INDEX

Explanations

elements that involve JSON object structures and data manipulation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ÃŃrk

-0.16

olib

-0.14

reso

-0.14

 éĩ

-0.14

IFO

-0.14

'gc

-0.13

ubern

-0.13

áº¤

-0.13

ojis

-0.13

nou

-0.13

POSITIVE LOGITS

erna

0.16

ount

0.14

asy

0.14

 ÑħÐ¾Ð·ÑıÐ¹

0.14

ector

0.14

otos

0.14

 createSelector

0.14

_dual

0.14

empty

0.14

anky

0.13

Activations Density 0.015%

elements that involve JSON object structures and data manipulation

No Comments

No Known Activations