Training in progress, epoch 1, checkpoint
Browse files- checkpoint-112/config.json +59 -0
- checkpoint-112/generation_config.json +6 -0
- checkpoint-112/model.safetensors +3 -0
- checkpoint-112/optimizer.pt +3 -0
- checkpoint-112/rng_state_0.pth +3 -0
- checkpoint-112/rng_state_1.pth +3 -0
- checkpoint-112/rng_state_2.pth +3 -0
- checkpoint-112/rng_state_3.pth +3 -0
- checkpoint-112/scheduler.pt +3 -0
- checkpoint-112/trainer_state.json +818 -0
- checkpoint-112/training_args.bin +3 -0
checkpoint-112/config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen2ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"bos_token_id": 151643,
|
| 7 |
+
"eos_token_id": 151643,
|
| 8 |
+
"hidden_act": "silu",
|
| 9 |
+
"hidden_size": 1536,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 8960,
|
| 12 |
+
"layer_types": [
|
| 13 |
+
"full_attention",
|
| 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 |
+
],
|
| 42 |
+
"max_position_embeddings": 131072,
|
| 43 |
+
"max_window_layers": 28,
|
| 44 |
+
"model_type": "qwen2",
|
| 45 |
+
"num_attention_heads": 12,
|
| 46 |
+
"num_hidden_layers": 28,
|
| 47 |
+
"num_key_value_heads": 2,
|
| 48 |
+
"rms_norm_eps": 1e-06,
|
| 49 |
+
"rope_scaling": null,
|
| 50 |
+
"rope_theta": 1000000.0,
|
| 51 |
+
"sliding_window": null,
|
| 52 |
+
"tie_word_embeddings": true,
|
| 53 |
+
"torch_dtype": "bfloat16",
|
| 54 |
+
"transformers_version": "4.55.2",
|
| 55 |
+
"use_cache": true,
|
| 56 |
+
"use_mrope": false,
|
| 57 |
+
"use_sliding_window": false,
|
| 58 |
+
"vocab_size": 151667
|
| 59 |
+
}
|
checkpoint-112/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": 151643,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"transformers_version": "4.55.2"
|
| 6 |
+
}
|
checkpoint-112/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:953bd36ef62ac6470c2ab74e62c95777852f8602622dc705949309f2216873c9
|
| 3 |
+
size 3086640776
|
checkpoint-112/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3c52123b7dfcff19fbf07e406d49e5adf041c2dc0b1c487214805e220258054
|
| 3 |
+
size 6055930
|
checkpoint-112/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8dadb1d7a35d2e493d316615f04f3e23834b6b99d6a0c3c8e577e518ec7c7e5
|
| 3 |
+
size 14960
|
checkpoint-112/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4c154b6a63e0b1f98f7d2847944398f99f1657d35e8eddf7fdf0ae2c24b0552
|
| 3 |
+
size 15024
|
checkpoint-112/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f784c6a9507b51189f2caffbd178ea9882103b75852e31c15f47fdae6a43af1d
|
| 3 |
+
size 15024
|
checkpoint-112/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34b023e05bc2d12b91dc436d4922b990d50ec8dc56d40dc3e36b3bb34fc81341
|
| 3 |
+
size 15024
|
checkpoint-112/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d20553bc1148553b273e373b131a0c82ef7d73af94469106fa6681165deca328
|
| 3 |
+
size 1064
|
checkpoint-112/trainer_state.json
ADDED
|
@@ -0,0 +1,818 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 112,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008988764044943821,
|
| 14 |
+
"grad_norm": 105.5,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 5.1855,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.017977528089887642,
|
| 21 |
+
"grad_norm": 101.5,
|
| 22 |
+
"learning_rate": 1.4705882352941177e-06,
|
| 23 |
+
"loss": 5.0139,
|
| 24 |
+
"step": 2
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02696629213483146,
