Training in progress, step 2500, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 9446744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea03fbd5faff9829b79932a9492534fbbbe2845de9ce69e896c0b8d109c1a825
|
| 3 |
size 9446744
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4879947
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8ca4bfe766f5a9ce1a39e0d776749658d826fc560902b47178ff40c41d18b94a
|
| 3 |
size 4879947
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd840ceb0cbd2bc41560fadd05ab11cb9d3690eebf99ba42e453854e5f372ed8
|
| 3 |
size 14917
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e68bcbbf919727508b1f5613e7b10a32a3e07fdef6c3370ef48c8724f2e31e4
|
| 3 |
size 14917
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:861ce13e6ca091acee9a68ebfc5ca38479baf4b537c37b3949f071f77b81e9f0
|
| 3 |
size 1383
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f5bce3ff1203929d6808ab229d6e6d4d185a3da8ef87a3b682b0eec04e6bacf2
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -2008,6 +2008,506 @@
|
|
| 2008 |
"mean_token_accuracy": 0.6531724959611893,
|
| 2009 |
"num_tokens": 11896615.0,
|
| 2010 |
"step": 2000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2011 |
}
|
| 2012 |
],
|
| 2013 |
"logging_steps": 10,
|
|
@@ -2027,7 +2527,7 @@
|
|
| 2027 |
"attributes": {}
|
| 2028 |
}
|
| 2029 |
},
|
| 2030 |
-
"total_flos":
|
| 2031 |
"train_batch_size": 2,
|
| 2032 |
"trial_name": null,
|
| 2033 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 4.295723189340211,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 2500,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 2008 |
"mean_token_accuracy": 0.6531724959611893,
|
| 2009 |
"num_tokens": 11896615.0,
|
| 2010 |
"step": 2000
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"entropy": 1.8952298507094383,
|
| 2014 |
+
"epoch": 3.453900709219858,
|
| 2015 |
+
"grad_norm": 0.7292787432670593,
|
| 2016 |
+
"learning_rate": 0.00011964,
|
| 2017 |
+
"loss": 1.9302806854248047,
|
| 2018 |
+
"mean_token_accuracy": 0.6462091594934464,
|
| 2019 |
+
"num_tokens": 11954949.0,
|
| 2020 |
+
"step": 2010
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"entropy": 1.8723753660917282,
|
| 2024 |
+
"epoch": 3.4710939179024285,
|
| 2025 |
+
"grad_norm": 0.730530858039856,
|
| 2026 |
+
"learning_rate": 0.00011923999999999999,
|
| 2027 |
+
"loss": 1.9216194152832031,
|
| 2028 |
+
"mean_token_accuracy": 0.6504904717206955,
|
| 2029 |
+
"num_tokens": 12013803.0,
|
| 2030 |
+
"step": 2020
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"entropy": 1.8673277243971824,
|
| 2034 |
+
"epoch": 3.488287126584999,
|
| 2035 |
+
"grad_norm": 0.7530126571655273,
|
| 2036 |
+
"learning_rate": 0.00011884,
|
| 2037 |
+
"loss": 1.968985366821289,
|
| 2038 |
+
"mean_token_accuracy": 0.646847129613161,
|
| 2039 |
+
"num_tokens": 12073284.0,
|
| 2040 |
+
"step": 2030
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"entropy": 1.8757897645235062,
|
| 2044 |
+
"epoch": 3.505480335267569,
