INDEX

Explanations

phrases related to excessive or under-evaluated contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 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

httphttps

-0.45

aarrggbb

-0.42

Clik

-0.42

:+:

-0.41

Hentet

-0.40

tagHelperRunner

-0.39

Pil

-0.39

IntoConstraints

-0.38

SPJ

-0.38

 ligiloj

-0.37

POSITIVE LOGITS

 Overs

0.95

过度

0.85

 excess

0.82

 oversized

0.80

 Excess

0.79

Overs

0.79

 exceso

0.79

excess

0.78

 overdose

0.76

Excess

0.75

Activations Density 0.400%

phrases related to excessive or under-evaluated contexts

No Comments

No Known Activations