INDEX

Explanations

questions and expressions of curiosity or self-reflection

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

conversational endings

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

+#+#

-0.82

 referenties

-0.50

 disponibilités

-0.47

oa̍t

-0.47

ThroughAttribute

-0.47

expandindo

-0.45

rrggbb

-0.45

Tikang

-0.44

setVerticalGroup

-0.44

 CreateTagHelper

-0.44

POSITIVE LOGITS

hood

0.41

 AssemblyCompany

0.38

runOn

0.35

urtstag

0.35

heck

0.35

getSupport

0.35

honest

0.35

iastes

0.35

Composição

0.35

itness

0.34

Activations Density 0.144%

questions and expressions of curiosity or self-reflection

conversational endings

No Comments

No Known Activations