INDEX

Explanations

technical terms and code-related elements in a programming context

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

channel or Session

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

GOTREF

-0.75

AndEndTag

-0.74

IntoConstraints

-0.74

 CreateTagHelper

-0.73

expandindo

-0.71

fjspx

-0.67

Datuak

-0.65

QMetaType

-0.64

 endforeach

-0.63

 Signalez

-0.59

POSITIVE LOGITS

 channels

0.46

 channel

0.44

 canales

0.39

 Channels

0.39

channels

0.38

channel

0.38

Channel

0.36

 canali

0.35

Channels

0.35

 Channel

0.35

Activations Density 0.002%

technical terms and code-related elements in a programming context

channel or Session

No Comments

No Known Activations