INDEX

Explanations

elements related to analysis and categorization of various topics, particularly in studies or discussions

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

ensch

-0.17

 âĹĦ

-0.15

ulumi

-0.15

illance

-0.14

ventus

-0.14

igham

-0.13

utter

-0.13

åĩĨ

-0.13

mime

-0.13

bler

-0.13

POSITIVE LOGITS

 Doll

0.19

 flats

0.16

ovable

0.16

Hey

0.15

 embedded

0.15

airo

0.15

embedded

0.14

Bor

0.14

veral

0.14

 Howard

0.14

Activations Density 0.190%

elements related to analysis and categorization of various topics, particularly in studies or discussions

No Comments

No Known Activations