INDEX

Explanations

conditional statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

 Combination

-0.15

Rac

-0.14

onym

-0.14

icense

-0.14

rier

-0.14

ernel

-0.13

mÃ¡

-0.13

ãģĹãģ®

-0.13

ÅĦ

-0.13

isor

-0.13

POSITIVE LOGITS

Ã¨ge

0.17

ATUS

0.15

fen

0.14

ayet

0.14

afen

0.14

deniz

0.14

 press

0.13

lesc

0.13

lef

0.13

ubit

0.13

Activations Density 0.098%

conditional statements

No Comments

No Known Activations