INDEX

Explanations

programming-related keywords and syntax elements

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

the second question

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

tagHelperRunner

-1.03

 snippetHide

-0.88

AddTagHelper

-0.88

 JpaRepository

-0.82

<unused8>

-0.78

<unused79>

-0.78

<unused52>

-0.77

<unused16>

-0.77

<unused14>

-0.77

[@BOS@]

-0.77

POSITIVE LOGITS

 finishing

0.41

 finish

0.36

 finished

0.34

FINISH

0.31

 FINISH

0.31

 графи

0.30

finish

0.30

 completed

0.29

Finish

0.29

Finishing

0.27

Activations Density 0.001%

programming-related keywords and syntax elements

the second question

No Comments

No Known Activations