INDEX

Explanations

references to percentages and statistical increases or decreases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

Dop

-0.15

mant

-0.15

icks

-0.15

/media

-0.15

çķ

-0.15

esome

-0.15

ome

-0.14

linger

-0.14

Ã§os

-0.14

lements

-0.14

POSITIVE LOGITS

0.17

ITU

0.15

taÅŁ

0.15

itung

0.14

NGX

0.14

imbus

0.14

erken

0.13

 conce

0.13

cap

0.13

itmap

0.13

Activations Density 0.068%

references to percentages and statistical increases or decreases

No Comments

No Known Activations