INDEX

Explanations

phrases associated with subjective judgments and comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

cessite

-0.34

(!__

-0.30

TimeSpan

-0.29

 étoit

-0.29

わかります

-0.28

guisement

-0.28

分かります

-0.28

UnusedPrivate

-0.28

しまいました

-0.26

 avoient

-0.25

POSITIVE LOGITS

ьаж

0.67

Kariera

0.62

الحياه

0.62

IntoConstraints

0.61

 surla

0.61

RTEX

0.59

anches

0.59

Jeografia

0.59

EI

0.58

schild

0.58

Activations Density 0.391%

phrases associated with subjective judgments and comparisons

No Comments

No Known Activations