"bias": "none",
|
| 12 |
+
"corda_config": null,
|
| 13 |
+
"ensure_weight_tying": false,
|
| 14 |
+
"eva_config": null,
|
| 15 |
+
"exclude_modules": null,
|
| 16 |
+
"fan_in_fan_out": false,
|
| 17 |
+
"inference_mode": true,
|
| 18 |
+
"init_lora_weights": true,
|
| 19 |
+
"layer_replication": null,
|
| 20 |
+
"layers_pattern": null,
|
| 21 |
+
"layers_to_transform": null,
|
| 22 |
+
"loftq_config": {},
|
| 23 |
+
"lora_alpha": 32,
|
| 24 |
+
"lora_bias": false,
|
| 25 |
+
"lora_dropout": 0,
|
| 26 |
+
"lora_ga_config": null,
|
| 27 |
+
"megatron_config": null,
|
| 28 |
+
"megatron_core": "megatron.core",
|
| 29 |
+
"modules_to_save": [
|
| 30 |
+
"embed_tokens",
|
| 31 |
+
"lm_head"
|
| 32 |
+
],
|
| 33 |
+
"peft_type": "LORA",
|
| 34 |
+
"peft_version": "0.19.1",
|
| 35 |
+
"qalora_group_size": 16,
|
| 36 |
+
"r": 32,
|
| 37 |
+
"rank_pattern": {},
|
| 38 |
+
"revision": null,
|
| 39 |
+
"target_modules": [
|
| 40 |
+
"q_proj",
|
| 41 |
+
"o_proj",
|
| 42 |
+
"k_proj",
|
| 43 |
+
"down_proj",
|
| 44 |
+
"gate_proj",
|
| 45 |
+
"up_proj",
|
| 46 |
+
"v_proj"
|
| 47 |
+
],
|
| 48 |
+
"target_parameters": null,
|
| 49 |
+
"task_type": "CAUSAL_LM",
|
| 50 |
+
"trainable_token_indices": null,
|
| 51 |
+
"use_bdlora": null,
|
| 52 |
+
"use_dora": false,
|
| 53 |
+
"use_qalora": false,
|
| 54 |
+
"use_rslora": false
|
| 55 |
+
}
|
checkpoint-3000/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bc2a4ac72a3e0684ec977c9bfea1b7b017a64963da51d1643831ea3bbc949724
|
| 3 |
+
size 801203008
|
checkpoint-3000/chat_template.jinja
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}
|
| 2 |
+
{% if message['role'] == 'user' %}
|
| 3 |
+
{{ '<|user|>
|
| 4 |
+
' + message['content'] + eos_token }}
|
| 5 |
+
{% elif message['role'] == 'system' %}
|
| 6 |
+
{{ '<|system|>
|
| 7 |
+
' + message['content'] + eos_token }}
|
| 8 |
+
{% elif message['role'] == 'assistant' %}
|
| 9 |
+
{{ '<|assistant|>
|
| 10 |
+
' + message['content'] + eos_token }}
|
| 11 |
+
{% endif %}
|
| 12 |
+
{% if loop.last and add_generation_prompt %}
|
| 13 |
+
{{ '<|assistant|>' }}
|
| 14 |
+
{% endif %}
|
| 15 |
+
{% endfor %}
|
checkpoint-3000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b9fb38d5342dd24ab71bd0ffbb391b763693d63864f04f4644df6ea929871a4f
|
| 3 |
+
size 929648955
|
checkpoint-3000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2922886dc37f9211a4affea61093cbe4db90c0a37e91e7edac63ec0cf7b13d45
|
| 3 |
+
size 14645
|
checkpoint-3000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:abcab70bf125d1c16a3ffda2530ff0e0204d37053cf183ea9ed8b6ac39f1bbe1
|
| 3 |
+
size 1383
|
checkpoint-3000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bcde72b96528dbcd8895233e61a15bc38dab1dc645d75d41cc79340c74f6415d
|
| 3 |
+
size 1465
|
checkpoint-3000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3000/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3000/trainer_state.json
ADDED
|
@@ -0,0 +1,511 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 3000,
|
| 3 |
+
"best_metric": 3.2174906730651855,
|
| 4 |
+
"best_model_checkpoint": "/tmp/colab_artifacts_v2/pretrain_checkpoints/checkpoint-3000",
|
| 5 |
+
"epoch": 0.630119722747322,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.010501995379122032,
|
| 14 |
+
