Upload Rick dialogue fine-tuned model
Browse files
Crypto-adapters/adapter_config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": null,
|
| 6 |
+
"base_model_name_or_path": "Qwen/Qwen3-4B",
|
| 7 |
+
"bias": "none",
|
| 8 |
+
"corda_config": null,
|
| 9 |
+
"ensure_weight_tying": false,
|
| 10 |
+
"eva_config": null,
|
| 11 |
+
"exclude_modules": null,
|
| 12 |
+
"fan_in_fan_out": false,
|
| 13 |
+
"inference_mode": true,
|
| 14 |
+
"init_lora_weights": true,
|
| 15 |
+
"layer_replication": null,
|
| 16 |
+
"layers_pattern": null,
|
| 17 |
+
"layers_to_transform": null,
|
| 18 |
+
"loftq_config": {},
|
| 19 |
+
"lora_alpha": 128,
|
| 20 |
+
"lora_bias": false,
|
| 21 |
+
"lora_dropout": 0.05,
|
| 22 |
+
"megatron_config": null,
|
| 23 |
+
"megatron_core": "megatron.core",
|
| 24 |
+
"modules_to_save": null,
|
| 25 |
+
"peft_type": "LORA",
|
| 26 |
+
"peft_version": "0.18.0",
|
| 27 |
+
"qalora_group_size": 16,
|
| 28 |
+
"r": 64,
|
| 29 |
+
"rank_pattern": {},
|
| 30 |
+
"revision": null,
|
| 31 |
+
"target_modules": [
|
| 32 |
+
"gate_proj",
|
| 33 |
+
"q_proj",
|
| 34 |
+
"down_proj",
|
| 35 |
+
"o_proj",
|
| 36 |
+
"k_proj",
|
| 37 |
+
"up_proj",
|
| 38 |
+
"v_proj"
|
| 39 |
+
],
|
| 40 |
+
"target_parameters": null,
|
| 41 |
+
"task_type": "CAUSAL_LM",
|
| 42 |
+
"trainable_token_indices": null,
|
| 43 |
+
"use_dora": false,
|
| 44 |
+
"use_qalora": false,
|
| 45 |
+
"use_rslora": false
|
| 46 |
+
}
|
Crypto-adapters/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eb061e60695fc070d253b36257beb9a9218e90d31ecb6c0668b3e3f2db5c2177
|
| 3 |
+
size 528550256
|
Crypto-adapters/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c545c10c0f16ad53a06e19e15dad68399da0e7947fe7e1da741ad9af8c4c1e8d
|
| 3 |
+
size 6225
|
Crypto-adapters/training_history.json
ADDED
|
@@ -0,0 +1,1123 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"loss": 4.2889,
|
| 4 |
+
"grad_norm": 4.082701683044434,
|
| 5 |
+
"learning_rate": 4.090909090909091e-05,
|
| 6 |
+
"entropy": 1.8044173419475555,
|
| 7 |
+
"num_tokens": 4753.0,
|
| 8 |
+
"mean_token_accuracy": 0.41228369399905207,
|
| 9 |
+
"epoch": 0.018281535648994516,
|
| 10 |
+
"step": 10
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"loss": 1.9004,
|
| 14 |
+
"grad_norm": 1.8841173648834229,
|
| 15 |
+
"learning_rate": 8.636363636363636e-05,
|
| 16 |
+
"entropy": 1.8375275656580925,
|
| 17 |
+
"num_tokens": 9823.0,
|
| 18 |
+
"mean_token_accuracy": 0.6945700973272324,
|
| 19 |
+
"epoch": 0.03656307129798903,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"loss": 1.2667,
|
| 24 |
+
"grad_norm": 3.267770528793335,
|
| 25 |
+
"learning_rate": 0.0001318181818181818,
|
| 26 |
+
"entropy": 1.2633977964520455,
|
| 27 |
+
"num_tokens": 14808.0,
|
| 28 |
+
"mean_token_accuracy": 0.791448101401329,
|
| 29 |
+
"epoch": 0.054844606946983544,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"loss": 0.9755,
|
| 34 |
+
"grad_norm": 1.05254328250885,
|
| 35 |
+
"learning_rate": 0.00017727272727272728,
|
| 36 |
+
"entropy": 0.9446997016668319,
|
| 37 |
+
"num_tokens": 19622.0,
|
| 38 |
+
"mean_token_accuracy": 0.8207138180732727,
|
| 39 |
+
"epoch": 0.07312614259597806,
|
| 40 |
+
"step": 40
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"loss": 1.1448,
|
| 44 |
+
"grad_norm": 0.906599760055542,
|
| 45 |
+
"learning_rate": 0.00022272727272727272,
|
| 46 |
+
"entropy": 1.015735039114952,
|
| 47 |
+
"num_tokens": 24595.0,
|
| 48 |
+
"mean_token_accuracy": 0.8062975570559502,
|
| 49 |
+
"epoch": 0.09140767824497258,
|
| 50 |
+
"step": 50
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"loss": 1.1366,
|
| 54 |
+
"grad_norm": 1.0311369895935059,
|
| 55 |
+
"learning_rate": 0.0002681818181818182,
|
| 56 |
+
"entropy": 1.041224654763937,
|
| 57 |
+
"num_tokens": 29587.0,
|
| 58 |
+
"mean_token_accuracy": 0.8060152858495713,
|
| 59 |
+
"epoch": 0.10968921389396709,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"loss": 0.7749,
|
| 64 |
+
"grad_norm": 1.1635301113128662,
|
| 65 |
+
"learning_rate": 0.00031363636363636365,
|
| 66 |
+
"entropy": 0.7659586444497108,
|
| 67 |
+
"num_tokens": 34200.0,
|
| 68 |
+
"mean_token_accuracy": 0.8529342606663703,
|
| 69 |
+
"epoch": 0.12797074954296161,
|
| 70 |
+
"step": 70
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"loss": 1.0122,
|
| 74 |
+
"grad_norm": 1.0612354278564453,
|
| 75 |
+
"learning_rate": 0.00035909090909090906,
|
| 76 |
+
"entropy": 0.8900700502097607,
|
| 77 |
+
"num_tokens": 38881.0,
|
| 78 |
+
"mean_token_accuracy": 0.8307199895381927,
|
| 79 |
+
"epoch": 0.14625228519195613,
|
| 80 |
+
"step": 80
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"loss": 1.0642,
|
| 84 |
+
"grad_norm": 0.9732877016067505,
|
| 85 |
+
"learning_rate": 0.0004045454545454546,
|
| 86 |
+
"entropy": 0.9888352438807487,
|
| 87 |
+
"num_tokens": 43786.0,
|
| 88 |
+
"mean_token_accuracy": 0.8153566777706146,
|
| 89 |
+
"epoch": 0.16453382084095064,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"loss": 1.0782,
|
| 94 |
+
"grad_norm": 1.045189619064331,
|
| 95 |
+
"learning_rate": 0.00045000000000000004,
|
| 96 |
+
"entropy": 1.0296132363379003,
|
| 97 |
+
"num_tokens": 48748.0,
|
| 98 |
+
"mean_token_accuracy": 0.8004682093858719,
|
| 99 |
+
"epoch": 0.18281535648994515,
|
| 100 |
+
"step": 100
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"loss": 1.1802,
|
| 104 |
+
"grad_norm": 1.5044333934783936,
|
| 105 |
+
"learning_rate": 0.0004954545454545455,
|
| 106 |
+
"entropy": 1.0957321308553218,
|
| 107 |
+
"num_tokens": 53965.0,
