Checkpoint at step 40
Browse files- checkpoints/checkpoint-20/small_experts_and_gates.bin +3 -0
- checkpoints/checkpoint-40/config.json +38 -0
- checkpoints/checkpoint-40/generation_config.json +6 -0
- checkpoints/checkpoint-40/model-00001-of-00003.safetensors +3 -0
- checkpoints/checkpoint-40/model-00002-of-00003.safetensors +3 -0
- checkpoints/checkpoint-40/model-00003-of-00003.safetensors +3 -0
- checkpoints/checkpoint-40/model.safetensors.index.json +0 -0
- checkpoints/checkpoint-40/optimizer.pt +3 -0
- checkpoints/checkpoint-40/rng_state.pth +3 -0
- checkpoints/checkpoint-40/scheduler.pt +3 -0
- checkpoints/checkpoint-40/trainer_state.json +62 -0
- checkpoints/checkpoint-40/training_args.bin +3 -0
- logs/events.out.tfevents.1756831118.5624918391e3.1490.0 +2 -2
checkpoints/checkpoint-20/small_experts_and_gates.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0a1c589f5de5e24b6da0665fb82d1950a722bcadea7cbaab535b51df184e4bc
|
| 3 |
+
size 206711911
|
checkpoints/checkpoint-40/config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MyOlmoeForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"clip_qkv": null,
|
| 8 |
+
"eos_token_id": 50279,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 2048,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 1024,
|
| 13 |
+
"max_position_embeddings": 4096,
|
| 14 |
+
"max_small_expert_count": 64,
|
| 15 |
+
"model_type": "olmoe",
|
| 16 |
+
"norm_topk_prob": false,
|
| 17 |
+
"num_attention_heads": 16,
|
| 18 |
+
"num_experts": 64,
|
| 19 |
+
"num_experts_per_tok": 2,
|
| 20 |
+
"num_hidden_layers": 16,
|
| 21 |
+
"num_key_value_heads": 16,
|
| 22 |
+
"num_small_experts": 64,
|
| 23 |
+
"output_router_logits": false,
|
| 24 |
+
"pad_token_id": 1,
|
| 25 |
+
"rms_norm_eps": 1e-05,
|
| 26 |
+
"rope_scaling": null,
|
| 27 |
+
"rope_theta": 10000.0,
|
| 28 |
+
"router_aux_loss_coef": 0.01,
|
| 29 |
+
"small_expert_count": 64,
|
| 30 |
+
"small_expert_intermediate_ratio": 64,
|
| 31 |
+
"small_expert_intermediate_size": 0,
|
| 32 |
+
"small_expert_sparsity_coef": 0.1,
|
| 33 |
+
"tie_word_embeddings": false,
|
| 34 |
+
"torch_dtype": "bfloat16",
|
| 35 |
+
"transformers_version": "4.55.4",
|
| 36 |
+
"use_cache": true,
|
| 37 |
+
"vocab_size": 50304
|
| 38 |
+
}
|
checkpoints/checkpoint-40/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": 50279,
|
| 4 |
+
"pad_token_id": 1,
|
| 5 |
+
"transformers_version": "4.55.4"
|
| 6 |
+
}
|
checkpoints/checkpoint-40/model-00001-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:114400ea3c1f6b4bf8985e9f050651f88d39f370453e65343bbaa9d1d34d65a2
|
| 3 |
+
size 4998915096
|
checkpoints/checkpoint-40/model-00002-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2231902fc66ad75a3b8f3686c1a68d275554007b4678fad5d73d74f364a78a16
|
| 3 |
+
size 4998953920
|
checkpoints/checkpoint-40/model-00003-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fec36ff27e7cffe29172cd7c3426cec7a63e832e4a36fd3dd8d4cf337082135e
|
| 3 |
+
size 4046769808
|
checkpoints/checkpoint-40/model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints/checkpoint-40/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5592a107e3807fc0b70ae71b93a98f020ad60d21c3c068b7be1bbbac2001b89e
|
| 3 |
+
size 391093394
|
checkpoints/checkpoint-40/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
|
| 3 |
+
size 14645
|
checkpoints/checkpoint-40/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b724dee474d2f28b5e7cd28f30ad6358e8e4bd90e6205eb9f6ebd8fb296e614
|
| 3 |
+
size 1465
|
checkpoints/checkpoint-40/trainer_state.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.001962263225347535,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 40,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0004905658063368838,
|
| 14 |
+
"grad_norm": 55.75,
|
| 15 |
+
"learning_rate": 1.471550032701112e-07,
|
| 16 |
+
"loss": 106.74,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0009811316126737675,
|
| 21 |
+
"grad_norm": 53.5,
|
| 22 |
+
"learning_rate": 3.106605624591236e-07,
|
| 23 |
+
"loss": 106.5866,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0014716974190106514,
|
| 28 |
+
"grad_norm": 52.75,
|
| 29 |
+
"learning_rate": 4.7416612164813603e-07,
|
| 30 |
+
"loss": 107.8943,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.001962263225347535,
|
| 35 |
+
"grad_norm": 68.5,
|
| 36 |
+
"learning_rate": 6.376716808371485e-07,
|
| 37 |
+
"loss": 109.0438,
|
| 38 |
+
"step": 40
|
| 39 |
+
}
|
| 40 |
+
],
|
| 41 |
+
"logging_steps": 10,
|
| 42 |
+
"max_steps": 61155,
|
| 43 |
+
"num_input_tokens_seen": 0,
|
| 44 |
+
"num_train_epochs": 3,
|
| 45 |
+
"save_steps": 20,
|
| 46 |
+
"stateful_callbacks": {
|
| 47 |
+
"TrainerControl": {
|
| 48 |
+
"args": {
|
| 49 |
+
"should_epoch_stop": false,
|
| 50 |
+
"should_evaluate": false,
|
| 51 |
+
"should_log": false,
|
| 52 |
+
"should_save": true,
|
| 53 |
+
"should_training_stop": false
|
| 54 |
+
},
|
| 55 |
+
"attributes": {}
|
| 56 |
+
}
|
| 57 |
+
},
|
| 58 |
+
"total_flos": 1.0882488276615168e+17,
|
| 59 |
+
"train_batch_size": 2,
|
| 60 |
+
"trial_name": null,
|
| 61 |
+
"trial_params": null
|
| 62 |
+
}
|
checkpoints/checkpoint-40/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ffa490bde32401dd6d70c4f1b1cff8f5df114f94b79824d3a47b2ae8c00b822d
|
| 3 |
+
size 5713
|
logs/events.out.tfevents.1756831118.5624918391e3.1490.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b893f207f0aecf3668043c1d7255468fc0a035f088b014a4f0616628302defc
|
| 3 |
+
size 6072
|