INDEX

Explanations

negations and phrases questioning the state or quality of something

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

idon

-0.19

htt

-0.15

oly

-0.15

ellen

-0.14

mic

-0.14

.slim

-0.13

zf

-0.13

 Lilly

-0.13

may

-0.13

 Armstrong

-0.13

POSITIVE LOGITS

Ã³ng

0.15

 rather

0.15

ãĥĥ

0.14

Ð¾Ð´Ð°Ð²

0.14

ometrics

0.14

Calibri

0.13

rather

0.13

ighest

0.13

ī´

0.13

aw

0.13

Activations Density 0.025%

negations and phrases questioning the state or quality of something

No Comments

No Known Activations