INDEX

Explanations

references to structured lists and profiles of characters or entities

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

-0.18

↵

-0.16

 Barry

-0.15

 Wind

-0.15

 list

-0.14

dib

-0.14

 pres

-0.14

POSITIVE LOGITS

ersh

0.15

turnstile

0.14

endid

0.14

endale

0.14

odont

0.14

ï»¿using

0.14

RequestMethod

0.14

layÄ±cÄ±

0.14

scriptId

0.14

AZY

0.13

Activations Density 0.020%

references to structured lists and profiles of characters or entities

No Comments

No Known Activations