INDEX

Explanations

terms related to classification and redefinition processes

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

re- prefix, action verbs

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

LookAnd

-0.52

 صوتيه

-0.50

 ब्रेकडाउन

-0.49

cupertino

-0.49

 Administrativna

-0.49

ExtendWith

-0.48

ParallelGroup

-0.48

plier

-0.48

GenerationType

-0.47

tanleria

-0.47

POSITIVE LOGITS

 reorgan

0.57

 rebuilding

0.55

 reread

0.53

 rejoin

0.52

 recast

0.51

 tái

0.50

재

0.50

 reorganized

0.49

 rebuilt

0.49

 rebuild

0.49

Activations Density 0.059%

terms related to classification and redefinition processes

re- prefix, action verbs

No Comments

No Known Activations