|
| 28 |
+
"grad_norm": 140.0,
|
| 29 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 30 |
+
"loss": 6.6301,
|
| 31 |
+
"step": 3
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.035955056179775284,
|
| 35 |
+
"grad_norm": 102.5,
|
| 36 |
+
"learning_rate": 4.411764705882353e-06,
|
| 37 |
+
"loss": 5.0946,
|
| 38 |
+
"step": 4
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0449438202247191,
|
| 42 |
+
"grad_norm": 113.0,
|
| 43 |
+
"learning_rate": 5.882352941176471e-06,
|
| 44 |
+
"loss": 5.5796,
|
| 45 |
+
"step": 5
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05393258426966292,
|
| 49 |
+
"grad_norm": 129.0,
|
| 50 |
+
"learning_rate": 7.3529411764705884e-06,
|
| 51 |
+
"loss": 6.2941,
|
| 52 |
+
"step": 6
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06292134831460675,
|
| 56 |
+
"grad_norm": 104.5,
|
| 57 |
+
"learning_rate": 8.823529411764707e-06,
|
| 58 |
+
"loss": 5.088,
|
| 59 |
+
"step": 7
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07191011235955057,
|
| 63 |
+
"grad_norm": 120.0,
|
| 64 |
+
"learning_rate": 1.0294117647058824e-05,
|
| 65 |
+
"loss": 5.5442,
|
| 66 |
+
"step": 8
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.08089887640449438,
|
| 70 |
+
"grad_norm": 140.0,
|
| 71 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 72 |
+
"loss": 4.9847,
|
| 73 |
+
"step": 9
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0898876404494382,
|
| 77 |
+
"grad_norm": 142.0,
|
| 78 |
+
"learning_rate": 1.323529411764706e-05,
|
| 79 |
+
"loss": 4.8134,
|
| 80 |
+
"step": 10
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09887640449438202,
|
| 84 |
+
"grad_norm": 213.0,
|
| 85 |
+
"learning_rate": 1.4705882352941177e-05,
|
| 86 |
+
"loss": 4.1162,
|
| 87 |
+
"step": 11
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10786516853932585,
|
| 91 |
+
"grad_norm": 229.0,
|
| 92 |
+
"learning_rate": 1.6176470588235296e-05,
|
| 93 |
+
"loss": 4.9429,
|
| 94 |
+
"step": 12
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.11685393258426967,
|
| 98 |
+
"grad_norm": 225.0,
|
| 99 |
+
"learning_rate": 1.7647058823529414e-05,
|
| 100 |
+
"loss": 3.9659,
|
| 101 |
+
"step": 13
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.1258426966292135,
|
| 105 |
+
"grad_norm": 1280.0,
|
| 106 |
+
"learning_rate": 1.9117647058823528e-05,
|
| 107 |
+
"loss": 5.1684,
|
| 108 |
+
"step": 14
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.1348314606741573,
|
| 112 |
+
"grad_norm": 504.0,
|
| 113 |
+
"learning_rate": 2.058823529411765e-05,
|
| 114 |
+
"loss": 4.3175,
|
| 115 |
+
"step": 15
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.14382022471910114,
|
| 119 |
+
"grad_norm": 235.0,
|
| 120 |
+
"learning_rate": 2.2058823529411766e-05,
|
| 121 |
+
"loss": 3.3212,
|
| 122 |
+
"step": 16
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.15280898876404495,
|
| 126 |
+
"grad_norm": 179.0,
|
| 127 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 128 |
+
"loss": 3.4494,
|
| 129 |
+
"step": 17
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.16179775280898875,
|
| 133 |
+
"grad_norm": 102.5,
|
| 134 |
+
"learning_rate": 2.5e-05,
|
| 135 |
+
"loss": 3.5728,
|
| 136 |
+
"step": 18
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1707865168539326,
|
| 140 |
+
"grad_norm": 45.25,
|
| 141 |
+
"learning_rate": 2.647058823529412e-05,
|
| 142 |
