INDEX

Explanations

patterns or structures in code or data syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

 Hilda

-0.69

uha

-0.65

ustra

-0.64

 official

-0.63

makeatother

-0.63

istically

-0.62

jo

-0.62

 paper

-0.62

 Paper

-0.62

AX

-0.61

POSITIVE LOGITS

\{\\

1.13

ViewFeatures

0.91

mrow

0.86

 nakalista

0.82

↵

0.75

}")]

0.73

/******/

0.71

BeginContext

0.71

})));

0.69

<tbody>

0.69

Activations Density 0.150%

patterns or structures in code or data syntax

No Comments

No Known Activations