INDEX

Explanations

references to educational institutions and programs

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

veis

-0.16

istique

-0.16

amage

-0.16

assen

-0.14

IRD

-0.14

itÃ©s

-0.14

udd

-0.14

 Å¡t

-0.14

 stan

-0.14

anax

-0.13

POSITIVE LOGITS

Ri

0.16

exion

0.15

jug

0.14

nor

0.14

Ø¹Ø§Ø±

0.13

 performing

0.13

ÏģÏį

0.13

bilt

0.13

 offence

0.13

 perform

0.13

Activations Density 0.005%

references to educational institutions and programs

No Comments

No Known Activations