INDEX

Explanations

intensifiers of degree

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

acts as an intensifier

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

usercontent

-0.55

PDC

-0.55

setDo

-0.54

ReusableCell

-0.52

IoT

-0.50

Lid

-0.50

 cartera

-0.50

CDP

-0.50

SDP

-0.50

NOPQRST

-0.49

POSITIVE LOGITS

 extremely

1.09

extremely

0.94

Extremely

0.85

 highly

0.84

 Extremely

0.83

 extrêmement

0.79

 extremadamente

0.75

 exceedingly

0.72

 very

0.72

 extraordinarily

0.71

Activations Density 0.004%

intensifiers of degree

acts as an intensifier

No Comments

No Known Activations