INDEX

Explanations

timestamps and time-related abbreviations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

uce

-0.16

eniz

-0.16

itness

-0.15

eker

-0.14

ovice

-0.14

ipo

-0.14

aviest

-0.14

obox

-0.14

ÑĦÐ¸

-0.14

innie

-0.14

POSITIVE LOGITS

 Sphere

0.16

owitz

0.15

kovi

0.14

ï»¿#

0.14

 ç½

0.14

Ã²ng

0.13

éľŀ

0.13

/epl

0.13

 sphere

0.13

 handjob

0.13

Activations Density 0.011%

timestamps and time-related abbreviations

No Comments

No Known Activations