INDEX

Explanations

suffixes or segments related to programming and technology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

eneric

-0.19

asco

-0.15

andle

-0.15

-0.14

emann

-0.14

IRECTION

-0.14

enville

-0.14

unci

-0.14

isco

-0.14

iband

-0.14

POSITIVE LOGITS

ars

0.17

nts

0.17

ãĥ¼ãĥł

0.16

rary

0.15

ÙĪÙħÛĮ

0.14

 Bernardino

0.14

mys

0.14

ally

0.14

antes

0.14

 postage

0.14

Activations Density 0.110%

suffixes or segments related to programming and technology

No Comments

No Known Activations