|
| 108 |
+
"mean_token_accuracy": 0.7918785288929939,
|
| 109 |
+
"epoch": 0.20109689213893966,
|
| 110 |
+
"step": 110
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"loss": 1.0926,
|
| 114 |
+
"grad_norm": 1.3921072483062744,
|
| 115 |
+
"learning_rate": 0.0004998968011877129,
|
| 116 |
+
"entropy": 0.9945384554564953,
|
| 117 |
+
"num_tokens": 58917.0,
|
| 118 |
+
"mean_token_accuracy": 0.8105567440390586,
|
| 119 |
+
"epoch": 0.21937842778793418,
|
| 120 |
+
"step": 120
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"loss": 1.1154,
|
| 124 |
+
"grad_norm": 1.3083264827728271,
|
| 125 |
+
"learning_rate": 0.0004995401739351536,
|
| 126 |
+
"entropy": 1.0778948426246644,
|
| 127 |
+
"num_tokens": 63922.0,
|
| 128 |
+
"mean_token_accuracy": 0.8031231343746186,
|
| 129 |
+
"epoch": 0.2376599634369287,
|
| 130 |
+
"step": 130
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"loss": 1.0033,
|
| 134 |
+
"grad_norm": 2.4141838550567627,
|
| 135 |
+
"learning_rate": 0.0004989292075819091,
|
| 136 |
+
"entropy": 0.9672931365668773,
|
| 137 |
+
"num_tokens": 68697.0,
|
| 138 |
+
"mean_token_accuracy": 0.8235540971159935,
|
| 139 |
+
"epoch": 0.25594149908592323,
|
| 140 |
+
"step": 140
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"loss": 1.1671,
|
| 144 |
+
"grad_norm": 1.0534119606018066,
|
| 145 |
+
"learning_rate": 0.0004980645248438745,
|
| 146 |
+
"entropy": 1.0584807768464088,
|
| 147 |
+
"num_tokens": 73708.0,
|
| 148 |
+
"mean_token_accuracy": 0.7992900639772416,
|
| 149 |
+
"epoch": 0.2742230347349177,
|
| 150 |
+
"step": 150
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"loss": 0.9638,
|
| 154 |
+
"grad_norm": 2.2693159580230713,
|
| 155 |
+
"learning_rate": 0.0004969470070325699,
|
| 156 |
+
"entropy": 0.926125219464302,
|
| 157 |
+
"num_tokens": 78399.0,
|
| 158 |
+
"mean_token_accuracy": 0.8329230979084968,
|
| 159 |
+
"epoch": 0.29250457038391225,
|
| 160 |
+
"step": 160
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"loss": 1.1336,
|
| 164 |
+
"grad_norm": 0.9418737888336182,
|
| 165 |
+
"learning_rate": 0.0004955777931568797,
|
| 166 |
+
"entropy": 1.0604520812630653,
|
| 167 |
+
"num_tokens": 83346.0,
|
| 168 |
+
"mean_token_accuracy": 0.8043849408626557,
|
| 169 |
+
"epoch": 0.31078610603290674,
|
| 170 |
+
"step": 170
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"loss": 1.0936,
|
| 174 |
+
"grad_norm": 1.2109498977661133,
|
| 175 |
+
"learning_rate": 0.0004939582787621393,
|
| 176 |
+
"entropy": 0.9868959777057171,
|
| 177 |
+
"num_tokens": 88297.0,
|
| 178 |
+
"mean_token_accuracy": 0.8075228631496429,
|
| 179 |
+
"epoch": 0.3290676416819013,
|
| 180 |
+
"step": 180
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"loss": 0.9634,
|
| 184 |
+
"grad_norm": 1.1354761123657227,
|
| 185 |
+
"learning_rate": 0.0004920901145077526,
|
| 186 |
+
"entropy": 0.9269359841942787,
|
| 187 |
+
"num_tokens": 93120.0,
|
| 188 |
+
"mean_token_accuracy": 0.8204815790057183,
|
| 189 |
+
"epoch": 0.3473491773308958,
|
| 190 |
+
"step": 190
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"loss": 0.902,
|
| 194 |
+
"grad_norm": 1.1963611841201782,
|
| 195 |
+
"learning_rate": 0.0004899752044847881,
|
| 196 |
+
"entropy": 0.809983392059803,
|
| 197 |
+
"num_tokens": 97836.0,
|
| 198 |
+
"mean_token_accuracy": 0.8354423463344574,
|
| 199 |
+
"epoch": 0.3656307129798903,
|
| 200 |
+
"step": 200
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"loss": 1.0727,
|
| 204 |
+
"grad_norm": 1.2923297882080078,
|
| 205 |
+
"learning_rate": 0.0004876157042752697,
|
| 206 |
+
"entropy": 1.0803617157042027,
|
| 207 |
+
"num_tokens": 102707.0,
|
| 208 |
+
"mean_token_accuracy": 0.804425984621048,
|
| 209 |
+
"epoch": 0.38391224862888484,
|
| 210 |
+
"step": 210
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"loss": 1.0431,
|
| 214 |
+
"grad_norm": 1.0821493864059448,
|
| 215 |
+
"learning_rate": 0.00048501401875514173,
|
| 216 |
+
"entropy": 0.9232283338904381,
|
| 217 |
+
"num_tokens": 107578.0,
|
| 218 |
+
"mean_token_accuracy": 0.8204686373472214,
|
| 219 |
+
"epoch": 0.40219378427787933,
|
| 220 |
+
"step": 220
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"loss": 1.0016,
|
| 224 |
+
"grad_norm": 1.3452186584472656,
|
| 225 |
+
"learning_rate": 0.0004821727996431435,
|
| 226 |
+
"entropy": 0.8946781434118748,
|
| 227 |
+
"num_tokens": 112492.0,
|
| 228 |
+
"mean_token_accuracy": 0.8133098736405373,
|
| 229 |
+
"epoch": 0.42047531992687387,
|
| 230 |
+
"step": 230
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"loss": 0.9545,
|
| 234 |
+
"grad_norm": 1.3005157709121704,
|
| 235 |
+
"learning_rate": 0.00047909494279809557,
|
| 236 |
+
"entropy": 0.9158567085862159,
|
| 237 |
+
"num_tokens": 117326.0,
|
| 238 |
+
"mean_token_accuracy": 0.8263604387640953,
|
| 239 |
+
"epoch": 0.43875685557586835,
|
| 240 |
+
"step": 240
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"loss": 1.0434,
|
| 244 |
+
"grad_norm": 1.3348708152770996,
|
| 245 |
+
"learning_rate": 0.0004757835852673506,
|
| 246 |
+
"entropy": 0.9666553139686584,
|
| 247 |
+
"num_tokens": 122188.0,
|
| 248 |
+
"mean_token_accuracy": 0.8108666554093361,
|
| 249 |
+
"epoch": 0.4570383912248629,
|
| 250 |
+
"step": 250
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"loss": 1.5233,
|
| 254 |
+
"grad_norm": 5.236101150512695,
|
| 255 |
+
"learning_rate": 0.00047224210208941694,
|
| 256 |
+
"entropy": 1.3487498432397842,
|
| 257 |
+
"num_tokens": 127137.0,
|
| 258 |
+
"mean_token_accuracy": 0.7540628828108311,
