INDEX

Explanations

phrases associated with gaining knowledge or accessing additional information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

usu

-0.14

.Mock

-0.13

áº¡t

-0.13

apur

-0.13

ousel

-0.13

 liberated

-0.13

entifier

-0.12

èĻŁ

-0.12

 Liberation

-0.12

 Baxter

-0.12

POSITIVE LOGITS

 noqa

0.15

//{{

0.15

rej

0.15

ebi

0.14

villa

0.14

vail

0.14

 Ø±ÙĪØ³ØªØ§

0.14

 WARRANT

0.14

 Ø§ÙĦØ³ÙĥØ§ÙĨ

0.14

endregion

0.14

Activations Density 0.159%

phrases associated with gaining knowledge or accessing additional information

No Comments

No Known Activations