INDEX

Explanations

numeric values related to events and participation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ieres

-0.15

ApiController

-0.14

.asp

-0.14

çĲ³

-0.13

ocht

-0.13

åĿ¡

-0.13

uropean

-0.13

ÄįÃŃ

-0.13

ATEGORY

-0.13

rene

-0.13

POSITIVE LOGITS

 COPYING

0.16

ivid

0.15

arak

0.14

 ourselves

0.14

igel

0.14

 [...]↵↵

0.14

uso

0.13

our

0.13

tridge

0.13

 mmap

0.13

Activations Density 0.146%

numeric values related to events and participation

No Comments

No Known Activations