+
"loss": 3.2919,
|
| 143 |
+
"step": 19
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.1797752808988764,
|
| 147 |
+
"grad_norm": 157.0,
|
| 148 |
+
"learning_rate": 2.7941176470588236e-05,
|
| 149 |
+
"loss": 3.2853,
|
| 150 |
+
"step": 20
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.18876404494382024,
|
| 154 |
+
"grad_norm": 110.5,
|
| 155 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 156 |
+
"loss": 4.1756,
|
| 157 |
+
"step": 21
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.19775280898876405,
|
| 161 |
+
"grad_norm": 163.0,
|
| 162 |
+
"learning_rate": 3.0882352941176475e-05,
|
| 163 |
+
"loss": 3.1835,
|
| 164 |
+
"step": 22
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.20674157303370785,
|
| 168 |
+
"grad_norm": 219.0,
|
| 169 |
+
"learning_rate": 3.235294117647059e-05,
|
| 170 |
+
"loss": 3.9426,
|
| 171 |
+
"step": 23
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2157303370786517,
|
| 175 |
+
"grad_norm": 44.75,
|
| 176 |
+
"learning_rate": 3.382352941176471e-05,
|
| 177 |
+
"loss": 2.5031,
|
| 178 |
+
"step": 24
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.2247191011235955,
|
| 182 |
+
"grad_norm": 58.25,
|
| 183 |
+
"learning_rate": 3.529411764705883e-05,
|
| 184 |
+
"loss": 2.7572,
|
| 185 |
+
"step": 25
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.23370786516853934,
|
| 189 |
+
"grad_norm": 37.5,
|
| 190 |
+
"learning_rate": 3.6764705882352945e-05,
|
| 191 |
+
"loss": 3.012,
|
| 192 |
+
"step": 26
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.24269662921348314,
|
| 196 |
+
"grad_norm": 47.5,
|
| 197 |
+
"learning_rate": 3.8235294117647055e-05,
|
| 198 |
+
"loss": 2.9611,
|
| 199 |
+
"step": 27
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.251685393258427,
|
| 203 |
+
"grad_norm": 50.25,
|
| 204 |
+
"learning_rate": 3.970588235294117e-05,
|
| 205 |
+
"loss": 2.7039,
|
| 206 |
+
"step": 28
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2606741573033708,
|
| 210 |
+
"grad_norm": 41.25,
|
| 211 |
+
"learning_rate": 4.11764705882353e-05,
|
| 212 |
+
"loss": 2.0191,
|
| 213 |
+
"step": 29
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.2696629213483146,
|
| 217 |
+
"grad_norm": 47.5,
|
| 218 |
+
"learning_rate": 4.2647058823529415e-05,
|
| 219 |
+
"loss": 2.1197,
|
| 220 |
+
"step": 30
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2786516853932584,
|
| 224 |
+
"grad_norm": 44.0,
|
| 225 |
+
"learning_rate": 4.411764705882353e-05,
|
| 226 |
+
"loss": 1.8918,
|
| 227 |
+
"step": 31
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2876404494382023,
|
| 231 |
+
"grad_norm": 44.75,
|
| 232 |
+
"learning_rate": 4.558823529411765e-05,
|
| 233 |
+
"loss": 2.3818,
|
| 234 |
+
"step": 32
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2966292134831461,
|
| 238 |
+
"grad_norm": 46.75,
|
| 239 |
+
"learning_rate": 4.705882352941177e-05,
|
| 240 |
+
"loss": 1.7817,
|
| 241 |
+
"step": 33
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3056179775280899,
|
| 245 |
+
"grad_norm": 49.0,
|
| 246 |
+
"learning_rate": 4.8529411764705885e-05,
|
| 247 |
+
"loss": 1.5977,
|
| 248 |
+
"step": 34
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3146067415730337,
|
| 252 |
+
"grad_norm": 49.25,
|
| 253 |
+
"learning_rate": 5e-05,
|
| 254 |
+
"loss": 1.4331,
|
| 255 |
+
"step": 35
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3235955056179775,
