INDEX

Explanations

instances of significant events, actions, or themes that imply transitions or changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 however

-0.23

 jednak

-0.16

 Ð¿ÑĢÐ°Ð²Ð´Ð°

-0.15

 Ð¾Ð´Ð½Ð°ÐºÐ¾

-0.15

however

-0.15

gado

-0.15

çĦ¶èĢĮ

-0.15

ÄĽl

-0.15

elerik

-0.15

 HOWEVER

-0.15

POSITIVE LOGITS

 they

0.22

it

0.21

 ÙģØ¥ÙĨ

0.19

they

0.18

æĿ¥è¯´

0.18

 Ð¾Ð½Ð¾

0.15

he

0.15

ula

0.15

 Ð¾Ð½Ð¸

0.15

å®ĥ

0.14

Activations Density 0.239%

instances of significant events, actions, or themes that imply transitions or changes

No Comments

No Known Activations