INDEX

Explanations

references to popular media and cultural events

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

ÑĢÑıÐ´Ñĥ

-0.17

awan

-0.15

ocale

-0.15

jez

-0.15

Ð»Ð¸ÐºÐ°

-0.14

athi

-0.14

uros

-0.14

istrovstvÃŃ

-0.13

ipples

-0.13

 helicopt

-0.13

POSITIVE LOGITS

 episode

0.26

 Segment

0.24

 Episode

0.23

 segment

0.22

 guests

0.22

 segments

0.21

 listeners

0.20

 guest

0.20

 talk

0.20

 listener

0.20

Activations Density 0.085%

references to popular media and cultural events

No Comments

No Known Activations