INDEX

Explanations

repeated instances of the word "connected."

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

connected to or with

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

<bos>

-0.47

if

-0.39

 Standardi

-0.38

FA

-0.37

try

-0.36

YC

-0.36

imdi

-0.36

fa

-0.36

ark

-0.35

 וי

-0.35

POSITIVE LOGITS

 Connected

1.59

 connected

1.56

connected

1.55

Connected

1.49

CONNECTED

1.23

 verbonden

1.04

 interconnected

1.02

 propOrder

0.95

 conectado

0.91

disconnected

0.90

Activations Density 0.017%

repeated instances of the word "connected."

connected to or with

No Comments

No Known Activations

repeated instances of the word "connected."

connected to or with

No Comments

No Known Activations