INDEX

Explanations

words related to tagging and managing resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

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

waitKey

-0.62

 FetchType

-0.59

GraphicsUnit

-0.56

ArgsConstructor

-0.55

nthetic

-0.52

 things

-0.52

GetMapping

-0.51

bootstrapcdn

-0.51

CloseOperation

-0.50

 thing

-0.49

POSITIVE LOGITS

 betweenstory

0.72

 enfans

0.68

 Grecs

0.68

 stället

0.66

berdayakan

0.66

 myſelf

0.65

eraard

0.62

 iprot

0.62

 gesprek

0.61

 zufolge

0.61

Activations Density 0.841%

words related to tagging and managing resources

No Comments

No Known Activations