INDEX

Explanations

references to notable achievements and recognitions in the UK or world context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

matchCondition

-0.58

RenderAtEndOf

-0.57

 للمعارف

-0.53

 transfieras

-0.53

uca

-0.53

UnsafeEnabled

-0.51

شهاد

-0.48

 متعلقه

-0.48

avu

-0.47

minmax

-0.46

POSITIVE LOGITS

 largest

1.21

 finest

1.19

 cleanest

1.09

 best

1.07

 fastest

1.06

 Largest

1.01

largest

1.01

 richest

1.00

 tallest

1.00

 most

0.98

Activations Density 0.198%

references to notable achievements and recognitions in the UK or world context

No Comments

No Known Activations