INDEX

Explanations

words related to logical reasoning and arguments

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

allah

-0.16

lage

-0.16

 Khoa

-0.15

alah

-0.14

elp

-0.13

heritance

-0.13

çĦ¡ãģĹãģ

-0.13

atr

-0.13

otas

-0.13

REP

-0.13

POSITIVE LOGITS

 scenario

0.22

 scenarios

0.21

 Scenario

0.18

Scenario

0.18

 example

0.17

 concept

0.17

sim

0.17

 this

0.16

 examples

0.15

 models

0.15

Activations Density 0.228%

words related to logical reasoning and arguments

No Comments

No Known Activations