INDEX

Explanations

words associated with caution and warnings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

EndInit

-0.50

yAxis

-0.41

BeginInit

-0.40

余

-0.38

java

-0.38

pro

-0.38

est

-0.38

redux

-0.37

HasValue

-0.37

Pro

-0.36

POSITIVE LOGITS

 there

2.08

there

1.62

 THERE

1.55

 thither

1.27

THERE

1.27

There

1.25

 There

1.25

 acolo

1.05

theres

0.94

 ahí

0.94

Activations Density 0.140%

words associated with caution and warnings

No Comments

No Known Activations