"grad_norm": 16.108776092529297,
|
| 15 |
+
"learning_rate": 4.716981132075472e-06,
|
| 16 |
+
"loss": 7.048248901367187,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.021003990758244065,
|
| 21 |
+
"grad_norm": 7.494509220123291,
|
| 22 |
+
"learning_rate": 9.958071278825997e-06,
|
| 23 |
+
"loss": 5.635718994140625,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0315059861373661,
|
| 28 |
+
"grad_norm": 3.1452040672302246,
|
| 29 |
+
"learning_rate": 1.5199161425576522e-05,
|
| 30 |
+
"loss": 4.769013366699219,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.04200798151648813,
|
| 35 |
+
"grad_norm": 6.104097843170166,
|
| 36 |
+
"learning_rate": 2.0440251572327044e-05,
|
| 37 |
+
"loss": 4.167159423828125,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05250997689561017,
|
| 42 |
+
"grad_norm": 2.9789626598358154,
|
| 43 |
+
"learning_rate": 2.5681341719077568e-05,
|
| 44 |
+
"loss": 4.021661987304688,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0630119722747322,
|
| 49 |
+
"grad_norm": 3.7343900203704834,
|
| 50 |
+
"learning_rate": 3.092243186582809e-05,
|
| 51 |
+
"loss": 3.8835012817382815,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.07351396765385423,
|
| 56 |
+
"grad_norm": 3.598313808441162,
|
| 57 |
+
"learning_rate": 3.6163522012578616e-05,
|
| 58 |
+
"loss": 3.719826354980469,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.08401596303297626,
|
| 63 |
+
"grad_norm": 2.9602694511413574,
|
| 64 |
+
"learning_rate": 4.140461215932914e-05,
|
| 65 |
+
"loss": 3.6351690673828125,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0945179584120983,
|
| 70 |
+
"grad_norm": 2.4005064964294434,
|
| 71 |
+
"learning_rate": 4.664570230607967e-05,
|
| 72 |
+
"loss": 3.552754211425781,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.10501995379122034,
|
| 77 |
+
"grad_norm": 3.6799378395080566,
|
| 78 |
+
"learning_rate": 4.978991596638656e-05,
|
| 79 |
+
"loss": 3.5413519287109376,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.10501995379122034,
|
| 84 |
+
"eval_loss": 4.811308860778809,
|
| 85 |
+
"eval_runtime": 24.7312,
|
| 86 |
+
"eval_samples_per_second": 10.958,
|
| 87 |
+
"eval_steps_per_second": 2.75,
|
| 88 |
+
"step": 500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.11552194917034236,
|
| 92 |
+
"grad_norm": 2.4428603649139404,
|
| 93 |
+
"learning_rate": 4.9206349206349204e-05,
|
| 94 |
+
"loss": 3.434654541015625,
|
| 95 |
+
"step": 550
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.1260239445494644,
|
| 99 |
+
"grad_norm": 2.2486448287963867,
|
| 100 |
+
"learning_rate": 4.862278244631186e-05,
|
| 101 |
+
"loss": 3.5255862426757814,
|
| 102 |
+
"step": 600
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.13652593992858644,
|
| 106 |
+
"grad_norm": 2.5440492630004883,
|
| 107 |
+
"learning_rate": 4.803921568627452e-05,
|
| 108 |
+
"loss": 3.265621337890625,
|
| 109 |
+
"step": 650
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.14702793530770847,
|
| 113 |
+
"grad_norm": 5.0713605880737305,
|
| 114 |
+
"learning_rate": 4.745564892623716e-05,
|
| 115 |
+
"loss": 3.3151104736328123,
|
| 116 |
+
"step": 700
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.1575299306868305,
|
| 120 |
+
"grad_norm": 3.032827615737915,
|
| 121 |
