INDEX

Explanations

phrases expressing gratitude and appreciation

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

aarrggbb

-0.74

 Majefty

-0.72

AddTagHelper

-0.72

 Theſe

-0.71

 ſche

-0.70

 Conſ

-0.70

 becauſe

-0.70

 ſtate

-0.70

uxxxx

-0.69

 ſtre

-0.68

POSITIVE LOGITS

 pleasure

2.15

pleasure

1.86

 privilege

1.71

 honor

1.71

 honour

1.52

 Pleasure

1.50

privilege

1.40

 honored

1.35

honor

1.32

 plaisir

1.28

Activations Density 0.164%

phrases expressing gratitude and appreciation

No Comments

No Known Activations