INDEX

Explanations

mentions of individuals, particularly in the context of entertainment or discussions around celebrity figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ëį°ìĿ´íĬ¸

-0.19

uitka

-0.16

_fwd

-0.15

/Typography

-0.15

.bridge

-0.15

iggins

-0.15

alama

-0.15

ILLA

-0.14

baugh

-0.14

DonaldTrump

-0.14

POSITIVE LOGITS

 Belle

0.15

lep

0.14

’s

0.14

CM

0.14

 Matter

0.14

çĮª

0.14

0.13

ism

0.13

PS

0.13

Activations Density 0.193%

mentions of individuals, particularly in the context of entertainment or discussions around celebrity figures

No Comments

No Known Activations