INDEX

Explanations

references to Netflix and its programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Loren

-0.17

unge

-0.15

ochen

-0.14

ç¤

-0.14

optera

-0.14

Fot

-0.14

 divider

-0.13

Zam

-0.13

mr

-0.13

Kre

-0.13

POSITIVE LOGITS

RD

0.16

ian

0.16

ÑĤÑĮ

0.15

ÑĥÑģÐ°

0.14

CEPTION

0.14

ioxide

0.14

Bru

0.14

_userdata

0.14

ipher

0.14

prene

0.14

Activations Density 0.004%

references to Netflix and its programming

No Comments

No Known Activations