+
"learning_rate": 4.6872082166199816e-05,
|
| 122 |
+
"loss": 3.3802804565429687,
|
| 123 |
+
"step": 750
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.16803192606595252,
|
| 127 |
+
"grad_norm": 2.8556971549987793,
|
| 128 |
+
"learning_rate": 4.6288515406162466e-05,
|
| 129 |
+
"loss": 3.175123596191406,
|
| 130 |
+
"step": 800
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.17853392144507457,
|
| 134 |
+
"grad_norm": 2.813127279281616,
|
| 135 |
+
"learning_rate": 4.5704948646125116e-05,
|
| 136 |
+
"loss": 3.1402743530273436,
|
| 137 |
+
"step": 850
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.1890359168241966,
|
| 141 |
+
"grad_norm": 3.5564844608306885,
|
| 142 |
+
"learning_rate": 4.512138188608777e-05,
|
| 143 |
+
"loss": 3.1378076171875,
|
| 144 |
+
"step": 900
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.19953791220331862,
|
| 148 |
+
"grad_norm": 4.420061111450195,
|
| 149 |
+
"learning_rate": 4.453781512605042e-05,
|
| 150 |
+
"loss": 3.09103271484375,
|
| 151 |
+
"step": 950
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.21003990758244068,
|
| 155 |
+
"grad_norm": 3.6192879676818848,
|
| 156 |
+
"learning_rate": 4.395424836601307e-05,
|
| 157 |
+
"loss": 3.145036315917969,
|
| 158 |
+
"step": 1000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.21003990758244068,
|
| 162 |
+
"eval_loss": 3.8735694885253906,
|
| 163 |
+
"eval_runtime": 24.3553,
|
| 164 |
+
"eval_samples_per_second": 11.127,
|
| 165 |
+
"eval_steps_per_second": 2.792,
|
| 166 |
+
"step": 1000
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.2205419029615627,
|
| 170 |
+
"grad_norm": 6.488380432128906,
|
| 171 |
+
"learning_rate": 4.337068160597573e-05,
|
| 172 |
+
"loss": 3.167205810546875,
|
| 173 |
+
"step": 1050
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.23104389834068473,
|
| 177 |
+
"grad_norm": 3.438765525817871,
|
| 178 |
+
"learning_rate": 4.278711484593838e-05,
|
| 179 |
+
"loss": 3.0527557373046874,
|
| 180 |
+
"step": 1100
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.24154589371980675,
|
| 184 |
+
"grad_norm": 3.3744046688079834,
|
| 185 |
+
"learning_rate": 4.220354808590103e-05,
|
| 186 |
+
"loss": 3.056681823730469,
|
| 187 |
+
"step": 1150
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.2520478890989288,
|
| 191 |
+
"grad_norm": 2.0418734550476074,
|
| 192 |
+
"learning_rate": 4.1619981325863684e-05,
|
| 193 |
+
"loss": 3.0335903930664063,
|
| 194 |
+
"step": 1200
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.26254988447805083,
|
| 198 |
+
"grad_norm": 3.0685362815856934,
|
| 199 |
+
"learning_rate": 4.1036414565826334e-05,
|
| 200 |
+
"loss": 2.931863098144531,
|
| 201 |
+
"step": 1250
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.2730518798571729,
|
| 205 |
+
"grad_norm": 2.3799123764038086,
|
| 206 |
+
"learning_rate": 4.0452847805788984e-05,
|
| 207 |
+
"loss": 3.0471902465820313,
|
| 208 |
+
"step": 1300
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.2835538752362949,
|
| 212 |
+
"grad_norm": 2.9344570636749268,
|
| 213 |
+
"learning_rate": 3.986928104575164e-05,
|
| 214 |
+
"loss": 2.888125,
|
| 215 |
+
"step": 1350
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.29405587061541694,
|
| 219 |
+
