INDEX

Explanations

references to large quantities or mass items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ingles

-0.17

emez

-0.15

ekil

-0.15

ILLISE

-0.14

oulos

-0.14

quil

-0.14

 wings

-0.14

 unilateral

-0.13

ureka

-0.13

POSITIVE LOGITS

Bulk

0.17

 Bulk

0.16

ture

0.15

pcl

0.15

bulk

0.14

ØŃÙĤ

0.14

apest

0.14

leyici

0.14

 bulk

0.14

spin

0.14

Activations Density 0.006%

references to large quantities or mass items

No Comments

No Known Activations