INDEX

Explanations

words related to ability or aptitude

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

-0.16

odzi

-0.16

run

-0.15

arpa

-0.15

ktor

-0.15

hatt

-0.15

rs

-0.15

iness

-0.15

head

-0.15

POSITIVE LOGITS

itude

0.40

itudes

0.31

ITUDE

0.28

itud

0.23

eka

0.21

ech

0.20

0.19

edd

0.17

amer

0.17

ly

0.17

Activations Density 0.008%

words related to ability or aptitude

No Comments

No Known Activations