INDEX

Explanations

temporal indicators associated with durations or time periods

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 viewType

-0.41

 quoque

-0.35

 types

-0.34

 poly

-0.33

nodeId

-0.30

Diweddarwch

-0.30

 fans

-0.30

 family

-0.30

mix

-0.30

 kura

-0.30

POSITIVE LOGITS

 consultato

0.69

 nakalista

0.69

 noDo

0.65

consulté

0.64

istoitu

0.64

endphp

0.63

 يتيمه

0.61

Décès

0.61

 Photocase

0.61

 Himo

0.60

Activations Density 0.027%

temporal indicators associated with durations or time periods

No Comments

No Known Activations