INDEX

Explanations

questions and expressions that prompt opinion or feedback

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

agan

-0.15

witch

-0.15

Ð±Ñĭ

-0.15

enumer

-0.14

onn

-0.14

atat

-0.14

vant

-0.14

acad

-0.14

ibri

-0.14

enger

-0.14

POSITIVE LOGITS

chyb

0.20

Ð²Ð¾ÑĢ

0.15

alers

0.15

cheng

0.14

ToProps

0.14

alim

0.14

_Impl

0.14

alon

0.14

åº¦

0.14

Activations Density 0.023%

questions and expressions that prompt opinion or feedback

No Comments

No Known Activations