INDEX

Explanations

code-related syntax, specifically focusing on hash key definitions and commands in a programming context

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

Programming code or symbols

np_acts-logits-general · gemini-2.0-flash

variable assignments or code syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

مقاله

-0.48

ogu

-0.45

 RecyclerView

-0.45

Fcn

-0.43

 Nigerian

-0.43

chak

-0.43

andowski

-0.43

akos

-0.41

ónico

-0.41

ActivityResult

-0.41

POSITIVE LOGITS

 CreateTagHelper

0.75

+#+#

0.71



0.70

клопе

0.69

 undef

0.66

writeFieldEnd

0.65

 scalar

0.65

 wireType

0.64

ছে

0.64

onAttach

0.61

Activations Density 0.217%

code-related syntax, specifically focusing on hash key definitions and commands in a programming context

Programming code or symbols

variable assignments or code syntax

No Comments

No Known Activations