INDEX

Explanations

phrases indicating repetition or redundancy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

nb

-0.17

vere

-0.16

ument

-0.16

INST

-0.16

æ¸Ī

-0.15

ished

-0.14

lege

-0.14

al

-0.14

 absolute

-0.14

emple

-0.14

POSITIVE LOGITS

ifar

0.17

ÐµÐ»ÑĮÐ·Ñı

0.15

 Ð¡ÑĤÐ°

0.14

entai

0.14

 Plugins

0.14

opc

0.14

UDA

0.13

 Persons

0.13

DeltaTime

0.13

èĮĤ

0.13

Activations Density 0.008%

phrases indicating repetition or redundancy

No Comments

No Known Activations