INDEX

Explanations

concepts related to comparison and categorization

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

conditional possibilities and specific contexts

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

 VIAF

-0.36

SystemColors

-0.35

 resultCode

-0.35

 Solidar

-0.30

 beaucoup

-0.30

эффици

-0.30

 many

-0.29

 трудо

-0.29

Sent

-0.29

包

-0.29

POSITIVE LOGITS

 ffilmiau

0.63

Datuak

0.59

 незавершена

0.56

Externé

0.55

 invokingState

0.54

principalTable

0.54

:✨

0.53

󠁴

0.52

ɵɵ

0.52

RTEX

0.52

Activations Density 0.092%

concepts related to comparison and categorization

conditional possibilities and specific contexts

No Comments

No Known Activations