INDEX

Explanations

occurrences of the word "gle" and its variations, indicating a focus on expressions of brightness or light

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

tiv

-0.17

HELL

-0.16

anÃ¡

-0.15

rud

-0.15

oha

-0.15

igua

-0.15

rar

-0.14

ACKET

-0.14

POSITIVE LOGITS

eful

0.32

efully

0.28

aming

0.26

esome

0.21

ason

0.21

ewood

0.21

eson

0.19

eman

0.18

asons

0.18

ful

0.18

Activations Density 0.002%

occurrences of the word "gle" and its variations, indicating a focus on expressions of brightness or light

No Comments

No Known Activations