INDEX

Explanations

mentions of various forms of the word "press."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

fromnode

-0.73



-0.73

 sherds

-0.72

 wattpad

-0.67

 intStringLen

-0.66

OnTop

-0.65

 hulls

-0.64

 iprot

-0.64

iprot

-0.64

 welds

-0.63

POSITIVE LOGITS

 press

0.94

 pressing

0.84

 Press

0.81

pressing

0.80

Water

0.76

 water

0.74

 Water

0.73

press

0.72

PRESS

0.67

water

0.66

Activations Density 0.173%

mentions of various forms of the word "press."

No Comments

No Known Activations