INDEX

Explanations

references to academic journals or articles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ift

-0.16

Mev

-0.16

iston

-0.15

edback

-0.15

icker

-0.15

 Lowell

-0.15

earer

-0.14

afen

-0.14

bidden

-0.14

 euch

-0.14

POSITIVE LOGITS

awe

0.15

=>$

0.14

aldi

0.14

lÃ¡d

0.14

ìĽĶ

0.14

ellido

0.14

riÃ³

0.14

 productivity

0.14

oxid

0.14

ctype

0.13

Activations Density 0.012%

references to academic journals or articles

No Comments

No Known Activations