INDEX

Explanations

geographical names and addresses

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

oty

-0.17

Ĥæķ°

-0.15

otec

-0.14

adj

-0.14

.flat

-0.14

oter

-0.14

Ð¾Ð±Ð¾Ð²

-0.14

idia

-0.14

æ¥½

-0.14

 Moss

-0.13

POSITIVE LOGITS

ìĸ¼

0.15

}elseif

0.14

IGHL

0.14

SenderId

0.14

ÅĤad

0.13

ENUM

0.13

GC

0.13

imon

0.13

IMENT

0.13

Activations Density 0.184%

geographical names and addresses

No Comments

No Known Activations