INDEX

Explanations

terms related to structured programs, situations, and conditions

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

urv

-0.16

ãĥĥ

-0.15

uards

-0.14

DCF

-0.13

malink

-0.13

_Tis

-0.13

:async

-0.13

/if

-0.13

rawn

-0.13

ruz

-0.13

POSITIVE LOGITS

iba

0.14

owej

0.13

URES

0.13

ifr

0.13

 Ø§ÙĨØ¯

0.13

bron

0.13

anke

0.13

âĢı

0.13

Ced

0.13

0.12

Activations Density 0.018%

terms related to structured programs, situations, and conditions

No Comments

No Known Activations