INDEX

Explanations

specific numerical dates and years

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

maz

-0.16

wed

-0.15

 spin

-0.14

 flux

-0.14

pl

-0.14

 groom

-0.13

 Rosenstein

-0.13

 pret

-0.13

idget

-0.13

heel

-0.13

POSITIVE LOGITS

affe

0.16

arez

0.15

umbo

0.15

Ctrls

0.15

hev

0.14

umsuz

0.14

Millis

0.14

istro

0.14

FAULT

0.14

èĥŀ

0.14

Activations Density 0.068%

specific numerical dates and years

No Comments

No Known Activations