|
| 259 |
+
"epoch": 0.4753199268738574,
|
| 260 |
+
"step": 260
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"loss": 1.0924,
|
| 264 |
+
"grad_norm": 2.207545280456543,
|
| 265 |
+
"learning_rate": 0.0004684741028540146,
|
| 266 |
+
"entropy": 0.9830687694251538,
|
| 267 |
+
"num_tokens": 132032.0,
|
| 268 |
+
"mean_token_accuracy": 0.809441576898098,
|
| 269 |
+
"epoch": 0.4936014625228519,
|
| 270 |
+
"step": 270
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"loss": 1.9378,
|
| 274 |
+
"grad_norm": 1.80031418800354,
|
| 275 |
+
"learning_rate": 0.0004644834280230692,
|
| 276 |
+
"entropy": 1.6831681944429875,
|
| 277 |
+
"num_tokens": 136810.0,
|
| 278 |
+
"mean_token_accuracy": 0.6984923191368579,
|
| 279 |
+
"epoch": 0.5118829981718465,
|
| 280 |
+
"step": 280
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"loss": 1.0963,
|
| 284 |
+
"grad_norm": 1.824531078338623,
|
| 285 |
+
"learning_rate": 0.00046027414501639396,
|
| 286 |
+
"entropy": 0.9923494562506676,
|
| 287 |
+
"num_tokens": 141832.0,
|
| 288 |
+
"mean_token_accuracy": 0.7938514590263367,
|
| 289 |
+
"epoch": 0.5301645338208409,
|
| 290 |
+
"step": 290
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"loss": 0.9917,
|
| 294 |
+
"grad_norm": 1.9162729978561401,
|
| 295 |
+
"learning_rate": 0.00045585054406604866,
|
| 296 |
+
"entropy": 0.938605896383524,
|
| 297 |
+
"num_tokens": 146777.0,
|
| 298 |
+
"mean_token_accuracy": 0.8107017129659653,
|
| 299 |
+
"epoch": 0.5484460694698354,
|
| 300 |
+
"step": 300
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"loss": 1.4443,
|
| 304 |
+
"grad_norm": 9.322174072265625,
|
| 305 |
+
"learning_rate": 0.0004512171338436021,
|
| 306 |
+
"entropy": 1.295152136683464,
|
| 307 |
+
"num_tokens": 151950.0,
|
| 308 |
+
"mean_token_accuracy": 0.7566543176770211,
|
| 309 |
+
"epoch": 0.56672760511883,
|
| 310 |
+
"step": 310
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"loss": 1.0499,
|
| 314 |
+
"grad_norm": 1.8375171422958374,
|
| 315 |
+
"learning_rate": 0.000446378636864754,
|
| 316 |
+
"entropy": 0.9144886791706085,
|
| 317 |
+
"num_tokens": 156537.0,
|
| 318 |
+
"mean_token_accuracy": 0.8165208205580712,
|
| 319 |
+
"epoch": 0.5850091407678245,
|
| 320 |
+
"step": 320
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"loss": 1.1253,
|
| 324 |
+
"grad_norm": 1.7732634544372559,
|
| 325 |
+
"learning_rate": 0.00044133998467599977,
|
| 326 |
+
"entropy": 1.0305592842400073,
|
| 327 |
+
"num_tokens": 161418.0,
|
| 328 |
+
"mean_token_accuracy": 0.7921615049242974,
|
| 329 |
+
"epoch": 0.603290676416819,
|
| 330 |
+
"step": 330
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"loss": 1.5316,
|
| 334 |
+
"grad_norm": 1.4331961870193481,
|
| 335 |
+
"learning_rate": 0.00043610631282824553,
|
| 336 |
+
"entropy": 1.325055579841137,
|
| 337 |
+
"num_tokens": 166279.0,
|
| 338 |
+
"mean_token_accuracy": 0.7350606312975287,
|
| 339 |
+
"epoch": 0.6215722120658135,
|
| 340 |
+
"step": 340
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"loss": 0.9966,
|
| 344 |
+
"grad_norm": 1.8411200046539307,
|
| 345 |
+
"learning_rate": 0.0004306829556424948,
|
| 346 |
+
"entropy": 0.8855753645300866,
|
| 347 |
+
"num_tokens": 171214.0,
|
| 348 |
+
"mean_token_accuracy": 0.8123403206467629,
|
| 349 |
+
"epoch": 0.6398537477148081,
|
| 350 |
+
"step": 350
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"loss": 1.1294,
|
| 354 |
+
"grad_norm": 1.0451457500457764,
|
| 355 |
+
"learning_rate": 0.00042507544077294274,
|
| 356 |
+
"entropy": 1.0525127850472926,
|
| 357 |
+
"num_tokens": 176265.0,
|
| 358 |
+
"mean_token_accuracy": 0.8036296904087067,
|
| 359 |
+
"epoch": 0.6581352833638026,
|
| 360 |
+
"step": 360
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"loss": 1.0186,
|
| 364 |
+
"grad_norm": 1.330788493156433,
|
| 365 |
+
"learning_rate": 0.00041928948357301935,
|
| 366 |
+
"entropy": 0.9291763827204704,
|
| 367 |
+
"num_tokens": 181157.0,
|
| 368 |
+
"mean_token_accuracy": 0.8202641427516937,
|
| 369 |
+
"epoch": 0.676416819012797,
|
| 370 |
+
"step": 370
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"loss": 0.9623,
|
| 374 |
+
"grad_norm": 1.2461735010147095,
|
| 375 |
+
"learning_rate": 0.00041333098127012326,
|
| 376 |
+
"entropy": 0.8654836975038052,
|
| 377 |
+
"num_tokens": 186130.0,
|
| 378 |
+
"mean_token_accuracy": 0.8226122915744781,
|
| 379 |
+
"epoch": 0.6946983546617916,
|
| 380 |
+
"step": 380
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"loss": 1.277,
|
| 384 |
+
"grad_norm": 1.0772674083709717,
|
| 385 |
+
"learning_rate": 0.00040720600695498484,
|
| 386 |
+
"entropy": 1.1583830490708351,
|
| 387 |
+
"num_tokens": 191459.0,
|
| 388 |
+
"mean_token_accuracy": 0.7833705425262452,
|
| 389 |
+
"epoch": 0.7129798903107861,
|
| 390 |
+
"step": 390
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"loss": 0.7887,
|
| 394 |
+
"grad_norm": 0.8381028771400452,
|
| 395 |
+
"learning_rate": 0.0004009208033917829,
|
| 396 |
+
"entropy": 0.7390592843294144,
|
| 397 |
+
"num_tokens": 196161.0,
|
| 398 |
+
"mean_token_accuracy": 0.8589992865920066,
|
| 399 |
+
"epoch": 0.7312614259597806,
|
| 400 |
+
"step": 400
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"loss": 0.9316,
|
| 404 |
+
"grad_norm": 1.2052291631698608,
|
| 405 |
+
"learning_rate": 0.00039448177665532574,
|
| 406 |
+
"entropy": 0.9087510347366333,
|
| 407 |
+
"num_tokens": 201115.0,
|
| 408 |
+
"mean_token_accuracy": 0.8239080429077148,
|
| 409 |
+
"epoch": 0.7495429616087751,
