INDEX

Explanations

geographical locations or references to specific places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ĥn

-0.19

enant

-0.18

pas

-0.16

emet

-0.16

oise

-0.16

croft

-0.15

ispers

-0.15

à¤¹à¤°

-0.15

hare

-0.15

existent

-0.15

POSITIVE LOGITS

iris

0.24

mium

0.21

born

0.21

ascript

0.20

Commerce

0.19

.path

0.19

wald

0.18

borne

0.18

commerce

0.18

mos

0.17

Activations Density 0.021%

geographical locations or references to specific places

No Comments

No Known Activations