INDEX

Explanations

phrases that indicate the process of creating or defining unique characteristics

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

lake

-0.15

ycin

-0.15

icamente

-0.14

pur

-0.14

ÐµÑĢÐ³

-0.14

zimmer

-0.14

)sender

-0.13

çĬ¶æħĭ

-0.13

ãĥ©ãĤ¹

-0.13

ode

-0.13

POSITIVE LOGITS

 successful

0.16

0.15

rÃ¡

0.15

emmel

0.15

drive

0.15

Ľi

0.14

usting

0.14

successful

0.14

 Rudy

0.14

 Successful

0.14

Activations Density 0.065%

phrases that indicate the process of creating or defining unique characteristics

No Comments

No Known Activations