|
| 2045 |
+
"grad_norm": 0.7031217813491821,
|
| 2046 |
+
"learning_rate": 0.00011844,
|
| 2047 |
+
"loss": 1.9071741104125977,
|
| 2048 |
+
"mean_token_accuracy": 0.6450003884732723,
|
| 2049 |
+
"num_tokens": 12126451.0,
|
| 2050 |
+
"step": 2040
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"entropy": 1.7986262783408165,
|
| 2054 |
+
"epoch": 3.5226735439501398,
|
| 2055 |
+
"grad_norm": 0.7223983407020569,
|
| 2056 |
+
"learning_rate": 0.00011804,
|
| 2057 |
+
"loss": 1.8450950622558593,
|
| 2058 |
+
"mean_token_accuracy": 0.6576410517096519,
|
| 2059 |
+
"num_tokens": 12183343.0,
|
| 2060 |
+
"step": 2050
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"entropy": 1.8884935915470122,
|
| 2064 |
+
"epoch": 3.53986675263271,
|
| 2065 |
+
"grad_norm": 0.7206518650054932,
|
| 2066 |
+
"learning_rate": 0.00011763999999999999,
|
| 2067 |
+
"loss": 1.9660964965820313,
|
| 2068 |
+
"mean_token_accuracy": 0.6422303304076195,
|
| 2069 |
+
"num_tokens": 12243607.0,
|
| 2070 |
+
"step": 2060
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"entropy": 1.8009026944637299,
|
| 2074 |
+
"epoch": 3.5570599613152805,
|
| 2075 |
+
"grad_norm": 0.7229637503623962,
|
| 2076 |
+
"learning_rate": 0.00011724000000000002,
|
| 2077 |
+
"loss": 1.851433563232422,
|
| 2078 |
+
"mean_token_accuracy": 0.6556052915751934,
|
| 2079 |
+
"num_tokens": 12304867.0,
|
| 2080 |
+
"step": 2070
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"entropy": 1.7949693977832795,
|
| 2084 |
+
"epoch": 3.574253169997851,
|
| 2085 |
+
"grad_norm": 0.6935518383979797,
|
| 2086 |
+
"learning_rate": 0.00011684000000000001,
|
| 2087 |
+
"loss": 1.8848058700561523,
|
| 2088 |
+
"mean_token_accuracy": 0.6580755174160003,
|
| 2089 |
+
"num_tokens": 12367633.0,
|
| 2090 |
+
"step": 2080
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"entropy": 1.8038981169462205,
|
| 2094 |
+
"epoch": 3.591446378680421,
|
| 2095 |
+
"grad_norm": 0.7003904581069946,
|
| 2096 |
+
"learning_rate": 0.00011644000000000002,
|
| 2097 |
+
"loss": 1.8867233276367188,
|
| 2098 |
+
"mean_token_accuracy": 0.655081395432353,
|
| 2099 |
+
"num_tokens": 12423928.0,
|
| 2100 |
+
"step": 2090
|
| 2101 |
+
},
|
| 2102 |
+
{
|
| 2103 |
+
"entropy": 1.850062020123005,
|
| 2104 |
+
"epoch": 3.6086395873629917,
|
| 2105 |
+
"grad_norm": 0.6852926015853882,
|
| 2106 |
+
"learning_rate": 0.00011604000000000002,
|
| 2107 |
+
"loss": 1.9325201034545898,
|
| 2108 |
+
"mean_token_accuracy": 0.6472255479544401,
|
| 2109 |
+
"num_tokens": 12479411.0,
|
| 2110 |
+
"step": 2100
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"entropy": 1.8294448778033257,
|
| 2114 |
+
"epoch": 3.625832796045562,
|
| 2115 |
+
"grad_norm": 0.7044693827629089,
|
| 2116 |
+
"learning_rate": 0.00011564000000000001,
|
| 2117 |
+
"loss": 1.8989273071289063,
|
| 2118 |
+
"mean_token_accuracy": 0.6499249216169118,
|
| 2119 |
+
"num_tokens": 12539175.0,
|
| 2120 |
+
"step": 2110
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"entropy": 1.8719267755746842,
