End of training
Browse files
all_results.json
CHANGED
|
@@ -1,5 +1,10 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
"total_flos": 9.830314275137126e+17,
|
| 4 |
"train_loss": 1.5905956603586673,
|
| 5 |
"train_runtime": 1497.0202,
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 2.595360517501831,
|
| 4 |
+
"eval_runtime": 0.8154,
|
| 5 |
+
"eval_samples": 15,
|
| 6 |
+
"eval_samples_per_second": 7.358,
|
| 7 |
+
"eval_steps_per_second": 1.226,
|
| 8 |
"total_flos": 9.830314275137126e+17,
|
| 9 |
"train_loss": 1.5905956603586673,
|
| 10 |
"train_runtime": 1497.0202,
|
config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_attn_implementation_autoset": true,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 128256,
|
| 9 |
+
"eos_token_id": 128257,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 128257,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"quantization_config": {
|
| 24 |
+
"_load_in_4bit": true,
|
| 25 |
+
"_load_in_8bit": false,
|
| 26 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
| 27 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 28 |
+
"bnb_4bit_quant_type": "nf4",
|
| 29 |
+
"bnb_4bit_use_double_quant": false,
|
| 30 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 31 |
+
"llm_int8_has_fp16_weight": false,
|
| 32 |
+
"llm_int8_skip_modules": null,
|
| 33 |
+
"llm_int8_threshold": 6.0,
|
| 34 |
+
"load_in_4bit": true,
|
| 35 |
+
"load_in_8bit": false,
|
| 36 |
+
"quant_method": "bitsandbytes"
|
| 37 |
+
},
|
| 38 |
+
"rms_norm_eps": 1e-05,
|
| 39 |
+
"rope_scaling": {
|
| 40 |
+
"factor": 32.0,
|
| 41 |
+
"high_freq_factor": 4.0,
|
| 42 |
+
"low_freq_factor": 1.0,
|
| 43 |
+
"original_max_position_embeddings": 8192,
|
| 44 |
+
"rope_type": "llama3"
|
| 45 |
+
},
|
| 46 |
+
"rope_theta": 500000.0,
|
| 47 |
+
"tie_word_embeddings": true,
|
| 48 |
+
"torch_dtype": "bfloat16",
|
| 49 |
+
"transformers_version": "4.50.3",
|
| 50 |
+
"use_cache": true,
|
| 51 |
+
"vocab_size": 128258
|
| 52 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 2.595360517501831,
|
| 4 |
+
"eval_runtime": 0.8154,
|
| 5 |
+
"eval_samples": 15,
|
| 6 |
+
"eval_samples_per_second": 7.358,
|
| 7 |
+
"eval_steps_per_second": 1.226
|
| 8 |
+
}
|
runs/Mar31_12-46-20_9ec6cd803560/events.out.tfevents.1743441273.9ec6cd803560.23533.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c4a045897c4dc269bc9b1f972fcdddd8b6d7c71d366e47332ed2a57eb6175a3
|
| 3 |
+
size 359
|