|
| 259 |
+
"grad_norm": 42.0,
|
| 260 |
+
"learning_rate": 4.999864732969518e-05,
|
| 261 |
+
"loss": 1.1944,
|
| 262 |
+
"step": 36
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3325842696629214,
|
| 266 |
+
"grad_norm": 33.5,
|
| 267 |
+
"learning_rate": 4.999458946515808e-05,
|
| 268 |
+
"loss": 0.9532,
|
| 269 |
+
"step": 37
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.3415730337078652,
|
| 273 |
+
"grad_norm": 40.5,
|
| 274 |
+
"learning_rate": 4.9987826845504914e-05,
|
| 275 |
+
"loss": 0.9695,
|
| 276 |
+
"step": 38
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.350561797752809,
|
| 280 |
+
"grad_norm": 36.5,
|
| 281 |
+
"learning_rate": 4.997836020254328e-05,
|
| 282 |
+
"loss": 0.948,
|
| 283 |
+
"step": 39
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3595505617977528,
|
| 287 |
+
"grad_norm": 76.0,
|
| 288 |
+
"learning_rate": 4.996619056069291e-05,
|
| 289 |
+
"loss": 1.0868,
|
| 290 |
+
"step": 40
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3685393258426966,
|
| 294 |
+
"grad_norm": 118.5,
|
| 295 |
+
"learning_rate": 4.995131923687488e-05,
|
| 296 |
+
"loss": 3.4243,
|
| 297 |
+
"step": 41
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.3775280898876405,
|
| 301 |
+
"grad_norm": 55.25,
|
| 302 |
+
"learning_rate": 4.9933747840369015e-05,
|
| 303 |
+
"loss": 1.7167,
|
| 304 |
+
"step": 42
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3865168539325843,
|
| 308 |
+
"grad_norm": 48.5,
|
| 309 |
+
"learning_rate": 4.991347827263982e-05,
|
| 310 |
+
"loss": 1.4509,
|
| 311 |
+
"step": 43
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3955056179775281,
|
| 315 |
+
"grad_norm": 49.5,
|
| 316 |
+
"learning_rate": 4.98905127271307e-05,
|
| 317 |
+
"loss": 1.1233,
|
| 318 |
+
"step": 44
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.4044943820224719,
|
| 322 |
+
"grad_norm": 42.0,
|
| 323 |
+
"learning_rate": 4.9864853689026556e-05,
|
| 324 |
+
"loss": 0.9881,
|
| 325 |
+
"step": 45
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.4134831460674157,
|
| 329 |
+
"grad_norm": 32.0,
|
| 330 |
+
"learning_rate": 4.98365039349849e-05,
|
| 331 |
+
"loss": 0.8365,
|
| 332 |
+
"step": 46
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.42247191011235957,
|
| 336 |
+
"grad_norm": 24.75,
|
| 337 |
+
"learning_rate": 4.9805466532835376e-05,
|
| 338 |
+
"loss": 0.675,
|
| 339 |
+
"step": 47
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.4314606741573034,
|
| 343 |
+
"grad_norm": 15.125,
|
| 344 |
+
"learning_rate": 4.9771744841247756e-05,
|
| 345 |
+
"loss": 0.8089,
|
| 346 |
+
"step": 48
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.4404494382022472,
|
| 350 |
+
"grad_norm": 11.375,
|
| 351 |
+
"learning_rate": 4.973534250936851e-05,
|
| 352 |
+
"loss": 0.7185,
|
| 353 |
+
"step": 49
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.449438202247191,
|
| 357 |
+
"grad_norm": 10.25,
|
| 358 |
+
"learning_rate": 4.9696263476425905e-05,
|
| 359 |
+
"loss": 0.6298,
|
| 360 |
+
"step": 50
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.4584269662921348,
|
| 364 |
+
"grad_norm": 6.25,
|
| 365 |
+
"learning_rate": 4.965451197130373e-05,
|
| 366 |
+
"loss": 0.5313,
|
| 367 |
+
"step": 51
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.46741573033707867,
|
| 371 |
+
"grad_norm": 5.5,
|
| 372 |
+
