INDEX

Explanations

numerical values and references to notable events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

idot

-0.14

Pai

-0.14

Va

-0.14

 downstream

-0.14

oco

-0.13

 lead

-0.13

oes

-0.13

va

-0.13

ÑĳÑĤ

-0.13

(New

-0.13

POSITIVE LOGITS

Sez

0.16

icot

0.15

ãĤ·ãĤ¢

0.15

Ø³Ø§ÙĦ

0.14

avia

0.14

abay

0.14

à¸µà¸ĭ

0.14

TTY

0.14

ÅĻÃŃz

0.14

isine

0.14

Activations Density 0.001%

numerical values and references to notable events

No Comments

No Known Activations