INDEX

Explanations

references to winning or victorious outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ikit

-0.16

-0.15

iki

-0.15

bounce

-0.15

 Welch

-0.15

eka

-0.14

Ã©nÃ©

-0.14

bah

-0.14

asi

-0.14

utes

-0.13

POSITIVE LOGITS

isia

0.16

itore

0.16

ÑĢÐ°Ð¹

0.16

 ustanov

0.15

olor

0.14

osu

0.14

Nej

0.14

olla

0.14

à¥ģà¤

0.14

arness

0.14

Activations Density 0.079%

references to winning or victorious outcomes

No Comments

No Known Activations