"learning_rate": 4.961009251208368e-05,
|
| 373 |
+
"loss": 0.6564,
|
| 374 |
+
"step": 52
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.4764044943820225,
|
| 378 |
+
"grad_norm": 4.0625,
|
| 379 |
+
"learning_rate": 4.956300990555643e-05,
|
| 380 |
+
"loss": 0.6241,
|
| 381 |
+
"step": 53
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.4853932584269663,
|
| 385 |
+
"grad_norm": 2.890625,
|
| 386 |
+
"learning_rate": 4.951326924670148e-05,
|
| 387 |
+
"loss": 0.5819,
|
| 388 |
+
"step": 54
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4943820224719101,
|
| 392 |
+
"grad_norm": 3.40625,
|
| 393 |
+
"learning_rate": 4.9460875918135804e-05,
|
| 394 |
+
"loss": 0.5702,
|
| 395 |
+
"step": 55
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.503370786516854,
|
| 399 |
+
"grad_norm": 2.484375,
|
| 400 |
+
"learning_rate": 4.940583558953138e-05,
|
| 401 |
+
"loss": 0.5426,
|
| 402 |
+
"step": 56
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.5123595505617977,
|
| 406 |
+
"grad_norm": 2.34375,
|
| 407 |
+
"learning_rate": 4.934815421700165e-05,
|
| 408 |
+
"loss": 0.5366,
|
| 409 |
+
"step": 57
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5213483146067416,
|
| 413 |
+
"grad_norm": 2.3125,
|
| 414 |
+
"learning_rate": 4.9287838042456994e-05,
|
| 415 |
+
"loss": 0.5249,
|
| 416 |
+
"step": 58
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5303370786516854,
|
| 420 |
+
"grad_norm": 2.90625,
|
| 421 |
+
"learning_rate": 4.9224893592929275e-05,
|
| 422 |
+
"loss": 0.5797,
|
| 423 |
+
"step": 59
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5393258426966292,
|
| 427 |
+
"grad_norm": 2.515625,
|
| 428 |
+
"learning_rate": 4.9159327679865516e-05,
|
| 429 |
+
"loss": 0.5651,
|
| 430 |
+
"step": 60
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5483146067415731,
|
| 434 |
+
"grad_norm": 1.984375,
|
| 435 |
+
"learning_rate": 4.909114739839079e-05,
|
| 436 |
+
"loss": 0.5264,
|
| 437 |
+
"step": 61
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5573033707865168,
|
| 441 |
+
"grad_norm": 1.4140625,
|
| 442 |
+
"learning_rate": 4.9020360126540474e-05,
|
| 443 |
+
"loss": 0.5069,
|
| 444 |
+
"step": 62
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5662921348314607,
|
| 448 |
+
"grad_norm": 1.3046875,
|
| 449 |
+
"learning_rate": 4.894697352446182e-05,
|
| 450 |
+
"loss": 0.4606,
|
| 451 |
+
"step": 63
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5752808988764045,
|
| 455 |
+
"grad_norm": 1.9453125,
|
| 456 |
+
"learning_rate": 4.887099553358502e-05,
|
| 457 |
+
"loss": 0.5082,
|
| 458 |
+
"step": 64
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5842696629213483,
|
| 462 |
+
"grad_norm": 1.8828125,
|
| 463 |
+
"learning_rate": 4.879243437576383e-05,
|
| 464 |
+
"loss": 0.5532,
|
| 465 |
+
"step": 65
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5932584269662922,
|
| 469 |
+
"grad_norm": 1.2890625,
|
| 470 |
+
"learning_rate": 4.8711298552385886e-05,
|
| 471 |
+
"loss": 0.518,
|
| 472 |
+
"step": 66
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.6022471910112359,
|
| 476 |
+
"grad_norm": 1.6796875,
|
| 477 |
+
"learning_rate": 4.862759684345269e-05,
|
| 478 |
+
"loss": 0.4805,
|
| 479 |
+
"step": 67
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6112359550561798,
|
| 483 |
+
"grad_norm": 2.046875,
|
| 484 |
+
"learning_rate": 4.8541338306629545e-05,
