INDEX

Explanations

numerical ratings or scores, particularly on a 5-point or 10-point scale.

oai_token-act-pair · claude-3-5-sonnet-20240620 Triggered by @bot

highly positive ratings and reviews of books or movies.

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

positive reviews for books or movies.

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

Fractions representing ratings or scores

np_acts-logits-general · gemini-2.0-flash

ratings and scores

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DeleteBehavior

-0.76

出版年

-0.74

GraphicsUnit

-0.69

rungsseite

-0.68

findpost

-0.67

 يتيمه

-0.65

WireFormatLite

-0.64

ColumnHeaders

-0.60

IndentedString

-0.60

Vidite

-0.59

POSITIVE LOGITS

 rating

1.33

 score

1.24

rating

1.11

 Rating

1.05

 ratings

1.03

score

1.02

 Score

1.00

Rating

1.00

 RATING

0.98

 scores

0.97

Activations Density 0.194%