INDEX

Explanations

references to formal discussions and meetings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

REC

-0.17

elle

-0.15

sla

-0.14

aman

-0.14

ponde

-0.14

ndef

-0.14

ikk

-0.14

ique

-0.14

Ø³ØªØ§ÙĨ

-0.14

ç¢¼

-0.14

POSITIVE LOGITS

ksam

0.15

ÙĦØ·

0.15

OTOR

0.15

Typ

0.15

.createFrom

0.15

ulia

0.14

 Incontri

0.14

Î·

0.14

iram

0.14

EST

0.14

Activations Density 0.203%

references to formal discussions and meetings

No Comments

No Known Activations