INDEX

Explanations

references to various educational institutions and their classifications

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

itten

-0.15

zel

-0.14

uss

-0.14

 Diana

-0.14

 Burb

-0.14

and

-0.14

 Oasis

-0.14

Î¸Î®

-0.13

zik

-0.13

 ranks

-0.13

POSITIVE LOGITS

 somebody

0.15

inoa

0.15

 someone

0.15

osate

0.15

 nÃło

0.15

_deinit

0.15

timeofday

0.15

someone

0.14

latlong

0.14

ildo

0.14

Activations Density 0.300%

references to various educational institutions and their classifications

No Comments

No Known Activations