INDEX

Explanations

specific references to education, courses, and instructional roles

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

Negative Logits

 academia

-0.16

.Suppress

-0.15

Ø§Ø¬Ø±

-0.14

iffer

-0.14

Ã¡Å¡

-0.14

upil

-0.13

ãĥ¼ãĥģ

-0.13

å°¿

-0.13

agne

-0.13

CADE

-0.13

POSITIVE LOGITS

uml

0.16

'gc

0.15

iques

0.15

 forg

0.14

pong

0.14

 EITHER

0.13

rror

0.13

eÄį

0.13

Activations Density 0.008%

specific references to education, courses, and instructional roles

No Known Activations