INDEX

Explanations

phrases indicating positive outcomes and beneficiaries related to various subjects

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

enberg

-0.16

rott

-0.15

 waste

-0.15

å¿µ

-0.14

ance

-0.14

 kettle

-0.14

å³°

-0.13

bubble

-0.13

 liken

-0.13

 Waste

-0.13

POSITIVE LOGITS

 benefit

0.21

çĽĬ

0.19

 benefits

0.17

 Benefit

0.17

 benefited

0.16

 compartments

0.16

 benefici

0.16

/english

0.16

 beneficiaries

0.15

.mdl

0.15

Activations Density 0.084%

phrases indicating positive outcomes and beneficiaries related to various subjects

No Comments

No Known Activations