INDEX

Explanations

concepts related to exceptions and specific conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

å¹

-0.15

 neither

-0.15

-0.14

endo

-0.14

è®¸å¤ļ

-0.14

arget

-0.14

ajs

-0.13

ale

-0.13

lobber

-0.13

idel

-0.13

POSITIVE LOGITS

thing

0.16

holm

0.16

lage

0.15

lest

0.15

remaining

0.15

bjerg

0.15

 remaining

0.15

gett

0.15

 thing

0.14

wheel

0.14

Activations Density 0.037%

concepts related to exceptions and specific conditions

No Comments

No Known Activations