INDEX

Explanations

questions related to knowledge tests or quizzes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 McDon

-0.15

lp

-0.15

å¹¸

-0.15

AAD

-0.15

ertools

-0.15

kers

-0.15

kop

-0.15

 Repeat

-0.14

preter

-0.14

piler

-0.14

POSITIVE LOGITS

wright

0.17

 Kenn

0.15

Wid

0.15

EFR

0.14

 relate

0.14

 correct

0.14

æŃ£ç¡®

0.14

 correctly

0.14

Activations Density 0.214%

questions related to knowledge tests or quizzes

No Comments

No Known Activations