INDEX

Explanations

words related to writing and documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

ctools

-0.53

বহ

-0.53

posedge

-0.52

mybatisplus

-0.51

 Mila

-0.51

ldquo

-0.50

 ligiloj

-0.50

favicon

-0.50

Viitteet

-0.50

ccm

-0.47

POSITIVE LOGITS

 write

1.02

 Write

0.96

 Writes

0.96

 writers

0.95

 writing

0.94

 writes

0.93

 WRITE

0.91

 Writing

0.90

write

0.89

 Writers

0.88

Activations Density 0.306%

words related to writing and documentation

No Comments

No Known Activations