INDEX

Explanations

sentences that contain high activation values, indicating important or impactful statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

te

-0.45

Newswire

-0.45

坦

-0.43

лев

-0.43

óa

-0.42

wj

-0.42

vania

-0.42

GV

-0.41

PostMapping

-0.41

mav

-0.41

POSITIVE LOGITS

+#+#

1.06

tagHelperRunner

0.93

 resourceCulture

0.90

 مشين

0.86

writeFieldEnd

0.82

WebVitals

0.82

 iconFacebook

0.81

(!__

0.79

SharedDtor

0.78

 BoxFit

0.77

Activations Density 0.251%

sentences that contain high activation values, indicating important or impactful statements

No Comments

No Known Activations