INDEX

Explanations

mentions of companies or organizations, particularly Google

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ie

-0.19

-0.18

uling

-0.17

èĬĤ

-0.14

–

-0.14

Hud

-0.13

rud

-0.13

èĢ³

-0.13

 Haram

-0.13

POSITIVE LOGITS

 apparently

0.17

Apparently

0.17

izik

0.15

 Apparently

0.15

 nond

0.15

 supposedly

0.14

_normalize

0.14

antz

0.14

 Microsoft

0.14

ÚĨÙĩ

0.14

Activations Density 0.000%

mentions of companies or organizations, particularly Google

No Comments

No Known Activations