INDEX

Explanations

phrases and terms associated with perceptions or evaluations of concepts

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

regarded as

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

BagLayout

-0.45

ymal

-0.44

년

-0.42

orari

-0.41

 modifications

-0.40

 nonUne

-0.40

雳

-0.40

getOutputStream

-0.39

the

-0.39

clearpage

-0.39

POSITIVE LOGITS

 regarded

0.57

IntoConstraints

0.57

MigrationBuilder

0.56

 CreateTagHelper

0.45

 behaving

0.45

thâu

0.45

CloseOperation

0.45

 ISNI

0.45

 SqlDbType

0.45

setVerticalGroup

0.43

Activations Density 0.067%

phrases and terms associated with perceptions or evaluations of concepts

regarded as

No Comments

No Known Activations