|
| 2124 |
+
"epoch": 3.6430260047281324,
|
| 2125 |
+
"grad_norm": 0.7180586457252502,
|
| 2126 |
+
"learning_rate": 0.00011524000000000001,
|
| 2127 |
+
"loss": 1.925216293334961,
|
| 2128 |
+
"mean_token_accuracy": 0.648950444161892,
|
| 2129 |
+
"num_tokens": 12598337.0,
|
| 2130 |
+
"step": 2120
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"entropy": 1.88923449665308,
|
| 2134 |
+
"epoch": 3.660219213410703,
|
| 2135 |
+
"grad_norm": 0.7464597821235657,
|
| 2136 |
+
"learning_rate": 0.00011484000000000002,
|
| 2137 |
+
"loss": 1.990826416015625,
|
| 2138 |
+
"mean_token_accuracy": 0.6456409864127636,
|
| 2139 |
+
"num_tokens": 12656592.0,
|
| 2140 |
+
"step": 2130
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"entropy": 1.8126205861568452,
|
| 2144 |
+
"epoch": 3.677412422093273,
|
| 2145 |
+
"grad_norm": 0.7253774404525757,
|
| 2146 |
+
"learning_rate": 0.00011444000000000001,
|
| 2147 |
+
"loss": 1.9414216995239257,
|
| 2148 |
+
"mean_token_accuracy": 0.6552157323807478,
|
| 2149 |
+
"num_tokens": 12717791.0,
|
| 2150 |
+
"step": 2140
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"entropy": 1.8930377542972565,
|
| 2154 |
+
"epoch": 3.6946056307758437,
|
| 2155 |
+
"grad_norm": 0.7404170036315918,
|
| 2156 |
+
"learning_rate": 0.00011404000000000001,
|
| 2157 |
+
"loss": 1.9364784240722657,
|
| 2158 |
+
"mean_token_accuracy": 0.6434980578720569,
|
| 2159 |
+
"num_tokens": 12775445.0,
|
| 2160 |
+
"step": 2150
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"entropy": 1.7652419656515121,
|
| 2164 |
+
"epoch": 3.711798839458414,
|
| 2165 |
+
"grad_norm": 0.688732385635376,
|
| 2166 |
+
"learning_rate": 0.00011364000000000002,
|
| 2167 |
+
"loss": 1.7636165618896484,
|
| 2168 |
+
"mean_token_accuracy": 0.6639453627169132,
|
| 2169 |
+
"num_tokens": 12834599.0,
|
| 2170 |
+
"step": 2160
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"entropy": 1.7745767116546631,
|
| 2174 |
+
"epoch": 3.7289920481409844,
|
| 2175 |
+
"grad_norm": 0.7011992335319519,
|
| 2176 |
+
"learning_rate": 0.00011324000000000001,
|
| 2177 |
+
"loss": 1.8347841262817384,
|
| 2178 |
+
"mean_token_accuracy": 0.6586773280054331,
|
| 2179 |
+
"num_tokens": 12889887.0,
|
| 2180 |
+
"step": 2170
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"entropy": 1.7952505484223367,
|
| 2184 |
+
"epoch": 3.746185256823555,
|
| 2185 |
+
"grad_norm": 0.7646785378456116,
|
| 2186 |
+
"learning_rate": 0.00011284000000000001,
|
| 2187 |
+
"loss": 1.883163070678711,
|
| 2188 |
+
"mean_token_accuracy": 0.6589437790215016,
|
| 2189 |
+
"num_tokens": 12950286.0,
|
| 2190 |
+
"step": 2180
|
| 2191 |
+
},
|
| 2192 |
+
{
|
| 2193 |
+
"entropy": 1.8878965258598328,
|
| 2194 |
+
"epoch": 3.763378465506125,
|
| 2195 |
+
"grad_norm": 0.7722623944282532,
|
| 2196 |
+
"learning_rate": 0.00011244000000000001,
|
| 2197 |
+
"loss": 1.9674694061279296,
|
| 2198 |
+
"mean_token_accuracy": 0.6422343414276839,
|
| 2199 |
+
"num_tokens": 13011083.0,
