INDEX

Explanations

information related to students and their engagement or interactions, particularly in a faith context

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_19/width_16k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ciclop

-0.59

 masukan

-0.49

 Paral

-0.49

Turn

-0.48

 khóa

-0.47

matism

-0.46

diac

-0.46

clud

-0.45

bif

-0.45

AUX

-0.44

POSITIVE LOGITS

 presenting

1.17

Presenting

1.08

 presentation

1.07

 presentations

1.05

 Presenting

1.02

 Presentation

0.97

presentation

0.93

 Presentations

0.92

 présentation

0.91

 PRESENTATION

0.91

Activations Density 0.253%

information related to students and their engagement or interactions, particularly in a faith context

No Known Activations