INDEX

Explanations

content related to scientific and technical terminology

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

 CreateTagHelper

-0.65

 <>",

-0.53

ніципа

-0.53

thoven

-0.50

 muitas

-0.49

 nombreux

-0.48

uchter

-0.46

 muchas

-0.46

Compat

-0.45

Four

-0.45

POSITIVE LOGITS

one

2.21

一个是

1.65

one

1.53

 unul

1.40

uno

1.29

 één

1.26

 satu

1.20

 один

1.18

 یکی

1.13

 одна

1.10

Activations Density 0.429%

content related to scientific and technical terminology

No Comments

No Known Activations