INDEX

Explanations

proper nouns, particularly names, especially in academic or publication contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

adio

-0.17

ObjectContext

-0.17

rif

-0.17

unning

-0.16

gor

-0.15

hythm

-0.15

forder

-0.15

orc

-0.15

erif

-0.15

enor

-0.14

POSITIVE LOGITS

instein

0.17

 Champ

0.15

linger

0.15

 Tall

0.14

onian

0.14

burgh

0.14

Ã³s

0.14

á»Łi

0.14

uff

0.14

ives

0.14

Activations Density 0.112%

proper nouns, particularly names, especially in academic or publication contexts

No Comments

No Known Activations