|
| 2200 |
+
"step": 2190
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"entropy": 1.919720321893692,
|
| 2204 |
+
"epoch": 3.780571674188695,
|
| 2205 |
+
"grad_norm": 0.7656893134117126,
|
| 2206 |
+
"learning_rate": 0.00011204000000000002,
|
| 2207 |
+
"loss": 1.9919773101806642,
|
| 2208 |
+
"mean_token_accuracy": 0.6393908958882093,
|
| 2209 |
+
"num_tokens": 13069376.0,
|
| 2210 |
+
"step": 2200
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"entropy": 1.77825688123703,
|
| 2214 |
+
"epoch": 3.797764882871266,
|
| 2215 |
+
"grad_norm": 0.8324808478355408,
|
| 2216 |
+
"learning_rate": 0.00011164000000000001,
|
| 2217 |
+
"loss": 1.8173160552978516,
|
| 2218 |
+
"mean_token_accuracy": 0.659475727379322,
|
| 2219 |
+
"num_tokens": 13124851.0,
|
| 2220 |
+
"step": 2210
|
| 2221 |
+
},
|
| 2222 |
+
{
|
| 2223 |
+
"entropy": 1.8232837438583374,
|
| 2224 |
+
"epoch": 3.8149580915538364,
|
| 2225 |
+
"grad_norm": 0.741481363773346,
|
| 2226 |
+
"learning_rate": 0.00011124000000000001,
|
| 2227 |
+
"loss": 1.860748291015625,
|
| 2228 |
+
"mean_token_accuracy": 0.6524971850216389,
|
| 2229 |
+
"num_tokens": 13182576.0,
|
| 2230 |
+
"step": 2220
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"entropy": 1.8588940657675266,
|
| 2234 |
+
"epoch": 3.8321513002364065,
|
| 2235 |
+
"grad_norm": 0.7748705148696899,
|
| 2236 |
+
"learning_rate": 0.00011084000000000002,
|
| 2237 |
+
"loss": 1.9206954956054687,
|
| 2238 |
+
"mean_token_accuracy": 0.6516353718936443,
|
| 2239 |
+
"num_tokens": 13242703.0,
|
| 2240 |
+
"step": 2230
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"entropy": 1.823398308455944,
|
| 2244 |
+
"epoch": 3.849344508918977,
|
| 2245 |
+
"grad_norm": 0.6341049671173096,
|
| 2246 |
+
"learning_rate": 0.00011044,
|
| 2247 |
+
"loss": 1.8718917846679688,
|
| 2248 |
+
"mean_token_accuracy": 0.6596139155328273,
|
| 2249 |
+
"num_tokens": 13303181.0,
|
| 2250 |
+
"step": 2240
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"entropy": 1.8098929420113563,
|
| 2254 |
+
"epoch": 3.866537717601547,
|
| 2255 |
+
"grad_norm": 0.6672969460487366,
|
| 2256 |
+
"learning_rate": 0.00011004000000000001,
|
| 2257 |
+
"loss": 1.8999752044677733,
|
| 2258 |
+
"mean_token_accuracy": 0.6594760783016682,
|
| 2259 |
+
"num_tokens": 13364371.0,
|
| 2260 |
+
"step": 2250
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"entropy": 1.7795367375016213,
|
| 2264 |
+
"epoch": 3.8837309262841178,
|
| 2265 |
+
"grad_norm": 0.6343891024589539,
|
| 2266 |
+
"learning_rate": 0.00010964000000000001,
|
| 2267 |
+
"loss": 1.827276611328125,
|
| 2268 |
+
"mean_token_accuracy": 0.6668465688824654,
|
| 2269 |
+
"num_tokens": 13425450.0,
|
| 2270 |
+
"step": 2260
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"entropy": 1.8673226684331894,
|
| 2274 |
+
"epoch": 3.9009241349666883,
|
| 2275 |
+
"grad_norm": 0.7357877492904663,
|
| 2276 |
+
"learning_rate": 0.00010924,
|
| 2277 |
+
"loss": 1.9206443786621095,
|
| 2278 |
+
