INDEX

Explanations

expressions of following or leading in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

pra

-0.15

 backgrounds

-0.14

.getElementsByName

-0.14

atoria

-0.14

anches

-0.14

Î¹Î²

-0.14

 queryInterface

-0.13

rame

-0.13

 useClass

-0.13

pez

-0.13

POSITIVE LOGITS

 footsteps

0.46

 lead

0.40

 path

0.32

lead

0.31

 Lead

0.30

 leads

0.30

Lead

0.29

 steps

0.29

 suit

0.29

 example

0.29

Activations Density 0.090%

expressions of following or leading in various contexts

No Comments

No Known Activations