INDEX

Explanations

mentions of loss and queries related to information and identity

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

preposition followed by possessive or name

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

UAGE

-0.64

 Roskov

-0.63

Rohy

-0.62

 оригіналу

-0.61

UMBIA

-0.61

Vidite

-0.60

клопе

-0.60

 محفوظة

-0.59

 eventdata

-0.59

kháu

-0.58

POSITIVE LOGITS

enumi

0.62

 vastaan

0.57

 userSchema

0.57

 marito

0.56

 reçoit

0.54

 régionale

0.54

 Schmitz

0.53

nocache

0.53

 nélkül

0.52

 Jerusalén

0.52

Activations Density 0.019%

mentions of loss and queries related to information and identity

preposition followed by possessive or name

No Comments

No Known Activations