INDEX

Explanations

phrases indicating observation or expectation

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

isman

-0.15

pop

-0.15

Ð»ÐµÑĤ

-0.14

LETE

-0.14

ÐºÑĥÐ»

-0.14

nder

-0.13

thesize

-0.13

chalk

-0.13

 chalk

-0.13

fte

-0.13

POSITIVE LOGITS

colo

0.15

ommen

0.15

pig

0.15

\OptionsResolver

0.15

 sÃ¡t

0.14

HEL

0.14

oplevel

0.14

pector

0.14

gne

0.14

coli

0.14

Activations Density 0.055%

phrases indicating observation or expectation

No Comments

No Known Activations