INDEX

Explanations

the word "that" frequently and in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Efq

-0.80

 propOrder

-0.78

DoubleQuotes

-0.67

 Offisielt

-0.65

 consisting

-0.63

 credere

-0.63

ρίζ

-0.63

 Monfieur

-0.63

voorbeeld

-0.63

XmlRootElement

-0.62

POSITIVE LOGITS

 دیکھیے

0.62

 Roskov

0.51

ItemLayout

0.50

 оригіналу

0.48

כשיו

0.48

клопе

0.48

RegressionTest

0.45

 typelib

0.45

횟

0.45

OVERALL

0.44

Activations Density 0.238%

the word "that" frequently and in various contexts

No Comments

No Known Activations