INDEX

Explanations

terms related to adjustable settings and modifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

aver

-0.18

.extern

-0.16

neau

-0.16

æľ

-0.16

rung

-0.15

inch

-0.15

usty

-0.14

ious

-0.14

POOL

-0.14

HEMA

-0.14

POSITIVE LOGITS

 controls

0.18

ths

0.17

 controlling

0.16

sb

0.16

-controlled

0.16

æİ§

0.15

 setbacks

0.15

 controlled

0.15

 Kont

0.15

alue

0.15

Activations Density 0.075%

terms related to adjustable settings and modifications

No Comments

No Known Activations