INDEX

Explanations

occurrences of the word "to" in various contexts

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

if [subject] were to [verb]

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

 cherchés

-0.72

 propOrder

-0.66

 ویکی‌پدی

-0.63

 للاسماء

-0.58

Portale

-0.58

 estekak

-0.57

Tikang

-0.56

IUrlHelper

-0.55

 autorytatywna

-0.53

 Taktlose

-0.52

POSITIVE LOGITS

 would

0.49

 Would

0.47

bleau

0.44

 Should

0.44

 Sider

0.43

Would

0.43

 skulle

0.43

 WOULD

0.42

 prospect

0.42

would

0.41

Activations Density 0.010%

occurrences of the word "to" in various contexts

if [subject] were to [verb]

No Comments

No Known Activations