INDEX

Explanations

references to medical terms and procedures

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

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥ³ãĥĳ

-0.14

.dsl

-0.14

 Grat

-0.14

raquo

-0.14

 Ø§Ø°

-0.13

illin

-0.13

.bc

-0.13

ductive

-0.13

 èĦ

-0.13

Bomb

-0.13

POSITIVE LOGITS

amam

0.18

no

0.14

 Holy

0.14

 Zion

0.14

kil

0.13

 Holmes

0.13

iten

0.13

 mang

0.12

 Styled

0.12

Nos

0.12

Activations Density 0.088%

references to medical terms and procedures

No Comments

No Known Activations