INDEX

Explanations

the word "from" and error messages related to TypeErrors

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

appeal from

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

InjectAttribute

-0.58

 MessageBoxIcon

-0.55

 HttpNotFound

-0.54

 ſche

-0.54

HideFlags

-0.52

nloa

-0.48

TextHelper

-0.47

ungeon

-0.47

ंदीखरीदारी

-0.47

ftagPool

-0.46

POSITIVE LOGITS

 from

1.72

 FROM

1.20

from

1.16

 từ

1.13

 From

1.03

 dari

1.02

จาก

1.00

From

0.92

 från

0.92

 από

0.85

Activations Density 0.009%

the word "from" and error messages related to TypeErrors

appeal from

No Comments

No Known Activations