INDEX

Explanations

corporate or business-related terminology and associations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

arro

-0.16

ears

-0.13

abbo

-0.13

lech

-0.13

ÑĢÐµÐ±

-0.13

eyer

-0.13

enties

-0.12

ardash

-0.12

YSTEM

-0.12

elay

-0.12

POSITIVE LOGITS

ckill

0.14

ãĥ³ãĥĶ

0.14

qrst

0.14

 trÃŃ

0.13

âĢı

0.13

ua

0.12

Mel

0.12

 whiteColor

0.12

cession

0.12

rena

0.12

Activations Density 0.470%

corporate or business-related terminology and associations

No Comments

No Known Activations