INDEX

Explanations

fragments related to origins or foundational aspects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

arehouse

-0.17

ãĤ¥

-0.15

Åº

-0.15

BAB

-0.14

asil

-0.14

ÑĥÐºÐ¸

-0.14

leared

-0.13

Å¾it

-0.13

bsp

-0.13

valuator

-0.13

POSITIVE LOGITS

chez

0.17

aidu

0.15

sez

0.14

esson

0.14

ledik

0.14

ival

0.14

ifest

0.14

ãĥªãĤ¹

0.14

ieder

0.14

 cuanto

0.14

Activations Density 0.018%

fragments related to origins or foundational aspects

No Comments

No Known Activations