INDEX

Explanations

terms related to user interfaces and programming errors in a technical context

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

UI elements and navigation

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

 uParam

-0.48

 errorMessage

-0.45

MarshalTo

-0.45

xase

-0.44

 '\\;'

-0.40

__()

-0.40

protoimpl

-0.40

erschrank

-0.40

yntaxException

-0.40

libft

-0.39

POSITIVE LOGITS

tagHelperRunner

0.54

TagMode

0.47

tangentMode

0.43

BindingResult

0.43

 fué

0.43

Kjelder

0.43

Viited

0.42

webElementXpaths

0.41

 paik

0.41

plugged

0.41

Activations Density 0.158%

terms related to user interfaces and programming errors in a technical context

UI elements and navigation

No Comments

No Known Activations