INDEX

Explanations

structures or comparisons of differences

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

Jer

-0.15

æĹ¢

-0.15

sn

-0.14

issor

-0.14

LIK

-0.14

 surprisingly

-0.14

gate

-0.14

Jer

-0.13

ave

-0.13

 fairly

-0.13

POSITIVE LOGITS

 slightly

0.19

unwrap

0.14

.scalablytyped

0.14

æĽ´åĬł

0.14

 additional

0.14

orsk

0.14

ç¨į

0.13

ensa

0.13

ãĥ©ãĤ¯

0.13

 ãĦ

0.13

Activations Density 0.055%

structures or comparisons of differences

No Comments

No Known Activations