INDEX

Explanations

references to television or radio programs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

uj

-0.17

Ju

-0.17

ju

-0.16

ayan

-0.16

uida

-0.15

ore

-0.15

à¤¾à¤¯à¤¨

-0.15

Ket

-0.15

ors

-0.14

ork

-0.14

POSITIVE LOGITS

.LayoutStyle

0.18

uddy

0.17

abbo

0.17

 Mixer

0.15

lems

0.15

LEGRO

0.15

ichert

0.15

à¥Ŀ

0.15

oval

0.15

ucht

0.14

Activations Density 0.030%

references to television or radio programs

No Comments

No Known Activations