"grad_norm": 2.8988893032073975,
|
| 220 |
+
"learning_rate": 3.928571428571429e-05,
|
| 221 |
+
"loss": 2.940912170410156,
|
| 222 |
+
"step": 1400
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.304557865994539,
|
| 226 |
+
"grad_norm": 8.145448684692383,
|
| 227 |
+
"learning_rate": 3.870214752567694e-05,
|
| 228 |
+
"loss": 2.9622201538085937,
|
| 229 |
+
"step": 1450
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.315059861373661,
|
| 233 |
+
"grad_norm": 3.801460027694702,
|
| 234 |
+
"learning_rate": 3.811858076563959e-05,
|
| 235 |
+
"loss": 2.9859747314453124,
|
| 236 |
+
"step": 1500
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.315059861373661,
|
| 240 |
+
"eval_loss": 3.5815439224243164,
|
| 241 |
+
"eval_runtime": 24.3722,
|
| 242 |
+
"eval_samples_per_second": 11.119,
|
| 243 |
+
"eval_steps_per_second": 2.79,
|
| 244 |
+
"step": 1500
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.32556185675278304,
|
| 248 |
+
"grad_norm": 2.2608940601348877,
|
| 249 |
+
"learning_rate": 3.753501400560224e-05,
|
| 250 |
+
"loss": 2.870184326171875,
|
| 251 |
+
"step": 1550
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.33606385213190504,
|
| 255 |
+
"grad_norm": 3.5865588188171387,
|
| 256 |
+
"learning_rate": 3.6951447245564896e-05,
|
| 257 |
+
"loss": 2.9996963500976563,
|
| 258 |
+
"step": 1600
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.3465658475110271,
|
| 262 |
+
"grad_norm": 3.1756653785705566,
|
| 263 |
+
"learning_rate": 3.6367880485527545e-05,
|
| 264 |
+
"loss": 2.9004736328125,
|
| 265 |
+
"step": 1650
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.35706784289014915,
|
| 269 |
+
"grad_norm": 3.698146104812622,
|
| 270 |
+
"learning_rate": 3.5784313725490195e-05,
|
| 271 |
+
"loss": 2.9160443115234376,
|
| 272 |
+
"step": 1700
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.36756983826927114,
|
| 276 |
+
"grad_norm": 2.639841079711914,
|
| 277 |
+
"learning_rate": 3.520074696545285e-05,
|
| 278 |
+
"loss": 2.928147888183594,
|
| 279 |
+
"step": 1750
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.3780718336483932,
|
| 283 |
+
"grad_norm": 3.2465906143188477,
|
| 284 |
+
"learning_rate": 3.46171802054155e-05,
|
| 285 |
+
"loss": 2.837327575683594,
|
| 286 |
+
"step": 1800
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.38857382902751525,
|
| 290 |
+
"grad_norm": 3.475036859512329,
|
| 291 |
+
"learning_rate": 3.403361344537815e-05,
|
| 292 |
+
"loss": 2.87112548828125,
|
| 293 |
+
"step": 1850
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.39907582440663725,
|
| 297 |
+
"grad_norm": 2.9756319522857666,
|
| 298 |
+
"learning_rate": 3.34500466853408e-05,
|
| 299 |
+
"loss": 2.870127868652344,
|
| 300 |
+
"step": 1900
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.4095778197857593,
|
| 304 |
+
"grad_norm": 2.9511425495147705,
|
| 305 |
+
"learning_rate": 3.286647992530346e-05,
|
| 306 |
+
"loss": 2.887370910644531,
|
| 307 |
+
"step": 1950
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 0.42007981516488135,
|
| 311 |
+
"grad_norm": 2.802246332168579,
|
| 312 |
+
"learning_rate": 3.228291316526611e-05,
|
| 313 |
+
"loss": 2.8228793334960938,
|
| 314 |
+
"step": 2000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.42007981516488135,