|
| 410 |
+
"step": 410
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"loss": 0.7695,
|
| 414 |
+
"grad_norm": 1.2508081197738647,
|
| 415 |
+
"learning_rate": 0.00038789548960178037,
|
| 416 |
+
"entropy": 0.7133356228470802,
|
| 417 |
+
"num_tokens": 205917.0,
|
| 418 |
+
"mean_token_accuracy": 0.849728773534298,
|
| 419 |
+
"epoch": 0.7678244972577697,
|
| 420 |
+
"step": 420
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"loss": 0.9197,
|
| 424 |
+
"grad_norm": 1.1393840312957764,
|
| 425 |
+
"learning_rate": 0.0003811686551796058,
|
| 426 |
+
"entropy": 0.877166099101305,
|
| 427 |
+
"num_tokens": 210755.0,
|
| 428 |
+
"mean_token_accuracy": 0.8312482610344887,
|
| 429 |
+
"epoch": 0.7861060329067642,
|
| 430 |
+
"step": 430
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"loss": 0.9159,
|
| 434 |
+
"grad_norm": 0.9043849110603333,
|
| 435 |
+
"learning_rate": 0.0003743081295875069,
|
| 436 |
+
"entropy": 0.8492414981126786,
|
| 437 |
+
"num_tokens": 215637.0,
|
| 438 |
+
"mean_token_accuracy": 0.8339142099022865,
|
| 439 |
+
"epoch": 0.8043875685557587,
|
| 440 |
+
"step": 440
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"loss": 0.8508,
|
| 444 |
+
"grad_norm": 0.9570634961128235,
|
| 445 |
+
"learning_rate": 0.0003673209052863843,
|
| 446 |
+
"entropy": 0.7381240986287594,
|
| 447 |
+
"num_tokens": 220483.0,
|
| 448 |
+
"mean_token_accuracy": 0.842980583012104,
|
| 449 |
+
"epoch": 0.8226691042047533,
|
| 450 |
+
"step": 450
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"loss": 0.9842,
|
| 454 |
+
"grad_norm": 1.2218940258026123,
|
| 455 |
+
"learning_rate": 0.00036021410387240013,
|
| 456 |
+
"entropy": 0.8695680610835552,
|
| 457 |
+
"num_tokens": 225469.0,
|
| 458 |
+
"mean_token_accuracy": 0.8326078191399574,
|
| 459 |
+
"epoch": 0.8409506398537477,
|
| 460 |
+
"step": 460
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"loss": 0.9256,
|
| 464 |
+
"grad_norm": 0.8783425092697144,
|
| 465 |
+
"learning_rate": 0.0003529949688184265,
|
| 466 |
+
"entropy": 0.9371243461966514,
|
| 467 |
+
"num_tokens": 230278.0,
|
| 468 |
+
"mean_token_accuracy": 0.8261681437492371,
|
| 469 |
+
"epoch": 0.8592321755027422,
|
| 470 |
+
"step": 470
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"loss": 0.8089,
|
| 474 |
+
"grad_norm": 0.9584093689918518,
|
| 475 |
+
"learning_rate": 0.0003456708580912725,
|
| 476 |
+
"entropy": 0.7249465331435203,
|
| 477 |
+
"num_tokens": 234976.0,
|
| 478 |
+
"mean_token_accuracy": 0.85187376588583,
|
| 479 |
+
"epoch": 0.8775137111517367,
|
| 480 |
+
"step": 480
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"loss": 0.9988,
|
| 484 |
+
"grad_norm": 1.0121896266937256,
|
| 485 |
+
"learning_rate": 0.0003382492366522158,
|
| 486 |
+
"entropy": 0.8962770074605941,
|
| 487 |
+
"num_tokens": 239994.0,
|
| 488 |
+
"mean_token_accuracy": 0.8272333428263664,
|
| 489 |
+
"epoch": 0.8957952468007313,
|
| 490 |
+
"step": 490
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"loss": 0.8,
|
| 494 |
+
"grad_norm": 1.0218371152877808,
|
| 495 |
+
"learning_rate": 0.00033073766884848235,
|
| 496 |
+
"entropy": 0.7547712914645672,
|
| 497 |
+
"num_tokens": 244672.0,
|
| 498 |
+
"mean_token_accuracy": 0.8585162028670311,
|
| 499 |
+
"epoch": 0.9140767824497258,
|
| 500 |
+
"step": 500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"loss": 0.7155,
|
| 504 |
+
"grad_norm": 1.0712863206863403,
|
| 505 |
+
"learning_rate": 0.00032314381070342813,
|
| 506 |
+
"entropy": 0.6817032858729363,
|
| 507 |
+
"num_tokens": 249226.0,
|
| 508 |
+
"mean_token_accuracy": 0.8629905194044113,
|
| 509 |
+
"epoch": 0.9323583180987203,
|
| 510 |
+
"step": 510
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"loss": 0.9886,
|
| 514 |
+
"grad_norm": 0.9817427396774292,
|
| 515 |
+
"learning_rate": 0.0003154754021132827,
|
| 516 |
+
"entropy": 0.9045170530676842,
|
| 517 |
+
"num_tokens": 254245.0,
|
| 518 |
+
"mean_token_accuracy": 0.8155439347028732,
|
| 519 |
+
"epoch": 0.9506398537477148,
|
| 520 |
+
"step": 520
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"loss": 0.7596,
|
| 524 |
+
"grad_norm": 0.8646839261054993,
|
| 525 |
+
"learning_rate": 0.0003077402589584061,
|
| 526 |
+
"entropy": 0.7170153304934501,
|
| 527 |
+
"num_tokens": 258959.0,
|
| 528 |
+
"mean_token_accuracy": 0.8573983281850814,
|
| 529 |
+
"epoch": 0.9689213893967094,
|
| 530 |
+
"step": 530
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"loss": 1.0173,
|
| 534 |
+
"grad_norm": 1.1162272691726685,
|
| 535 |
+
"learning_rate": 0.00029994626513710085,
|
| 536 |
+
"entropy": 0.956288554891944,
|
| 537 |
+
"num_tokens": 263941.0,
|
| 538 |
+
"mean_token_accuracy": 0.8194100826978683,
|
| 539 |
+
"epoch": 0.9872029250457038,
|
| 540 |
+
"step": 540
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"eval_loss": 0.8383127450942993,
|
| 544 |
+
"eval_runtime": 149.8997,
|
| 545 |
+
"eval_samples_per_second": 1.628,
|
| 546 |
+
"eval_steps_per_second": 1.628,
|
| 547 |
+
"eval_entropy": 0.7990527804024884,
|
| 548 |
+
"eval_num_tokens": 267313.0,
|
| 549 |
+
"eval_mean_token_accuracy": 0.8396265785225102,
|
| 550 |
+
"epoch": 1.0,
|
| 551 |
+
"step": 547
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"loss": 0.8334,
|
| 555 |
+
"grad_norm": 0.8393658995628357,
|
| 556 |
+
"learning_rate": 0.00029210136453009753,
|
| 557 |
+
"entropy": 0.8176384001970292,
|
| 558 |
+
"num_tokens": 268869.0,
|
| 559 |
+
"mean_token_accuracy": 0.8477066770195961,
|
| 560 |
+
"epoch": 1.0054844606946984,
