INDEX

Explanations

elements related to performances and entertainment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 Madden

-0.15

.stamp

-0.14

 Laurel

-0.14

Disposition

-0.14

ummer

-0.13

retty

-0.13

quin

-0.13

ongs

-0.13

etak

-0.13

åĳ¨æľŁ

-0.13

POSITIVE LOGITS

 rein

0.17

ldr

0.15

reon

0.15

ouden

0.15

 reins

0.15

SED

0.15

 Rank

0.15

Rank

0.14

tainment

0.14

à¹Ģà¸ĭà¸Ńà¸£

0.14

Activations Density 0.028%

elements related to performances and entertainment

No Comments

No Known Activations