INDEX

Explanations

terms related to capability, ability, and conditions describing functionality or state

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

adh

-0.15

ãĥķãĤ

-0.15

aises

-0.14

rij

-0.13

wend

-0.13

zen

-0.13

.sponge

-0.13

 Tarif

-0.13

 Inline

-0.13

quina

-0.13

POSITIVE LOGITS

strar

0.17

 loose

0.17

ters

0.16

ropp

0.16

_SYS

0.15

atron

0.15

_sys

0.14

antz

0.14

aders

0.14

rouw

0.14

Activations Density 0.003%

terms related to capability, ability, and conditions describing functionality or state

No Comments

No Known Activations