INDEX

Explanations

instances of strong claims or actions related to authority and influence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 RSSSF

-0.77

nextDouble

-0.69

 myſelf

-0.65

HtmlAttribute

-0.64

িনি

-0.62

 CanadaChoose

-0.61

simus

-0.60

 purpoſe

-0.60

HostException

-0.60

 Allociné

-0.58

POSITIVE LOGITS

its

4.26

Its

3.47

Its

3.42

its

2.96

 itself

2.58

它的

2.44

 Оно

2.15

 Itself

2.10

ITS

2.01

itself

2.01

Activations Density 3.709%

instances of strong claims or actions related to authority and influence

No Comments

No Known Activations