INDEX

Explanations

architectural and historical descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

uci

-0.15

 validationResult

-0.15

.AF

-0.14

åĨł

-0.14

forge

-0.14

idot

-0.14

 tinh

-0.13

hte

-0.13

zia

-0.13

asso

-0.13

POSITIVE LOGITS

ìĻ¸

0.14

 bulk

0.13

 Cald

0.13

dap

0.13

ĥĿ

0.13

Ngb

0.13

 centr

0.13

 originally

0.13

Activations Density 0.061%

architectural and historical descriptions

No Comments

No Known Activations