INDEX

Explanations

action verbs related to tasks and authority, especially those involving control or management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ality

-0.18

zb

-0.17

zh

-0.16

editary

-0.16

gs

-0.16

ycl

-0.16

xs

-0.15

zk

-0.15

abouts

-0.15

uality

-0.14

POSITIVE LOGITS

 factor

0.23

 factors

0.23

redients

0.20

factor

0.20

 Factors

0.19

redient

0.19

-factor

0.18

ÑģÑı

0.18

_FACTOR

0.17

 agent

0.17

Activations Density 0.128%

action verbs related to tasks and authority, especially those involving control or management

No Comments

No Known Activations