|
| 561 |
+
"step": 550
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"loss": 0.6929,
|
| 565 |
+
"grad_norm": 1.470503568649292,
|
| 566 |
+
"learning_rate": 0.00028421355290390507,
|
| 567 |
+
"entropy": 0.601496777124703,
|
| 568 |
+
"num_tokens": 273700.0,
|
| 569 |
+
"mean_token_accuracy": 0.8613295748829841,
|
| 570 |
+
"epoch": 1.023765996343693,
|
| 571 |
+
"step": 560
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"loss": 0.583,
|
| 575 |
+
"grad_norm": 0.6013614535331726,
|
| 576 |
+
"learning_rate": 0.0002762908697612765,
|
| 577 |
+
"entropy": 0.5732227236032486,
|
| 578 |
+
"num_tokens": 278597.0,
|
| 579 |
+
"mean_token_accuracy": 0.8821299687027931,
|
| 580 |
+
"epoch": 1.0420475319926874,
|
| 581 |
+
"step": 570
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"loss": 0.7185,
|
| 585 |
+
"grad_norm": 1.5905576944351196,
|
| 586 |
+
"learning_rate": 0.0002683413901470984,
|
| 587 |
+
"entropy": 0.6473550666123629,
|
| 588 |
+
"num_tokens": 283394.0,
|
| 589 |
+
"mean_token_accuracy": 0.8632880464196205,
|
| 590 |
+
"epoch": 1.0603290676416819,
|
| 591 |
+
"step": 580
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"loss": 0.5452,
|
| 595 |
+
"grad_norm": 1.0172004699707031,
|
| 596 |
+
"learning_rate": 0.0002603732164180539,
|
| 597 |
+
"entropy": 0.5780091550201177,
|
| 598 |
+
"num_tokens": 288104.0,
|
| 599 |
+
"mean_token_accuracy": 0.8878793254494667,
|
| 600 |
+
"epoch": 1.0786106032906764,
|
| 601 |
+
"step": 590
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"loss": 0.545,
|
| 605 |
+
"grad_norm": 0.912875235080719,
|
| 606 |
+
"learning_rate": 0.000252394469984449,
|
| 607 |
+
"entropy": 0.47236425392329695,
|
| 608 |
+
"num_tokens": 292815.0,
|
| 609 |
+
"mean_token_accuracy": 0.8934838846325874,
|
| 610 |
+
"epoch": 1.0968921389396709,
|
| 611 |
+
"step": 600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"loss": 0.6716,
|
| 615 |
+
"grad_norm": 1.3273028135299683,
|
| 616 |
+
"learning_rate": 0.0002444132830326187,
|
| 617 |
+
"entropy": 0.6132044233381748,
|
| 618 |
+
"num_tokens": 297908.0,
|
| 619 |
+
"mean_token_accuracy": 0.8674399435520173,
|
| 620 |
+
"epoch": 1.1151736745886653,
|
| 621 |
+
"step": 610
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"loss": 0.6113,
|
| 625 |
+
"grad_norm": 1.288959264755249,
|
| 626 |
+
"learning_rate": 0.00023643779023634987,
|
| 627 |
+
"entropy": 0.5818431481719017,
|
| 628 |
+
"num_tokens": 302705.0,
|
| 629 |
+
"mean_token_accuracy": 0.8799894049763679,
|
| 630 |
+
"epoch": 1.13345521023766,
|
| 631 |
+
"step": 620
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"loss": 0.61,
|
| 635 |
+
"grad_norm": 0.9242374897003174,
|
| 636 |
+
"learning_rate": 0.0002284761204657696,
|
| 637 |
+
"entropy": 0.5832867488265038,
|
| 638 |
+
"num_tokens": 307457.0,
|
| 639 |
+
"mean_token_accuracy": 0.8836823180317879,
|
| 640 |
+
"epoch": 1.1517367458866545,
|
| 641 |
+
"step": 630
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"loss": 0.5441,
|
| 645 |
+
"grad_norm": 0.7086995840072632,
|
| 646 |
+
"learning_rate": 0.00022053638850214703,
|
| 647 |
+
"entropy": 0.5097997933626175,
|
| 648 |
+
"num_tokens": 312253.0,
|
| 649 |
+
"mean_token_accuracy": 0.894115723669529,
|
| 650 |
+
"epoch": 1.170018281535649,
|
| 651 |
+
"step": 640
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"loss": 0.5176,
|
| 655 |
+
"grad_norm": 1.3160265684127808,
|
| 656 |
+
"learning_rate": 0.00021262668676705695,
|
| 657 |
+
"entropy": 0.466005721129477,
|
| 658 |
+
"num_tokens": 316908.0,
|
| 659 |
+
"mean_token_accuracy": 0.8894715309143066,
|
| 660 |
+
"epoch": 1.1882998171846435,
|
| 661 |
+
"step": 650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"loss": 0.6213,
|
| 665 |
+
"grad_norm": 0.756281316280365,
|
| 666 |
+
"learning_rate": 0.00020475507707433177,
|
| 667 |
+
"entropy": 0.6145222604274749,
|
| 668 |
+
"num_tokens": 321671.0,
|
| 669 |
+
"mean_token_accuracy": 0.874506875872612,
|
| 670 |
+
"epoch": 1.206581352833638,
|
| 671 |
+
"step": 660
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"loss": 0.6087,
|
| 675 |
+
"grad_norm": 1.3001129627227783,
|
| 676 |
+
"learning_rate": 0.0001969295824132107,
|
| 677 |
+
"entropy": 0.5633690148591995,
|
| 678 |
+
"num_tokens": 326633.0,
|
| 679 |
+
"mean_token_accuracy": 0.884332463145256,
|
| 680 |
+
"epoch": 1.2248628884826325,
|
| 681 |
+
"step": 670
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"loss": 0.7631,
|
| 685 |
+
"grad_norm": 1.119253396987915,
|
| 686 |
+
"learning_rate": 0.00018915817877105927,
|
| 687 |
+
"entropy": 0.6663524687290192,
|
| 688 |
+
"num_tokens": 331740.0,
|
| 689 |
+
"mean_token_accuracy": 0.851144264638424,
|
| 690 |
+
"epoch": 1.2431444241316272,
|
| 691 |
+
"step": 680
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"loss": 0.5761,
|
| 695 |
+
"grad_norm": 0.6832223534584045,
|
| 696 |
+
"learning_rate": 0.00018144878700399546,
|
| 697 |
+
"entropy": 0.5777830693870782,
|
| 698 |
+
"num_tokens": 336945.0,
|
| 699 |
+
"mean_token_accuracy": 0.886911803483963,
|
| 700 |
+
"epoch": 1.2614259597806217,
|
| 701 |
+
"step": 690
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"loss": 0.6277,
|
| 705 |
+
"grad_norm": 0.8378806114196777,
|
| 706 |
+
"learning_rate": 0.000173809264763707,
|
| 707 |
+
"entropy": 0.5720893319696188,
|
| 708 |
+
"num_tokens": 341819.0,
|
| 709 |
+
"mean_token_accuracy": 0.8724584490060806,
|
| 710 |
+
"epoch": 1.2797074954296161,
|
| 711 |
+
"step": 700
