INDEX

Explanations

references to media features and appearances, particularly on covers of magazines and shows

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

iÄį

-0.16

ĤŃ

-0.15

 Ø¨ÙĪÙĦ

-0.15

msp

-0.14

anos

-0.14

ACLE

-0.14

ushi

-0.13

ustos

-0.13

isoft

-0.13

pton

-0.13

POSITIVE LOGITS

.grpc

0.16

ildo

0.14

CIS

0.14

0.13

op

0.13

awner

0.13

llib

0.13

æ©

0.13

Activations Density 0.206%

references to media features and appearances, particularly on covers of magazines and shows

No Comments

No Known Activations