"mean_token_accuracy": 0.647479448094964,
|
| 2279 |
+
"num_tokens": 13485806.0,
|
| 2280 |
+
"step": 2270
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"entropy": 1.806484942883253,
|
| 2284 |
+
"epoch": 3.9181173436492585,
|
| 2285 |
+
"grad_norm": 0.7172144055366516,
|
| 2286 |
+
"learning_rate": 0.00010884000000000001,
|
| 2287 |
+
"loss": 1.8789045333862304,
|
| 2288 |
+
"mean_token_accuracy": 0.6594084780663252,
|
| 2289 |
+
"num_tokens": 13544934.0,
|
| 2290 |
+
"step": 2280
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"entropy": 1.7970930591225625,
|
| 2294 |
+
"epoch": 3.935310552331829,
|
| 2295 |
+
"grad_norm": 0.7578801512718201,
|
| 2296 |
+
"learning_rate": 0.00010844000000000001,
|
| 2297 |
+
"loss": 1.8405040740966796,
|
| 2298 |
+
"mean_token_accuracy": 0.6608923889696598,
|
| 2299 |
+
"num_tokens": 13606653.0,
|
| 2300 |
+
"step": 2290
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"entropy": 1.8469372361898422,
|
| 2304 |
+
"epoch": 3.952503761014399,
|
| 2305 |
+
"grad_norm": 0.7626324892044067,
|
| 2306 |
+
"learning_rate": 0.00010804,
|
| 2307 |
+
"loss": 1.8629837036132812,
|
| 2308 |
+
"mean_token_accuracy": 0.6560039456933737,
|
| 2309 |
+
"num_tokens": 13663938.0,
|
| 2310 |
+
"step": 2300
|
| 2311 |
+
},
|
| 2312 |
+
{
|
| 2313 |
+
"entropy": 1.836122378706932,
|
| 2314 |
+
"epoch": 3.9696969696969697,
|
| 2315 |
+
"grad_norm": 0.7074365615844727,
|
| 2316 |
+
"learning_rate": 0.00010764,
|
| 2317 |
+
"loss": 1.8942070007324219,
|
| 2318 |
+
"mean_token_accuracy": 0.647238065674901,
|
| 2319 |
+
"num_tokens": 13722549.0,
|
| 2320 |
+
"step": 2310
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"entropy": 1.821449062973261,
|
| 2324 |
+
"epoch": 3.9868901783795403,
|
| 2325 |
+
"grad_norm": 0.6956577301025391,
|
| 2326 |
+
"learning_rate": 0.00010724000000000001,
|
| 2327 |
+
"loss": 1.8947336196899414,
|
| 2328 |
+
"mean_token_accuracy": 0.6528103355318308,
|
| 2329 |
+
"num_tokens": 13785922.0,
|
| 2330 |
+
"step": 2320
|
| 2331 |
+
},
|
| 2332 |
+
{
|
| 2333 |
+
"entropy": 1.839719023023333,
|
| 2334 |
+
"epoch": 4.003438641736514,
|
| 2335 |
+
"grad_norm": 0.6865222454071045,
|
| 2336 |
+
"learning_rate": 0.00010684,
|
| 2337 |
+
"loss": 1.8803377151489258,
|
| 2338 |
+
"mean_token_accuracy": 0.6526942384707464,
|
| 2339 |
+
"num_tokens": 13844647.0,
|
| 2340 |
+
"step": 2330
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"entropy": 1.855065654218197,
|
| 2344 |
+
"epoch": 4.020631850419084,
|
| 2345 |
+
"grad_norm": 0.7424384355545044,
|
| 2346 |
+
"learning_rate": 0.00010644,
|
| 2347 |
+
"loss": 1.9461166381835937,
|
| 2348 |
+
"mean_token_accuracy": 0.6463506512343884,
|
| 2349 |
+
"num_tokens": 13904724.0,
|
| 2350 |
+
"step": 2340
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"entropy": 1.7508789122104644,
|
| 2354 |
+
"epoch": 4.037825059101655,
|
| 2355 |
+
"grad_norm": 0.6670609712600708,
|
| 2356 |
+
"learning_rate": 0.00010604000000000001,
|
| 2357 |
+
"loss": 1.781893539428711,
