INDEX

Explanations

instances of error or validation messages in code

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

code, non-english words

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

Datuak

-0.63

StoryboardSegue

-0.59

AnchorStyles

-0.57

tagHelperRunner

-0.56

IUrlHelper

-0.56

RTSC

-0.55

GraphicsUnit

-0.55

+#+

-0.54

fromCharCode

-0.53

findpost

-0.51

POSITIVE LOGITS

 umana

0.41

 umani

0.37

ResumeLayout

0.37

 despair

0.36

 geweest

0.36

 iniziale

0.36

nadequate

0.35

itonitis

0.35

colgroup

0.35

 ISNI

0.35

Activations Density 0.016%

instances of error or validation messages in code

code, non-english words

No Comments

No Known Activations