INDEX

Explanations

references to emotional states or conditions related to loss and support

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

.gb

-0.16

istrovstvÃŃ

-0.15

è¾°

-0.14

krom

-0.14

awy

-0.14

OAD

-0.14

 fitte

-0.14

TemplateName

-0.14

rypto

-0.14

abe

-0.14

POSITIVE LOGITS

 itself

0.22

 consisting

0.18

 consists

0.17

 consist

0.16

urch

0.16

-wide

0.16

 composition

0.15

wide

0.15

 wide

0.15

Activations Density 0.303%

references to emotional states or conditions related to loss and support

No Comments

No Known Activations