INDEX

Explanations

references to errors or mistakes in programming contexts

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

errors or empty checks

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

 mxArray

-0.54

ynb

-0.53

'},

-0.52

>",

-0.52

 thiệu

-0.51

genta

-0.50

 собою

-0.49

}$

-0.48

FieldNumber

-0.48

"],

-0.47

POSITIVE LOGITS

 NSCoder

0.63

ConstraintMaker

0.58

PRWEB

0.58

 متعلقه

0.57

Bioaccumulative

0.52

verifyException

0.51

EXPECT

0.50

 corredores

0.50

ntax

0.50

각주

0.49

Activations Density 0.029%

references to errors or mistakes in programming contexts

errors or empty checks

No Comments

No Known Activations