INDEX

Explanations

numbers and alphanumeric codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

-0.13

ani

-0.12

âĢ¦

-0.12

_asm

-0.12

RICT

-0.12

ÂĿ

-0.11

unic

-0.11

ORIZONTAL

-0.11

ÂŃ

-0.11

POSITIVE LOGITS

antro

0.15

eview

0.15

<typeof

0.15

erva

0.14

ephir

0.14

 DEALINGS

0.13

ADDE

0.13

 mÃ¤nner

0.13

"group

0.13

ekim

0.13

Activations Density 0.023%

numbers and alphanumeric codes

No Comments

No Known Activations