INDEX
Explanations
The neuron activates on occurrences of the word “mining” (and its close morphological variants, like “miner” or “mineral”), flagging mentions of mining-related terms.
New Auto-Interp
Negative Logits
сахар
-0.07
BÖL
-0.07
βάλ
-0.06
.bel
-0.06
конструк
-0.06
osal
-0.06
езда
-0.06
فرد
-0.06
templates
-0.06
hepatitis
-0.06
POSITIVE LOGITS
mining
0.14
Mining
0.13
Mine
0.12
Mining
0.11
mine
0.11
mines
0.10
Mines
0.10
Miner
0.08
Mine
0.08
ynn
0.08
Activations Density 0.008%