|
| 485 |
+
"loss": 0.5163,
|
| 486 |
+
"step": 68
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.6202247191011236,
|
| 490 |
+
"grad_norm": 5.15625,
|
| 491 |
+
"learning_rate": 4.8452532276265364e-05,
|
| 492 |
+
"loss": 0.7373,
|
| 493 |
+
"step": 69
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6292134831460674,
|
| 497 |
+
"grad_norm": 3.984375,
|
| 498 |
+
"learning_rate": 4.836118836238253e-05,
|
| 499 |
+
"loss": 0.662,
|
| 500 |
+
"step": 70
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6382022471910113,
|
| 504 |
+
"grad_norm": 0.859375,
|
| 505 |
+
"learning_rate": 4.8267316449637054e-05,
|
| 506 |
+
"loss": 0.4725,
|
| 507 |
+
"step": 71
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.647191011235955,
|
| 511 |
+
"grad_norm": 1.234375,
|
| 512 |
+
"learning_rate": 4.817092669624883e-05,
|
| 513 |
+
"loss": 0.4659,
|
| 514 |
+
"step": 72
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6561797752808989,
|
| 518 |
+
"grad_norm": 1.6640625,
|
| 519 |
+
"learning_rate": 4.8072029532902426e-05,
|
| 520 |
+
"loss": 0.4787,
|
| 521 |
+
"step": 73
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.6651685393258427,
|
| 525 |
+
"grad_norm": 1.453125,
|
| 526 |
+
"learning_rate": 4.797063566161834e-05,
|
| 527 |
+
"loss": 0.4605,
|
| 528 |
+
"step": 74
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6741573033707865,
|
| 532 |
+
"grad_norm": 1.3125,
|
| 533 |
+
"learning_rate": 4.786675605459487e-05,
|
| 534 |
+
"loss": 0.4901,
|
| 535 |
+
"step": 75
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6831460674157304,
|
| 539 |
+
"grad_norm": 1.0625,
|
| 540 |
+
"learning_rate": 4.77604019530208e-05,
|
| 541 |
+
"loss": 0.47,
|
| 542 |
+
"step": 76
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6921348314606741,
|
| 546 |
+
"grad_norm": 1.171875,
|
| 547 |
+
"learning_rate": 4.76515848658589e-05,
|
| 548 |
+
"loss": 0.4549,
|
| 549 |
+
"step": 77
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.701123595505618,
|
| 553 |
+
"grad_norm": 1.5234375,
|
| 554 |
+
"learning_rate": 4.754031656860059e-05,
|
| 555 |
+
"loss": 0.4886,
|
| 556 |
+
"step": 78
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.7101123595505618,
|
| 560 |
+
"grad_norm": 1.28125,
|
| 561 |
+
"learning_rate": 4.7426609101991605e-05,
|
| 562 |
+
"loss": 0.4847,
|
| 563 |
+
"step": 79
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7191011235955056,
|
| 567 |
+
"grad_norm": 1.34375,
|
| 568 |
+
"learning_rate": 4.7310474770729e-05,
|
| 569 |
+
"loss": 0.4684,
|
| 570 |
+
"step": 80
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.7280898876404495,
|
| 574 |
+
"grad_norm": 1.4765625,
|
| 575 |
+
"learning_rate": 4.719192614212969e-05,
|
| 576 |
+
"loss": 0.4757,
|
| 577 |
+
"step": 81
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.7370786516853932,
|
| 581 |
+
"grad_norm": 1.234375,
|
| 582 |
+
"learning_rate": 4.707097604477045e-05,
|
| 583 |
+
"loss": 0.4765,
|
| 584 |
+
"step": 82
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7460674157303371,
|
| 588 |
+
"grad_norm": 1.3125,
|
| 589 |
+
"learning_rate": 4.694763756709967e-05,
|
| 590 |
+
"loss": 0.4733,
|
| 591 |
+
"step": 83
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.755056179775281,
|
| 595 |
+
"grad_norm": 1.1328125,
|
| 596 |
+
"learning_rate": 4.682192405602106e-05,
|
| 597 |
+
"loss": 0.4672,
|
