INDEX

Explanations

news and announcements related to entertainment and media

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

actions occurring in sequence

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

FSM

-0.43

Clever

-0.42

 ...]

-0.41

 Brave

-0.41

 {...

-0.40

 Contrast

-0.40

 pere

-0.39

menea

-0.39

bê

-0.39

wasm

-0.38

POSITIVE LOGITS

AddTagHelper

0.66

 referrerpolicy

0.49

 bowiem

0.49

tvguidetime

0.47

 totiž

0.47

 חיצוניים

0.46

 AssemblyCulture

0.45

 تضيفلها

0.45

 محفوظة

0.44

XmlAccessType

0.44

Activations Density 0.037%

news and announcements related to entertainment and media

actions occurring in sequence

No Comments

No Known Activations