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"loss": 0.6422,
|
| 715 |
+
"grad_norm": 1.3611880540847778,
|
| 716 |
+
"learning_rate": 0.00016624739848868853,
|
| 717 |
+
"entropy": 0.5773471333086491,
|
| 718 |
+
"num_tokens": 346672.0,
|
| 719 |
+
"mean_token_accuracy": 0.8712103441357613,
|
| 720 |
+
"epoch": 1.2979890310786106,
|
| 721 |
+
"step": 710
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"loss": 0.5019,
|
| 725 |
+
"grad_norm": 0.6666797399520874,
|
| 726 |
+
"learning_rate": 0.00015877089546806122,
|
| 727 |
+
"entropy": 0.5135227143764496,
|
| 728 |
+
"num_tokens": 351651.0,
|
| 729 |
+
"mean_token_accuracy": 0.8887041434645653,
|
| 730 |
+
"epoch": 1.3162705667276051,
|
| 731 |
+
"step": 720
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"loss": 0.5872,
|
| 735 |
+
"grad_norm": 0.8829151391983032,
|
| 736 |
+
"learning_rate": 0.00015138737598606447,
|
| 737 |
+
"entropy": 0.5499209146946669,
|
| 738 |
+
"num_tokens": 356663.0,
|
| 739 |
+
"mean_token_accuracy": 0.8893190011382103,
|
| 740 |
+
"epoch": 1.3345521023765996,
|
| 741 |
+
"step": 730
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"loss": 0.5956,
|
| 745 |
+
"grad_norm": 1.4903333187103271,
|
| 746 |
+
"learning_rate": 0.0001441043655552252,
|
| 747 |
+
"entropy": 0.5730689261108637,
|
| 748 |
+
"num_tokens": 361623.0,
|
| 749 |
+
"mean_token_accuracy": 0.8744840949773789,
|
| 750 |
+
"epoch": 1.352833638025594,
|
| 751 |
+
"step": 740
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"loss": 0.8425,
|
| 755 |
+
"grad_norm": 1.1822813749313354,
|
| 756 |
+
"learning_rate": 0.00013692928724612203,
|
| 757 |
+
"entropy": 0.7603746997192502,
|
| 758 |
+
"num_tokens": 366800.0,
|
| 759 |
+
"mean_token_accuracy": 0.8339792802929878,
|
| 760 |
+
"epoch": 1.3711151736745886,
|
| 761 |
+
"step": 750
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"loss": 0.5213,
|
| 765 |
+
"grad_norm": 1.104277491569519,
|
| 766 |
+
"learning_rate": 0.00012986945412156038,
|
| 767 |
+
"entropy": 0.5225177899003028,
|
| 768 |
+
"num_tokens": 371627.0,
|
| 769 |
+
"mean_token_accuracy": 0.8935010746121407,
|
| 770 |
+
"epoch": 1.389396709323583,
|
| 771 |
+
"step": 760
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"loss": 0.5954,
|
| 775 |
+
"grad_norm": 0.457518607378006,
|
| 776 |
+
"learning_rate": 0.00012293206178287184,
|
| 777 |
+
"entropy": 0.5619303321465849,
|
| 778 |
+
"num_tokens": 376616.0,
|
| 779 |
+
"mean_token_accuracy": 0.8800176531076431,
|
| 780 |
+
"epoch": 1.4076782449725778,
|
| 781 |
+
"step": 770
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"loss": 0.5962,
|
| 785 |
+
"grad_norm": 1.0970666408538818,
|
| 786 |
+
"learning_rate": 0.00011612418103593401,
|
| 787 |
+
"entropy": 0.5024011623114347,
|
| 788 |
+
"num_tokens": 381540.0,
|
| 789 |
+
"mean_token_accuracy": 0.88609908670187,
|
| 790 |
+
"epoch": 1.4259597806215722,
|
| 791 |
+
"step": 780
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"loss": 0.4754,
|
| 795 |
+
"grad_norm": 0.6814817190170288,
|
| 796 |
+
"learning_rate": 0.0001094527506843849,
|
| 797 |
+
"entropy": 0.4739612426608801,
|
| 798 |
+
"num_tokens": 386304.0,
|
| 799 |
+
"mean_token_accuracy": 0.8971261441707611,
|
| 800 |
+
"epoch": 1.4442413162705667,
|
| 801 |
+
"step": 790
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"loss": 0.5653,
|
| 805 |
+
"grad_norm": 1.4108277559280396,
|
| 806 |
+
"learning_rate": 0.00010292457045737893,
|
| 807 |
+
"entropy": 0.5391225486993789,
|
| 808 |
+
"num_tokens": 391148.0,
|
| 809 |
+
"mean_token_accuracy": 0.884946446120739,
|
| 810 |
+
"epoch": 1.4625228519195612,
|
| 811 |
+
"step": 800
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"loss": 0.5766,
|
| 815 |
+
"grad_norm": 0.701485812664032,
|
| 816 |
+
"learning_rate": 9.654629407909162e-05,
|
| 817 |
+
"entropy": 0.5549372952431441,
|
| 818 |
+
"num_tokens": 396181.0,
|
| 819 |
+
"mean_token_accuracy": 0.8827219530940056,
|
| 820 |
+
"epoch": 1.4808043875685557,
|
| 821 |
+
"step": 810
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"loss": 0.5183,
|
| 825 |
+
"grad_norm": 1.1837406158447266,
|
| 826 |
+
"learning_rate": 9.032442248703665e-05,
|
| 827 |
+
"entropy": 0.521338826790452,
|
| 828 |
+
"num_tokens": 401065.0,
|
| 829 |
+
"mean_token_accuracy": 0.8859279066324234,
|
| 830 |
+
"epoch": 1.4990859232175504,
|
| 831 |
+
"step": 820
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"loss": 0.4086,
|
| 835 |
+
"grad_norm": 0.6805317997932434,
|
| 836 |
+
"learning_rate": 8.426529720610934e-05,
|
| 837 |
+
"entropy": 0.40612360909581185,
|
| 838 |
+
"num_tokens": 405858.0,
|
| 839 |
+
"mean_token_accuracy": 0.9128091052174568,
|
| 840 |
+
"epoch": 1.517367458866545,
|
| 841 |
+
"step": 830
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"loss": 0.5668,
|
| 845 |
+
"grad_norm": 1.0712112188339233,
|
| 846 |
+
"learning_rate": 7.83750938851061e-05,
|
| 847 |
+
"entropy": 0.5340927181765437,
|
| 848 |
+
"num_tokens": 410763.0,
|
| 849 |
+
"mean_token_accuracy": 0.8831513956189155,
|
| 850 |
+
"epoch": 1.5356489945155394,
|
| 851 |
+
"step": 840
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"loss": 0.5611,
|
| 855 |
+
"grad_norm": 1.1453123092651367,
|
| 856 |
+
"learning_rate": 7.265981600231233e-05,
|
| 857 |
+
"entropy": 0.49723163414746524,
|
| 858 |
+
"num_tokens": 415814.0,
|
| 859 |
+
"mean_token_accuracy": 0.8880010843276978,
|
| 860 |
+
"epoch": 1.5539305301645339,
|
| 861 |
+
"step": 850
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"loss": 0.3383,
|
| 865 |
