INDEX

Explanations

negations and questions related to personal circumstances or opinions

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

 whose

-0.17

unday

-0.16

who

-0.16

’Ãł

-0.16

-0.15

Ã¤s

-0.15

 sire

-0.14

Roe

-0.14

leftright

-0.14

inand

-0.14

POSITIVE LOGITS

sav

0.19

pou

0.17

aur

0.15

 compt

0.15

 eskort

0.15

cle

0.15

 mange

0.15

 conn

0.15

ROID

0.15

gles

0.15

Activations Density 0.014%

negations and questions related to personal circumstances or opinions

No Comments

No Known Activations