INDEX

Explanations

references to community values and relationships

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

hausen

-0.15

adge

-0.14

 scre

-0.14

coli

-0.14

(æľ¨

-0.14

ivent

-0.14

imbus

-0.14

uset

-0.13

zb

-0.13

Ø®

-0.13

POSITIVE LOGITS

reich

0.14

entrant

0.14

ught

0.14

uchos

0.13

tl

0.13

 aray

0.13

 compt

0.13

awe

0.13

even

0.13

ocoder

0.13

Activations Density 0.056%

references to community values and relationships

No Comments

No Known Activations