INDEX

Explanations

research-related terminology and phrases associated with studies and investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

plug

-0.15

 readings

-0.15

$__

-0.14

ields

-0.14

ivo

-0.14

ei

-0.14

ãģł

-0.14

villa

-0.14

zano

-0.14

_DIST

-0.13

POSITIVE LOGITS

how

0.22

 whether

0.20

 relationships

0.18

relationships

0.16

whether

0.16

how

0.16

 Relationships

0.15

.opendaylight

0.15

 cÃ³mo

0.15

Į¨

0.15

Activations Density 0.097%

research-related terminology and phrases associated with studies and investigations

No Comments

No Known Activations