INDEX

Explanations

references to tubes and connected components in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 Snell

-0.79

resiasi

-0.77

makeConstraints

-0.75

SpringRunner

-0.73

hofen

-0.72

سكرية

-0.71

jeev

-0.70

Dak

-0.70

 tramontina

-0.70

ństw

-0.69

POSITIVE LOGITS

:"-"`

0.91

%)$

0.81

\}\\

0.74

 Jacobson

0.74

Zoe

0.69

Vea

0.69

 unknownFields

0.69

Noi

0.68

Kra

0.68

}]

0.68

Activations Density 0.897%

references to tubes and connected components in various contexts

No Comments

No Known Activations