INDEX

Explanations

graphical formatting specifications related to figure dimensions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ensi

-0.17

aday

-0.16

inh

-0.15

ah

-0.15

 values

-0.15

æľĹ

-0.15

 Values

-0.15

elow

-0.14

igon

-0.14

ÏħÎ³

-0.14

POSITIVE LOGITS

woods

0.16

çĪµ

0.15

ILTER

0.14

agner

0.14

 NodeList

0.14

wife

0.14

erokee

0.14

ulp

0.14

à¸§à¸Ļ

0.14

æ´¥

0.13

Activations Density 0.007%

graphical formatting specifications related to figure dimensions

No Comments

No Known Activations