INDEX

Explanations

words and phrases related to academic and technical terminology in research 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

Diweddarwch

-0.65

WireFormatLite

-0.64

 fubject

-0.61

vég

-0.59

Rohy

-0.58

zmán

-0.56

atare

-0.56

 transférez

-0.56

 виправивши

-0.55

 fhort

-0.55

POSITIVE LOGITS

era

0.91

err

0.85

er

0.85

ERA

0.78

ero

0.78

erv

0.77

eras

0.77

ner

0.77

NER

0.72

ier

0.71

Activations Density 2.104%

words and phrases related to academic and technical terminology in research contexts

No Comments

No Known Activations