INDEX

Explanations

references to chess and chess-related accomplishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 pump

-0.19

 Pump

-0.17

xt

-0.16

.sdk

-0.15

 pumps

-0.15

 Pompe

-0.15

neys

-0.14

gado

-0.14

 Principal

-0.14

kyt

-0.14

POSITIVE LOGITS

	freopen

0.17

.gstatic

0.16

æ£ĭ

0.14

 slova

0.14

aggio

0.14

ennes

0.14

à¹ĩ

0.14

\Tests

0.13

ulp

0.13

 unfold

0.13

Activations Density 0.023%

references to chess and chess-related accomplishments

No Comments

No Known Activations