INDEX

Explanations

terms related to watches and their features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

inves

-0.42

 invokingState

-0.41

 inves

-0.39

condition

-0.39

 condition

-0.38

 zeitung

-0.37

grenze

-0.37

 condição

-0.37

 condizione

-0.37

Mot

-0.36

POSITIVE LOGITS

RenderAtEndOf

0.53

MLLoader

0.49

 iconic

0.49

 silueta

0.48

 EconPapers

0.48

 mergeFrom

0.45

 ✭✭

0.45

 lenker

0.43

 anniversary

0.43

 classic

0.42

Activations Density 0.005%

terms related to watches and their features

No Comments

No Known Activations