INDEX

Explanations

numerical identifiers or codes related to various contexts

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

Weg

-0.16

-analytics

-0.15

gram

-0.15

ood

-0.15

oad

-0.15

ouver

-0.14

ayo

-0.14

çīĻ

-0.14

ä¼į

-0.14

weg

-0.14

POSITIVE LOGITS

elik

0.14

_cv

0.14

angu

0.14

 cotton

0.14

iminal

0.14

ains

0.14

è¹

0.14

lew

0.13

heimer

0.13

xdc

0.13

Activations Density 0.000%

numerical identifiers or codes related to various contexts

No Comments

No Known Activations