INDEX

Explanations

numerical or monetary values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

anto

-0.17

oulos

-0.15

din

-0.15

bian

-0.14

odom

-0.14

isel

-0.13

lod

-0.13

 tiener

-0.13

landers

-0.13

ither

-0.13

POSITIVE LOGITS

 Pound

0.15

quina

0.14

Ãłn

0.14

.TryParse

0.14

uid

0.14

ÏĦÏħ

0.14

/colors

0.13

 Colors

0.13

unes

0.13

Ð¾Ð·Ð´

0.13

Activations Density 0.054%

numerical or monetary values

No Comments

No Known Activations