INDEX

Explanations

numerical values, particularly those related to programming or data structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

itung

-0.15

hof

-0.14

OLUME

-0.14

Ðħ

-0.14

omite

-0.14

Ã¸r

-0.14

à¸Ĳà¸²à¸Ļ

-0.14

overy

-0.13

hold

-0.13

NGTH

-0.13

POSITIVE LOGITS

vais

0.16

ellig

0.15

unes

0.14

viron

0.14

iculture

0.14

ssel

0.13

hei

0.13

acional

0.13

ummer

0.13

eree

0.13

Activations Density 0.090%

numerical values, particularly those related to programming or data structures

No Comments

No Known Activations