INDEX

Explanations

reported statements and claims made by officials or authorities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

chein

-0.16

kez

-0.15

sak

-0.14

 Charter

-0.14

ak

-0.14

Ã©li

-0.14

 Jury

-0.14

ital

-0.14

itan

-0.13

connect

-0.13

POSITIVE LOGITS

_bulk

0.15

UBLE

0.15

entina

0.15

ArrayOf

0.14

PageRoute

0.14

ÃĸL

0.14

leck

0.14

-wsj

0.14

ivant

0.14

:selected

0.14

Activations Density 0.065%

reported statements and claims made by officials or authorities

No Comments

No Known Activations