INDEX

Explanations

named entities, such as people and places, as well as significance of specific terms or identifiers

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

fnlp/Llama-Scope-R1-Distill/800M-Slimpajama-0-OpenR1-Math-220k/L15R

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Hzfinfdu/SlimPajama-3B

Features

32,768

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Penal

-0.16

å¾

-0.14

dun

-0.13

dev

-0.13

Î½Ïī

-0.13

shan

-0.13

GET

-0.13

reator

-0.13

-www

-0.12

POSITIVE LOGITS

again

0.17

Again

0.15

ramid

0.15

 again

0.15

 Again

0.15

 novamente

0.14

 çĻ

0.14

gratis

0.14

Gratis

0.14

iasm

0.14

Activations Density 0.048%

named entities, such as people and places, as well as significance of specific terms or identifiers

No Comments

No Known Activations