|
| 2358 |
+
"mean_token_accuracy": 0.6653038747608662,
|
| 2359 |
+
"num_tokens": 13963472.0,
|
| 2360 |
+
"step": 2350
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"entropy": 1.8165026590228082,
|
| 2364 |
+
"epoch": 4.0550182677842255,
|
| 2365 |
+
"grad_norm": 0.7823750376701355,
|
| 2366 |
+
"learning_rate": 0.00010564000000000001,
|
| 2367 |
+
"loss": 1.8847312927246094,
|
| 2368 |
+
"mean_token_accuracy": 0.6607359856367111,
|
| 2369 |
+
"num_tokens": 14019708.0,
|
| 2370 |
+
"step": 2360
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"entropy": 1.794335062801838,
|
| 2374 |
+
"epoch": 4.072211476466796,
|
| 2375 |
+
"grad_norm": 0.8262340426445007,
|
| 2376 |
+
"learning_rate": 0.00010524,
|
| 2377 |
+
"loss": 1.8576740264892577,
|
| 2378 |
+
"mean_token_accuracy": 0.6582343481481076,
|
| 2379 |
+
"num_tokens": 14076178.0,
|
| 2380 |
+
"step": 2370
|
| 2381 |
+
},
|
| 2382 |
+
{
|
| 2383 |
+
"entropy": 1.8828865155577659,
|
| 2384 |
+
"epoch": 4.089404685149366,
|
| 2385 |
+
"grad_norm": 0.784656822681427,
|
| 2386 |
+
"learning_rate": 0.00010484,
|
| 2387 |
+
"loss": 1.9146394729614258,
|
| 2388 |
+
"mean_token_accuracy": 0.6491621173918247,
|
| 2389 |
+
"num_tokens": 14133662.0,
|
| 2390 |
+
"step": 2380
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"entropy": 1.918326808512211,
|
| 2394 |
+
"epoch": 4.106597893831936,
|
| 2395 |
+
"grad_norm": 0.7571077346801758,
|
| 2396 |
+
"learning_rate": 0.00010444000000000001,
|
| 2397 |
+
"loss": 2.024713897705078,
|
| 2398 |
+
"mean_token_accuracy": 0.643079025298357,
|
| 2399 |
+
"num_tokens": 14196967.0,
|
| 2400 |
+
"step": 2390
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"entropy": 1.7909317679703236,
|
| 2404 |
+
"epoch": 4.123791102514507,
|
| 2405 |
+
"grad_norm": 0.7276471257209778,
|
| 2406 |
+
"learning_rate": 0.00010404,
|
| 2407 |
+
"loss": 1.845133399963379,
|
| 2408 |
+
"mean_token_accuracy": 0.6548417568206787,
|
| 2409 |
+
"num_tokens": 14256866.0,
|
| 2410 |
+
"step": 2400
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"entropy": 1.7750686906278133,
|
| 2414 |
+
"epoch": 4.140984311197077,
|
| 2415 |
+
"grad_norm": 0.668246328830719,
|
| 2416 |
+
"learning_rate": 0.00010364,
|
| 2417 |
+
"loss": 1.7945009231567384,
|
| 2418 |
+
"mean_token_accuracy": 0.6641525950282812,
|
| 2419 |
+
"num_tokens": 14318324.0,
|
| 2420 |
+
"step": 2410
|
| 2421 |
+
},
|
| 2422 |
+
{
|
| 2423 |
+
"entropy": 1.823828212916851,
|
| 2424 |
+
"epoch": 4.158177519879647,
|
| 2425 |
+
"grad_norm": 0.7596518993377686,
|
| 2426 |
+
"learning_rate": 0.00010324000000000001,
|
| 2427 |
+
"loss": 1.898871612548828,
|
| 2428 |
+
"mean_token_accuracy": 0.6519910141825676,
|
| 2429 |
+
"num_tokens": 14380775.0,
|
| 2430 |
+
"step": 2420
|
| 2431 |
+
},
|
| 2432 |
+
{
|
| 2433 |
+
"entropy": 1.7938876405358315,
|
| 2434 |
+
"epoch": 4.175370728562218,
|
| 2435 |
+
"grad_norm": 0.6834619641304016,
|
| 2436 |
+
"learning_rate": 0.00010284,
|
| 2437 |
+
"loss": 1.8518138885498048,
