INDEX

Explanations

references to academic journals and citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

inds

-0.16

PU

-0.15

amation

-0.15

Ø®Ø§ÙĨ

-0.15

auge

-0.14

ente

-0.14

arms

-0.14

 clutch

-0.14

Fre

-0.14

angers

-0.14

POSITIVE LOGITS

 FindObjectOfType

0.22

_Statics

0.15

ãĤĽ

0.15

ulumi

0.15

nodoc

0.15

inox

0.15

SGlobal

0.14

tir

0.14

IGNAL

0.14

 Arbor

0.14

Activations Density 0.004%

references to academic journals and citations

No Comments

No Known Activations