INDEX

Explanations

specific articles and terms related to medical or scientific contexts

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

ascar

-0.17

errer

-0.16

ÐºÑĢÐµÑĤ

-0.15

rico

-0.14

ntl

-0.14

ayet

-0.14

ibase

-0.14

ibrator

-0.13

abay

-0.13

POSITIVE LOGITS

 means

0.26

 virtue

0.19

means

0.18

gone

0.17

 Ð¼ÐµÑĢÐµ

0.16

 Means

0.15

-pass

0.15

-products

0.15

 cont

0.15

 mistake

0.14

Activations Density 0.325%

specific articles and terms related to medical or scientific contexts

No Comments

No Known Activations