INDEX

Explanations

technical details and methodologies related to processes and measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ninger

-0.14

nowrap

-0.14

ieri

-0.14

 Belle

-0.14

/gif

-0.14

ecial

-0.14

Ð¾ÑĩÐºÐ¸

-0.13

arella

-0.13

è¼Ķ

-0.13

Î½Î±Î½

-0.13

POSITIVE LOGITS

 approach

0.19

 method

0.19

 technique

0.15

 classic

0.15

appro

0.15

 scheme

0.15

Hin

0.15

classic

0.15

0.14

mil

0.14

Activations Density 0.182%

technical details and methodologies related to processes and measurements

No Comments

No Known Activations