INDEX

Explanations

HTML tags and attributes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

enes

-0.15

raman

-0.15

ormap

-0.15

é»

-0.15

kest

-0.15

igy

-0.14

cerer

-0.14

Ð»Ð°Ð²

-0.14

eken

-0.14

ysa

-0.13

POSITIVE LOGITS

 onResponse

0.16

 motive

0.14

åĨĨ

0.14

kah

0.14

Fur

0.14

fur

0.14

aÅ¾

0.14

elas

0.14

vro

0.13

uild

0.13

Activations Density 0.005%

HTML tags and attributes

No Comments

No Known Activations