INDEX

Explanations

terms related to error codes and messages in a technical context

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

ffmpeg format errors

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

 calzado

-0.39

 career

-0.39

 golfers

-0.38

IndentedString

-0.37

AndroidJUnit

-0.37

 ciudadana

-0.37

ufficio

-0.36

 artistique

-0.36

bukkit

-0.36

 depresión

-0.35

POSITIVE LOGITS

ffmpeg

0.73

oplayer

0.66

SourceChecksum

0.57

 Cinema

0.56

mpeg

0.55

 cinema

0.54

Tembelea

0.54

 Cine

0.54

 cine

0.54

 Movie

0.54

Activations Density 0.057%

terms related to error codes and messages in a technical context

ffmpeg format errors

No Comments

No Known Activations