INDEX

Explanations

expressions that solicit opinions or thoughts from others

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ount

-0.18

agan

-0.17

enumer

-0.17

vention

-0.15

onn

-0.15

PE

-0.14

Gam

-0.14

witch

-0.14

Î®ÏĤ

-0.14

ton

-0.13

POSITIVE LOGITS

åº¦

0.16

asz

0.15

chyb

0.14

iele

0.13

 integerValue

0.13

LogLevel

0.13

herits

0.13

iful

0.13

_connector

0.13

cha

0.13

Activations Density 0.018%

expressions that solicit opinions or thoughts from others

No Comments

No Known Activations