INDEX

Explanations

terms related to structural integrity and maintenance

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

uchen

-0.17

ardin

-0.16

alous

-0.16

veau

-0.16

ariate

-0.15

orex

-0.14

luv

-0.14

_SAMPL

-0.14

çħ§

-0.14

oleon

-0.14

POSITIVE LOGITS

 accuracy

0.15

ismic

0.15

 Lorenzo

0.15

Ð»Ð°Ñģ

0.15

nut

0.15

 compensate

0.14

avit

0.14

Nut

0.14

-center

0.14

accuracy

0.14

Activations Density 0.151%

terms related to structural integrity and maintenance

No Comments

No Known Activations