INDEX

Explanations

elements related to geographical or location-based information

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

aken

-0.17

Ð¾Ð»Ð¾Ð¼

-0.16

-0.15

blr

-0.15

opt

-0.14

rem

-0.14

unami

-0.14

elden

-0.14

 cath

-0.14

POSITIVE LOGITS

glo

0.15

Writes

0.14

IVO

0.14

SCII

0.14

CallCheck

0.14

Ãłng

0.14

ADDR

0.14

aat

0.14

ordan

0.14

ISP

0.14

Activations Density 0.023%

elements related to geographical or location-based information

No Comments

No Known Activations

elements related to geographical or location-based information

No Comments

No Known Activations