|
| 318 |
+
"eval_loss": 3.409285306930542,
|
| 319 |
+
"eval_runtime": 24.4492,
|
| 320 |
+
"eval_samples_per_second": 11.084,
|
| 321 |
+
"eval_steps_per_second": 2.781,
|
| 322 |
+
"step": 2000
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.43058181054400335,
|
| 326 |
+
"grad_norm": 3.337001085281372,
|
| 327 |
+
"learning_rate": 3.169934640522876e-05,
|
| 328 |
+
"loss": 2.8744744873046875,
|
| 329 |
+
"step": 2050
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.4410838059231254,
|
| 333 |
+
"grad_norm": 2.8167831897735596,
|
| 334 |
+
"learning_rate": 3.1115779645191413e-05,
|
| 335 |
+
"loss": 2.8364910888671875,
|
| 336 |
+
"step": 2100
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.4515858013022474,
|
| 340 |
+
"grad_norm": 3.531158685684204,
|
| 341 |
+
"learning_rate": 3.053221288515406e-05,
|
| 342 |
+
"loss": 2.7930868530273436,
|
| 343 |
+
"step": 2150
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 0.46208779668136946,
|
| 347 |
+
"grad_norm": 2.5563290119171143,
|
| 348 |
+
"learning_rate": 2.9948646125116713e-05,
|
| 349 |
+
"loss": 2.8546572875976564,
|
| 350 |
+
"step": 2200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.4725897920604915,
|
| 354 |
+
"grad_norm": 3.6629157066345215,
|
| 355 |
+
"learning_rate": 2.9365079365079366e-05,
|
| 356 |
+
"loss": 2.7838204956054686,
|
| 357 |
+
"step": 2250
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.4830917874396135,
|
| 361 |
+
"grad_norm": 3.887286424636841,
|
| 362 |
+
"learning_rate": 2.8781512605042016e-05,
|
| 363 |
+
"loss": 2.851587829589844,
|
| 364 |
+
"step": 2300
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 0.49359378281873556,
|
| 368 |
+
"grad_norm": 4.215712547302246,
|
| 369 |
+
"learning_rate": 2.819794584500467e-05,
|
| 370 |
+
"loss": 2.83611328125,
|
| 371 |
+
"step": 2350
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.5040957781978576,
|
| 375 |
+
"grad_norm": 3.636672019958496,
|
| 376 |
+
"learning_rate": 2.7614379084967322e-05,
|
| 377 |
+
"loss": 2.8077880859375,
|
| 378 |
+
"step": 2400
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.5145977735769797,
|
| 382 |
+
"grad_norm": 3.7146553993225098,
|
| 383 |
+
"learning_rate": 2.7030812324929972e-05,
|
| 384 |
+
"loss": 2.8354559326171875,
|
| 385 |
+
"step": 2450
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.5250997689561017,
|
| 389 |
+
"grad_norm": 2.1127307415008545,
|
| 390 |
+
"learning_rate": 2.6447245564892625e-05,
|
| 391 |
+
"loss": 2.7687191772460937,
|
| 392 |
+
"step": 2500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.5250997689561017,
|
| 396 |
+
"eval_loss": 3.3053934574127197,
|
| 397 |
+
"eval_runtime": 24.4584,
|
| 398 |
+
"eval_samples_per_second": 11.08,
|
| 399 |
+
"eval_steps_per_second": 2.78,
|
| 400 |
+
"step": 2500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.5356017643352237,
|
| 404 |
+
"grad_norm": 3.9465138912200928,
|
| 405 |
+
"learning_rate": 2.5863678804855278e-05,
|
| 406 |
+
"loss": 2.762879638671875,
|
| 407 |
+
"step": 2550
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.5461037597143458,
|
| 411 |
+
"grad_norm": 2.7259905338287354,
|
| 412 |
+
"learning_rate": 2.5280112044817928e-05,
|
| 413 |
+
"loss": 2.7311965942382814,
|
| 414 |
+
"step": 2600
