INDEX

Explanations

names and titles, particularly in a cultural or traditional context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

yr

-0.18

asco

-0.15

lac

-0.15

abic

-0.15

uais

-0.14

vier

-0.14

åĪĢ

-0.13

.CV

-0.13

LTR

-0.13

ryn

-0.13

POSITIVE LOGITS

odd

0.17

0.16

çĵ

0.16

iyon

0.15

anth

0.15

Vid

0.15

avig

0.15

Yug

0.15

vid

0.14

Rud

0.14

Activations Density 0.104%

names and titles, particularly in a cultural or traditional context

No Comments

No Known Activations