| 598 |
+
"step": 84
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7640449438202247,
|
| 602 |
+
"grad_norm": 0.96484375,
|
| 603 |
+
"learning_rate": 4.669384911544927e-05,
|
| 604 |
+
"loss": 0.4621,
|
| 605 |
+
"step": 85
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.7730337078651686,
|
| 609 |
+
"grad_norm": 0.91796875,
|
| 610 |
+
"learning_rate": 4.656342660483782e-05,
|
| 611 |
+
"loss": 0.4391,
|
| 612 |
+
"step": 86
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7820224719101123,
|
| 616 |
+
"grad_norm": 1.15625,
|
| 617 |
+
"learning_rate": 4.6430670637679295e-05,
|
| 618 |
+
"loss": 0.4448,
|
| 619 |
+
"step": 87
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7910112359550562,
|
| 623 |
+
"grad_norm": 1.3984375,
|
| 624 |
+
"learning_rate": 4.6295595579978046e-05,
|
| 625 |
+
"loss": 0.453,
|
| 626 |
+
"step": 88
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.8,
|
| 630 |
+
"grad_norm": 1.3828125,
|
| 631 |
+
"learning_rate": 4.615821604869564e-05,
|
| 632 |
+
"loss": 0.4506,
|
| 633 |
+
"step": 89
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8089887640449438,
|
| 637 |
+
"grad_norm": 1.0546875,
|
| 638 |
+
"learning_rate": 4.601854691016907e-05,
|
| 639 |
+
"loss": 0.4382,
|
| 640 |
+
"step": 90
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8179775280898877,
|
| 644 |
+
"grad_norm": 0.98828125,
|
| 645 |
+
"learning_rate": 4.587660327850203e-05,
|
| 646 |
+
"loss": 0.4251,
|
| 647 |
+
"step": 91
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.8269662921348314,
|
| 651 |
+
"grad_norm": 0.89453125,
|
| 652 |
+
"learning_rate": 4.573240051392935e-05,
|
| 653 |
+
"loss": 0.3714,
|
| 654 |
+
"step": 92
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.8359550561797753,
|
| 658 |
+
"grad_norm": 0.90625,
|
| 659 |
+
"learning_rate": 4.5585954221154856e-05,
|
| 660 |
+
"loss": 0.4261,
|
| 661 |
+
"step": 93
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.8449438202247191,
|
| 665 |
+
"grad_norm": 1.0234375,
|
| 666 |
+
"learning_rate": 4.543728024766265e-05,
|
| 667 |
+
"loss": 0.4283,
|
| 668 |
+
"step": 94
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.8539325842696629,
|
| 672 |
+
"grad_norm": 1.2109375,
|
| 673 |
+
"learning_rate": 4.528639468200226e-05,
|
| 674 |
+
"loss": 0.4429,
|
| 675 |
+
"step": 95
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.8629213483146068,
|
| 679 |
+
"grad_norm": 1.296875,
|
| 680 |
+
"learning_rate": 4.513331385204761e-05,
|
| 681 |
+
"loss": 0.4091,
|
| 682 |
+
"step": 96
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8719101123595505,
|
| 686 |
+
"grad_norm": 1.390625,
|
| 687 |
+
"learning_rate": 4.497805432323015e-05,
|
| 688 |
+
"loss": 0.4684,
|
| 689 |
+
"step": 97
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8808988764044944,
|
| 693 |
+
"grad_norm": 1.328125,
|
| 694 |
+
"learning_rate": 4.482063289674618e-05,
|
| 695 |
+
"loss": 0.4105,
|
| 696 |
+
"step": 98
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.8898876404494382,
|
| 700 |
+
"grad_norm": 1.390625,
|
| 701 |
+
"learning_rate": 4.466106660773885e-05,
|
| 702 |
+
"loss": 0.4638,
|
| 703 |
+
"step": 99
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.898876404494382,
|
| 707 |
+
"grad_norm": 1.2421875,
|
| 708 |
+
"learning_rate": 4.44993727234546e-05,
|
| 709 |
+
"loss": 0.3974,
|
| 710 |
+
"step": 100
