INDEX

Explanations

common words and punctuation, possibly relating to conversational structure

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

selling cost

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

the neuron detects contracted/clitic tokens (apostrophes and the pieces of contractions like 'll, n't, 's, etc.).

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

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

 CreateTagHelper

-0.70

enterOuterAlt

-0.69

########.

-0.63

 Мексичка

-0.58

setViewportView

-0.57

 MessageBoxIcon

-0.54

клопе

-0.54

TestBed

-0.54

FailureListener

-0.54

下载附件

-0.52

POSITIVE LOGITS

your

0.43

 democrática

0.42

 yourself

0.39

 chcesz

0.39

ɚ

0.38

 descans

0.38

Your

0.37

 your

0.37

 temperatura

0.36

 Jehová

0.36

Activations Density 0.036%

common words and punctuation, possibly relating to conversational structure

selling cost

the neuron detects contracted/clitic tokens (apostrophes and the pieces of contractions like 'll, n't, 's, etc.).

No Comments

No Known Activations