INDEX

Explanations

references to physical locations and structures

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

'icon

-0.15

åĿĤ

-0.14

ayah

-0.14

mix

-0.14

ibold

-0.14

ucer

-0.14

omik

-0.14

erus

-0.13

HEEL

-0.13

POSITIVE LOGITS

klady

0.15

tables

0.15

inkel

0.15

asa

0.14

ÏĨÎ¿

0.14

chor

0.14

 Lover

0.14

bage

0.14

adge

0.14

/vendors

0.14

Activations Density 0.085%

references to physical locations and structures

No Comments

No Known Activations