+
"grad_norm": 0.4844640791416168,
|
| 866 |
+
"learning_rate": 6.712528874657012e-05,
|
| 867 |
+
"entropy": 0.3304383518174291,
|
| 868 |
+
"num_tokens": 420421.0,
|
| 869 |
+
"mean_token_accuracy": 0.9282727792859078,
|
| 870 |
+
"epoch": 1.5722120658135283,
|
| 871 |
+
"step": 860
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"loss": 0.4785,
|
| 875 |
+
"grad_norm": 1.027254343032837,
|
| 876 |
+
"learning_rate": 6.177715308006504e-05,
|
| 877 |
+
"entropy": 0.46275698505342006,
|
| 878 |
+
"num_tokens": 425011.0,
|
| 879 |
+
"mean_token_accuracy": 0.9057738676667213,
|
| 880 |
+
"epoch": 1.5904936014625228,
|
| 881 |
+
"step": 870
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"loss": 0.5399,
|
| 885 |
+
"grad_norm": 0.9211871027946472,
|
| 886 |
+
"learning_rate": 5.662085998888214e-05,
|
| 887 |
+
"entropy": 0.5016709078103304,
|
| 888 |
+
"num_tokens": 429980.0,
|
| 889 |
+
"mean_token_accuracy": 0.8931746035814285,
|
| 890 |
+
"epoch": 1.6087751371115173,
|
| 891 |
+
"step": 880
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"loss": 0.623,
|
| 895 |
+
"grad_norm": 0.6902950406074524,
|
| 896 |
+
"learning_rate": 5.1661664927191235e-05,
|
| 897 |
+
"entropy": 0.566886336915195,
|
| 898 |
+
"num_tokens": 435007.0,
|
| 899 |
+
"mean_token_accuracy": 0.8787783816456795,
|
| 900 |
+
"epoch": 1.6270566727605118,
|
| 901 |
+
"step": 890
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"loss": 0.5005,
|
| 905 |
+
"grad_norm": 0.9160484075546265,
|
| 906 |
+
"learning_rate": 4.690462246072516e-05,
|
| 907 |
+
"entropy": 0.47696283366531134,
|
| 908 |
+
"num_tokens": 439747.0,
|
| 909 |
+
"mean_token_accuracy": 0.8926343008875847,
|
| 910 |
+
"epoch": 1.6453382084095063,
|
| 911 |
+
"step": 900
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"loss": 0.6681,
|
| 915 |
+
"grad_norm": 1.0038708448410034,
|
| 916 |
+
"learning_rate": 4.23545811150089e-05,
|
| 917 |
+
"entropy": 0.5537581698969006,
|
| 918 |
+
"num_tokens": 444847.0,
|
| 919 |
+
"mean_token_accuracy": 0.8806583434343338,
|
| 920 |
+
"epoch": 1.6636197440585008,
|
| 921 |
+
"step": 910
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"loss": 0.4931,
|
| 925 |
+
"grad_norm": 0.7535556554794312,
|
| 926 |
+
"learning_rate": 3.8016178433591866e-05,
|
| 927 |
+
"entropy": 0.4691365586593747,
|
| 928 |
+
"num_tokens": 449555.0,
|
| 929 |
+
"mean_token_accuracy": 0.901521198451519,
|
| 930 |
+
"epoch": 1.6819012797074955,
|
| 931 |
+
"step": 920
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"loss": 0.5885,
|
| 935 |
+
"grad_norm": 0.9373224973678589,
|
| 936 |
+
"learning_rate": 3.389383625131942e-05,
|
| 937 |
+
"entropy": 0.5948667161166667,
|
| 938 |
+
"num_tokens": 454448.0,
|
| 939 |
+
"mean_token_accuracy": 0.8759057343006134,
|
| 940 |
+
"epoch": 1.70018281535649,
|
| 941 |
+
"step": 930
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"loss": 0.6568,
|
| 945 |
+
"grad_norm": 1.0174107551574707,
|
| 946 |
+
"learning_rate": 2.9991756187461e-05,
|
| 947 |
+
"entropy": 0.5920586219057441,
|
| 948 |
+
"num_tokens": 459535.0,
|
| 949 |
+
"mean_token_accuracy": 0.8757189273834228,
|
| 950 |
+
"epoch": 1.7184643510054844,
|
| 951 |
+
"step": 940
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"loss": 0.486,
|
| 955 |
+
"grad_norm": 0.9126308560371399,
|
| 956 |
+
"learning_rate": 2.631391536328992e-05,
|
| 957 |
+
"entropy": 0.4687887070700526,
|
| 958 |
+
"num_tokens": 464202.0,
|
| 959 |
+
"mean_token_accuracy": 0.9010345816612244,
|
| 960 |
+
"epoch": 1.736745886654479,
|
| 961 |
+
"step": 950
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"loss": 0.4731,
|
| 965 |
+
"grad_norm": 0.9621142148971558,
|
| 966 |
+
"learning_rate": 2.2864062348476906e-05,
|
| 967 |
+
"entropy": 0.47076144628226757,
|
| 968 |
+
"num_tokens": 469164.0,
|
| 969 |
+
"mean_token_accuracy": 0.899216802418232,
|
| 970 |
+
"epoch": 1.7550274223034736,
|
| 971 |
+
"step": 960
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"loss": 0.5731,
|
| 975 |
+
"grad_norm": 1.0501958131790161,
|
| 976 |
+
"learning_rate": 1.9645713340431997e-05,
|
| 977 |
+
"entropy": 0.5156827673316002,
|
| 978 |
+
"num_tokens": 474032.0,
|
| 979 |
+
"mean_token_accuracy": 0.8930515125393867,
|
| 980 |
+
"epoch": 1.7733089579524681,
|
| 981 |
+
"step": 970
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"loss": 0.5638,
|
| 985 |
+
"grad_norm": 1.393446922302246,
|
| 986 |
+
"learning_rate": 1.6662148580486704e-05,
|
| 987 |
+
"entropy": 0.5192593608051539,
|
| 988 |
+
"num_tokens": 479058.0,
|
| 989 |
+
"mean_token_accuracy": 0.8873833552002907,
|
| 990 |
+
"epoch": 1.7915904936014626,
|
| 991 |
+
"step": 980
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"loss": 0.5387,
|
| 995 |
+
"grad_norm": 1.067104697227478,
|
| 996 |
+
"learning_rate": 1.3916409010569925e-05,
|
| 997 |
+
"entropy": 0.4929318675771356,
|
| 998 |
+
"num_tokens": 483717.0,
|
| 999 |
+
"mean_token_accuracy": 0.8899582773447037,
|
| 1000 |
+
"epoch": 1.809872029250457,
|
| 1001 |
+
"step": 990
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"loss": 0.474,
|
| 1005 |
+
"grad_norm": 0.7692973017692566,
|
| 1006 |
+
"learning_rate": 1.1411293173785726e-05,
|
| 1007 |
+
"entropy": 0.4536196293309331,
|
| 1008 |
+
"num_tokens": 488517.0,
|
| 1009 |
+
"mean_token_accuracy": 0.9024506390094758,
|
| 1010 |
+
"epoch": 1.8281535648994516,
|
| 1011 |
+
"step": 1000
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"loss": 0.5862,
|
| 1015 |
+
"grad_norm": 0.9886012077331543,
|
