INDEX

Explanations

words associated with varying degrees of intensity or pressure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

BITS

-0.15

stitute

-0.15

liers

-0.14

immune

-0.14

ustum

-0.14

uin

-0.14

Cro

-0.14

.trip

-0.14

plers

-0.13

plib

-0.13

POSITIVE LOGITS

 minimum

0.15

Ð¾Ñĩ

0.14

ful

0.14

 Minimum

0.14

eft

0.14

çī§

0.14

Ø¹ÙħØ§ÙĦ

0.13

fu

0.13

hatt

0.13

 grounds

0.13

Activations Density 0.649%

words associated with varying degrees of intensity or pressure

No Comments

No Known Activations