INDEX

Explanations

uniquely formatted or encoded characters and their associated values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 ..."↵

-0.14

Âĸ

-0.14

Âł↵

-0.14

ÂĶ

-0.14

 âĢ¦

-0.14

 youngsters

-0.13

 âĢ¦↵

-0.13

}`

-0.12

zbo

-0.12

 fractional

-0.12

POSITIVE LOGITS

FML

0.40

FML

0.31

 fucking

0.31

 FUCK

0.29

 fucked

0.28

 fuck

0.27

 Fuck

0.27

 Fucking

0.27

Fuck

0.25

 cunt

0.24

Activations Density 0.007%

uniquely formatted or encoded characters and their associated values

No Comments

No Known Activations