INDEX

Explanations

elements related to historical events and exhibitions

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

-0.16

Ã©es

-0.15

StartPosition

-0.15

 Strand

-0.15

@js

-0.14

otope

-0.14

hod

-0.14

odos

-0.14

compat

-0.14

UniformLocation

-0.14

POSITIVE LOGITS

è©²

0.17

 decltype

0.17

wick

0.15

 outlier

0.15

ino

0.15

 à¤ªà¤¸

0.14

orks

0.14

éĤ£ä¸ª

0.14

 sole

0.14

è¿Ļä¸ª

0.14

Activations Density 0.265%

elements related to historical events and exhibitions

No Comments

No Known Activations