INDEX

Explanations

academic terms and references related to scientific studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

CloseOperation

-0.56

XtraReports

-0.47

sun

-0.45

writeFieldEnd

-0.44

shi

-0.44

ske

-0.43

snow

-0.43

 iNdEx

-0.43

sch

-0.42

ski

-0.42

POSITIVE LOGITS

sa

2.08

Sa

2.05

Sa

1.88

SA

1.87

SA

1.71

Sá

1.70

sa

1.60

 са

1.59

 Са

1.52

SAP

1.45

Activations Density 2.098%

academic terms and references related to scientific studies

No Comments

No Known Activations