|
| 2438 |
+
"mean_token_accuracy": 0.6622516691684723,
|
| 2439 |
+
"num_tokens": 14440862.0,
|
| 2440 |
+
"step": 2430
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"entropy": 1.8744625180959702,
|
| 2444 |
+
"epoch": 4.192563937244788,
|
| 2445 |
+
"grad_norm": 0.8088146448135376,
|
| 2446 |
+
"learning_rate": 0.00010244,
|
| 2447 |
+
"loss": 1.9542848587036132,
|
| 2448 |
+
"mean_token_accuracy": 0.6499028638005256,
|
| 2449 |
+
"num_tokens": 14500841.0,
|
| 2450 |
+
"step": 2440
|
| 2451 |
+
},
|
| 2452 |
+
{
|
| 2453 |
+
"entropy": 1.8284114554524422,
|
| 2454 |
+
"epoch": 4.209757145927359,
|
| 2455 |
+
"grad_norm": 0.82193523645401,
|
| 2456 |
+
"learning_rate": 0.00010204,
|
| 2457 |
+
"loss": 1.9107404708862306,
|
| 2458 |
+
"mean_token_accuracy": 0.6551219135522842,
|
| 2459 |
+
"num_tokens": 14564257.0,
|
| 2460 |
+
"step": 2450
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"entropy": 1.8538024842739105,
|
| 2464 |
+
"epoch": 4.226950354609929,
|
| 2465 |
+
"grad_norm": 0.7263757586479187,
|
| 2466 |
+
"learning_rate": 0.00010164,
|
| 2467 |
+
"loss": 1.8713863372802735,
|
| 2468 |
+
"mean_token_accuracy": 0.6510257624089718,
|
| 2469 |
+
"num_tokens": 14623019.0,
|
| 2470 |
+
"step": 2460
|
| 2471 |
+
},
|
| 2472 |
+
{
|
| 2473 |
+
"entropy": 1.756752038002014,
|
| 2474 |
+
"epoch": 4.244143563292499,
|
| 2475 |
+
"grad_norm": 0.7334346175193787,
|
| 2476 |
+
"learning_rate": 0.00010124,
|
| 2477 |
+
"loss": 1.7855047225952148,
|
| 2478 |
+
"mean_token_accuracy": 0.6687729060649872,
|
| 2479 |
+
"num_tokens": 14682191.0,
|
| 2480 |
+
"step": 2470
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"entropy": 1.7032470375299453,
|
| 2484 |
+
"epoch": 4.26133677197507,
|
| 2485 |
+
"grad_norm": 0.7168938517570496,
|
| 2486 |
+
"learning_rate": 0.00010084,
|
| 2487 |
+
"loss": 1.7648530960083009,
|
| 2488 |
+
"mean_token_accuracy": 0.6696467150002718,
|
| 2489 |
+
"num_tokens": 14739840.0,
|
| 2490 |
+
"step": 2480
|
| 2491 |
+
},
|
| 2492 |
+
{
|
| 2493 |
+
"entropy": 1.7426577515900135,
|
| 2494 |
+
"epoch": 4.27852998065764,
|
| 2495 |
+
"grad_norm": 0.7091065645217896,
|
| 2496 |
+
"learning_rate": 0.00010044000000000001,
|
| 2497 |
+
"loss": 1.8180946350097655,
|
| 2498 |
+
"mean_token_accuracy": 0.6640235505998134,
|
| 2499 |
+
"num_tokens": 14798444.0,
|
| 2500 |
+
"step": 2490
|
| 2501 |
+
},
|
| 2502 |
+
{
|
| 2503 |
+
"entropy": 1.8743537411093711,
|
| 2504 |
+
"epoch": 4.295723189340211,
|
| 2505 |
+
"grad_norm": 0.6376718878746033,
|
| 2506 |
+
"learning_rate": 0.00010004,
|
| 2507 |
+
"loss": 1.9534942626953125,
|
| 2508 |
+
"mean_token_accuracy": 0.6467559643089771,
|
| 2509 |
+
"num_tokens": 14861262.0,
|
| 2510 |
+
"step": 2500
|
| 2511 |
}
|
| 2512 |
],
|
| 2513 |
"logging_steps": 10,
|
|
|
|
| 2527 |
"attributes": {}
|
| 2528 |
}
|
| 2529 |
},
|
| 2530 |
+
"total_flos": 1.2204939073814528e+17,
|
| 2531 |
"train_batch_size": 2,
|
| 2532 |
"trial_name": null,
|
| 2533 |
"trial_params": null
|