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.9078651685393259,
|
| 714 |
+
"grad_norm": 1.3515625,
|
| 715 |
+
"learning_rate": 4.43355687413747e-05,
|
| 716 |
+
"loss": 0.4689,
|
| 717 |
+
"step": 101
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.9168539325842696,
|
| 721 |
+
"grad_norm": 1.1953125,
|
| 722 |
+
"learning_rate": 4.4169672387321734e-05,
|
| 723 |
+
"loss": 0.4262,
|
| 724 |
+
"step": 102
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.9258426966292135,
|
| 728 |
+
"grad_norm": 0.98046875,
|
| 729 |
+
"learning_rate": 4.4001701613541456e-05,
|
| 730 |
+
"loss": 0.4375,
|
| 731 |
+
"step": 103
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.9348314606741573,
|
| 735 |
+
"grad_norm": 0.9296875,
|
| 736 |
+
"learning_rate": 4.3831674596760086e-05,
|
| 737 |
+
"loss": 0.4434,
|
| 738 |
+
"step": 104
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.9438202247191011,
|
| 742 |
+
"grad_norm": 0.66015625,
|
| 743 |
+
"learning_rate": 4.3659609736217344e-05,
|
| 744 |
+
"loss": 0.4107,
|
| 745 |
+
"step": 105
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.952808988764045,
|
| 749 |
+
"grad_norm": 0.734375,
|
| 750 |
+
"learning_rate": 4.348552565167543e-05,
|
| 751 |
+
"loss": 0.39,
|
| 752 |
+
"step": 106
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.9617977528089887,
|
| 756 |
+
"grad_norm": 0.7890625,
|
| 757 |
+
"learning_rate": 4.330944118140407e-05,
|
| 758 |
+
"loss": 0.3889,
|
| 759 |
+
"step": 107
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.9707865168539326,
|
| 763 |
+
"grad_norm": 0.78515625,
|
| 764 |
+
"learning_rate": 4.3131375380141987e-05,
|
| 765 |
+
"loss": 0.3995,
|
| 766 |
+
"step": 108
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.9797752808988764,
|
| 770 |
+
"grad_norm": 0.7265625,
|
| 771 |
+
"learning_rate": 4.295134751703493e-05,
|
| 772 |
+
"loss": 0.3878,
|
| 773 |
+
"step": 109
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.9887640449438202,
|
| 777 |
+
"grad_norm": 0.90625,
|
| 778 |
+
"learning_rate": 4.276937707355044e-05,
|
| 779 |
+
"loss": 0.4239,
|
| 780 |
+
"step": 110
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9977528089887641,
|
| 784 |
+
"grad_norm": 1.015625,
|
| 785 |
+
"learning_rate": 4.2585483741369755e-05,
|
| 786 |
+
"loss": 0.4019,
|
| 787 |
+
"step": 111
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.0,
|
| 791 |
+
"grad_norm": 1.03125,
|
| 792 |
+
"learning_rate": 4.239968742025684e-05,
|
| 793 |
+
"loss": 0.3925,
|
| 794 |
+
"step": 112
|
| 795 |
+
}
|
| 796 |
+
],
|
| 797 |
+
"logging_steps": 1,
|
| 798 |
+
"max_steps": 336,
|
| 799 |
+
"num_input_tokens_seen": 0,
|
| 800 |
+
"num_train_epochs": 3,
|
| 801 |
+
"save_steps": 500,
|
| 802 |
+
"stateful_callbacks": {
|
| 803 |
+
"TrainerControl": {
|
| 804 |
+
"args": {
|
| 805 |
+
"should_epoch_stop": false,
|
| 806 |
+
"should_evaluate": false,
|
| 807 |
+
"should_log": false,
|
| 808 |
+
"should_save": true,
|
| 809 |
+
"should_training_stop": false
|
| 810 |
+
},
|
| 811 |
+
"attributes": {}
|
| 812 |
+
}
|
| 813 |
+
},
|
| 814 |
+
"total_flos": 1.1464243358466048e+17,
|
| 815 |
+
"train_batch_size": 4,
|
| 816 |
+
"trial_name": null,
|
| 817 |
+
"trial_params": null
|
| 818 |
+
}
|
checkpoint-112/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c14f0c2559ff2541be54fd4abbd909fcc8b6ae56ec977d4f1ee0355ffd1bcba
|
| 3 |
+
size 5560
|