INDEX

Explanations

keywords associated with awareness and introspection

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

unar

-0.19

etri

-0.17

yclopedia

-0.14

opers

-0.14

endo

-0.14

Ã¤n

-0.14

enberg

-0.13

jest

-0.13

ersh

-0.13

POSITIVE LOGITS

 underlying

0.15

 ActiveForm

0.14

å¼ĢæĶ¾

0.14

éľ²åĩº

0.14

ç«ľ

0.14

éľ²

0.13

steder

0.13

PX

0.13

 access

0.13

ãĥ³ãĥ

0.12

Activations Density 0.036%

keywords associated with awareness and introspection

No Comments

No Known Activations