INDEX

Explanations

locations and geographic references

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

Ñĳ

-0.15

 missed

-0.14

alc

-0.14

isp

-0.13

HQ

-0.13

vey

-0.13

 ÑģÐ¾Ð±Ð¾Ð¹

-0.13

irr

-0.13

 Wahl

-0.13

 sclerosis

-0.13

POSITIVE LOGITS

inning

0.17

xsd

0.15

igm

0.15

deme

0.15

oted

0.14

_EXTENSIONS

0.13

ineTransform

0.13

monkey

0.13

ziel

0.13

 district

0.13

Activations Density 0.086%

locations and geographic references

No Comments

No Known Activations