INDEX

Explanations

geographical locations and event details

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

subpackage

-0.18

Ð¾Ð³Ð¾Ð´

-0.17

ield

-0.15

_PTR

-0.15

stad

-0.15

paged

-0.14

agas

-0.14

Lam

-0.14

al

-0.14

Mah

-0.14

POSITIVE LOGITS

ãĥ«ãĥķ

0.15

Latin

0.14

reesome

0.14

.scope

0.14

gay

0.13

ê²

0.13

ean

0.13

 VÅ¡

0.13

SetFont

0.13

ãģĲ

0.13

Activations Density 0.002%

geographical locations and event details

No Comments

No Known Activations