INDEX

Explanations

references to mathematical and statistical concepts, particularly those related to modeling and evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

findpost

-0.80

 Monfieur

-0.80

AccessorTable

-0.73

 myſelf

-0.70

 automatiques

-0.70

IntoConstraints

-0.66

 leaſt

-0.66

iconque

-0.64

 Theſe

-0.63

 Diſ

-0.62

POSITIVE LOGITS

 easy

0.69

easy

0.60

RenderAtEndOf

0.59

็จ

0.59

 mudah

0.58

ViewFeatures

0.58

 facile

0.57

 fixed

0.57

 Easy

0.56

Fixed

0.56

Activations Density 0.809%

references to mathematical and statistical concepts, particularly those related to modeling and evaluation

No Comments

No Known Activations