INDEX

Explanations

terms associated with decision-making processes and policy formulation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 messenger

-0.18

BED

-0.17

raya

-0.17

anko

-0.16

brero

-0.16

mits

-0.16

pios

-0.15

 Mull

-0.15

(mut

-0.15

Made

-0.15

POSITIVE LOGITS

making

0.31

 making

0.31

-making

0.29

 makers

0.26

-M

0.26

mak

0.25

mak

0.25

makers

0.24

 Making

0.24

Mak

0.24

Activations Density 0.027%

terms associated with decision-making processes and policy formulation

No Comments

No Known Activations