| 1016 |
+
"learning_rate": 9.149354362050804e-06,
|
| 1017 |
+
"entropy": 0.5348154939711094,
|
| 1018 |
+
"num_tokens": 493643.0,
|
| 1019 |
+
"mean_token_accuracy": 0.887369342148304,
|
| 1020 |
+
"epoch": 1.846435100548446,
|
| 1021 |
+
"step": 1010
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"loss": 0.4364,
|
| 1025 |
+
"grad_norm": 0.6033039689064026,
|
| 1026 |
+
"learning_rate": 7.13289801370054e-06,
|
| 1027 |
+
"entropy": 0.4438053784891963,
|
| 1028 |
+
"num_tokens": 498402.0,
|
| 1029 |
+
"mean_token_accuracy": 0.9055494770407677,
|
| 1030 |
+
"epoch": 1.8647166361974405,
|
| 1031 |
+
"step": 1020
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"loss": 0.4822,
|
| 1035 |
+
"grad_norm": 1.0783859491348267,
|
| 1036 |
+
"learning_rate": 5.363979363714244e-06,
|
| 1037 |
+
"entropy": 0.44634236562997104,
|
| 1038 |
+
"num_tokens": 503273.0,
|
| 1039 |
+
"mean_token_accuracy": 0.9009472608566285,
|
| 1040 |
+
"epoch": 1.882998171846435,
|
| 1041 |
+
"step": 1030
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"loss": 0.464,
|
| 1045 |
+
"grad_norm": 1.3200268745422363,
|
| 1046 |
+
"learning_rate": 3.844401348955834e-06,
|
| 1047 |
+
"entropy": 0.44374721236526965,
|
| 1048 |
+
"num_tokens": 508179.0,
|
| 1049 |
+
"mean_token_accuracy": 0.9102602645754814,
|
| 1050 |
+
"epoch": 1.9012797074954295,
|
| 1051 |
+
"step": 1040
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"loss": 0.5956,
|
| 1055 |
+
"grad_norm": 0.8053908348083496,
|
| 1056 |
+
"learning_rate": 2.575712770564592e-06,
|
| 1057 |
+
"entropy": 0.567001193203032,
|
| 1058 |
+
"num_tokens": 513289.0,
|
| 1059 |
+
"mean_token_accuracy": 0.8815849825739861,
|
| 1060 |
+
"epoch": 1.919561243144424,
|
| 1061 |
+
"step": 1050
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"loss": 0.6047,
|
| 1065 |
+
"grad_norm": 1.3640791177749634,
|
| 1066 |
+
"learning_rate": 1.559206715368966e-06,
|
| 1067 |
+
"entropy": 0.609914462082088,
|
| 1068 |
+
"num_tokens": 518359.0,
|
| 1069 |
+
"mean_token_accuracy": 0.8840003907680511,
|
| 1070 |
+
"epoch": 1.9378427787934185,
|
| 1071 |
+
"step": 1060
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"loss": 0.4451,
|
| 1075 |
+
"grad_norm": 1.2007508277893066,
|
| 1076 |
+
"learning_rate": 7.959192379322078e-07,
|
| 1077 |
+
"entropy": 0.43388587329536676,
|
| 1078 |
+
"num_tokens": 523015.0,
|
| 1079 |
+
"mean_token_accuracy": 0.9105660766363144,
|
| 1080 |
+
"epoch": 1.9561243144424132,
|
| 1081 |
+
"step": 1070
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"loss": 0.4733,
|
| 1085 |
+
"grad_norm": 0.7846450209617615,
|
| 1086 |
+
"learning_rate": 2.8662830457340526e-07,
|
| 1087 |
+
"entropy": 0.46208970360457896,
|
| 1088 |
+
"num_tokens": 527800.0,
|
| 1089 |
+
"mean_token_accuracy": 0.9034222677350044,
|
| 1090 |
+
"epoch": 1.9744058500914077,
|
| 1091 |
+
"step": 1080
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"loss": 0.5183,
|
| 1095 |
+
"grad_norm": 0.9528534412384033,
|
| 1096 |
+
"learning_rate": 3.1853000439951984e-08,
|
| 1097 |
+
"entropy": 0.47328052408993243,
|
| 1098 |
+
"num_tokens": 532702.0,
|
| 1099 |
+
"mean_token_accuracy": 0.8960653573274613,
|
| 1100 |
+
"epoch": 1.9926873857404022,
|
| 1101 |
+
"step": 1090
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"eval_loss": 0.7137235999107361,
|
| 1105 |
+
"eval_runtime": 149.8883,
|
| 1106 |
+
"eval_samples_per_second": 1.628,
|
| 1107 |
+
"eval_steps_per_second": 1.628,
|
| 1108 |
+
"eval_entropy": 0.5778556458163456,
|
| 1109 |
+
"eval_num_tokens": 534626.0,
|
| 1110 |
+
"eval_mean_token_accuracy": 0.8677698337152356,
|
| 1111 |
+
"epoch": 2.0,
|
| 1112 |
+
"step": 1094
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"train_runtime": 6289.0757,
|
| 1116 |
+
"train_samples_per_second": 0.696,
|
| 1117 |
+
"train_steps_per_second": 0.174,
|
| 1118 |
+
"total_flos": 1.2079230284163072e+16,
|
| 1119 |
+
"train_loss": 0.8425550589395834,
|
| 1120 |
+
"epoch": 2.0,
|
| 1121 |
+
"step": 1094
|
| 1122 |
+
}
|
| 1123 |
+
]
|
README.md
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
---
|
| 2 |
base_model: Qwen/Qwen3-4B
|
| 3 |
library_name: peft
|
| 4 |
-
model_name:
|
| 5 |
tags:
|
| 6 |
- base_model:adapter:Qwen/Qwen3-4B
|
| 7 |
- lora
|
|
@@ -12,7 +12,7 @@ licence: license
|
|
| 12 |
pipeline_tag: text-generation
|
| 13 |
---
|
| 14 |
|
| 15 |
-
# Model Card for
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [Qwen/Qwen3-4B](https://huggingface.co/Qwen/Qwen3-4B).
|
| 18 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
|
@@ -38,8 +38,8 @@ This model was trained with SFT.
|
|
| 38 |
### Framework versions
|
| 39 |
|
| 40 |
- PEFT 0.18.0
|
| 41 |
-
- TRL: 0.
|
| 42 |
-
- Transformers: 4.57.
|
| 43 |
- Pytorch: 2.9.0+cu126
|
| 44 |
- Datasets: 4.0.0
|
| 45 |
- Tokenizers: 0.22.1
|
|
|
|
| 1 |
---
|
| 2 |
base_model: Qwen/Qwen3-4B
|
| 3 |
library_name: peft
|
| 4 |
+
model_name: CryptoStatuette-qwen-finetuned
|
| 5 |
tags:
|
| 6 |
- base_model:adapter:Qwen/Qwen3-4B
|
| 7 |
- lora
|
|
|
|
| 12 |
pipeline_tag: text-generation
|
| 13 |
---
|
| 14 |
|
| 15 |
+
# Model Card for CryptoStatuette-qwen-finetuned
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [Qwen/Qwen3-4B](https://huggingface.co/Qwen/Qwen3-4B).
|
| 18 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
|
|
|
| 38 |
### Framework versions
|
| 39 |
|
| 40 |
- PEFT 0.18.0
|
| 41 |
+
- TRL: 0.25.1
|
| 42 |
+
- Transformers: 4.57.2
|
| 43 |
- Pytorch: 2.9.0+cu126
|
| 44 |
- Datasets: 4.0.0
|
| 45 |
- Tokenizers: 0.22.1
|