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.5566057550934678,
|
| 418 |
+
"grad_norm": 2.7369556427001953,
|
| 419 |
+
"learning_rate": 2.469654528478058e-05,
|
| 420 |
+
"loss": 2.7701571655273436,
|
| 421 |
+
"step": 2650
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.5671077504725898,
|
| 425 |
+
"grad_norm": 3.810979127883911,
|
| 426 |
+
"learning_rate": 2.411297852474323e-05,
|
| 427 |
+
"loss": 2.7268853759765626,
|
| 428 |
+
"step": 2700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.5776097458517119,
|
| 432 |
+
"grad_norm": 2.9589717388153076,
|
| 433 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 434 |
+
"loss": 2.7360519409179687,
|
| 435 |
+
"step": 2750
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.5881117412308339,
|
| 439 |
+
"grad_norm": 3.193847894668579,
|
| 440 |
+
"learning_rate": 2.2945845004668533e-05,
|
| 441 |
+
"loss": 2.722780456542969,
|
| 442 |
+
"step": 2800
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.5986137366099559,
|
| 446 |
+
"grad_norm": 3.0324525833129883,
|
| 447 |
+
"learning_rate": 2.2362278244631187e-05,
|
| 448 |
+
"loss": 2.817607421875,
|
| 449 |
+
"step": 2850
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.609115731989078,
|
| 453 |
+
"grad_norm": 3.465670108795166,
|
| 454 |
+
"learning_rate": 2.177871148459384e-05,
|
| 455 |
+
"loss": 2.725840759277344,
|
| 456 |
+
"step": 2900
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.6196177273682,
|
| 460 |
+
"grad_norm": 3.059812307357788,
|
| 461 |
+
"learning_rate": 2.119514472455649e-05,
|
| 462 |
+
"loss": 2.73619140625,
|
| 463 |
+
"step": 2950
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 0.630119722747322,
|
| 467 |
+
"grad_norm": 3.341780662536621,
|
| 468 |
+
"learning_rate": 2.0611577964519143e-05,
|
| 469 |
+
"loss": 2.7529730224609374,
|
| 470 |
+
"step": 3000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.630119722747322,
|
| 474 |
+
"eval_loss": 3.2174906730651855,
|
| 475 |
+
"eval_runtime": 24.4218,
|
| 476 |
+
"eval_samples_per_second": 11.097,
|
| 477 |
+
"eval_steps_per_second": 2.784,
|
| 478 |
+
"step": 3000
|
| 479 |
+
}
|
| 480 |
+
],
|
| 481 |
+
"logging_steps": 50,
|
| 482 |
+
"max_steps": 4761,
|
| 483 |
+
"num_input_tokens_seen": 0,
|
| 484 |
+
"num_train_epochs": 1,
|
| 485 |
+
"save_steps": 500,
|
| 486 |
+
"stateful_callbacks": {
|
| 487 |
+
"EarlyStoppingCallback": {
|
| 488 |
+
"args": {
|
| 489 |
+
"early_stopping_patience": 3,
|
| 490 |
+
"early_stopping_threshold": 0.01
|
| 491 |
+
},
|
| 492 |
+
"attributes": {
|
| 493 |
+
"early_stopping_patience_counter": 0
|
| 494 |
+
}
|
| 495 |
+
},
|
| 496 |
+
"TrainerControl": {
|
| 497 |
+
"args": {
|
| 498 |
+
"should_epoch_stop": false,
|
| 499 |
+
"should_evaluate": false,
|
| 500 |
+
"should_log": false,
|
| 501 |
+
"should_save": true,
|
| 502 |
+
"should_training_stop": false
|
| 503 |
+
},
|
| 504 |
+
"attributes": {}
|
| 505 |
+
}
|
| 506 |
+
},
|
| 507 |
+
"total_flos": 4.283008443823718e+16,
|
| 508 |
+
"train_batch_size": 4,
|
| 509 |
+
"trial_name": null,
|
| 510 |
+
"trial_params": null
|
| 511 |
+
}
|
checkpoint-3000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4f1f7f3546d1cc3758bfa2cf3e663df947f7820b9154cedeaa390df7f9639303
|
| 3 |
+
size 5713
|