INDEX

Explanations

phrases related to logging into accounts or systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

 downs

-0.17

ivor

-0.15

oses

-0.15

osal

-0.14

uz

-0.14

eway

-0.14

errupt

-0.13

ÑĸÑĹÐ²

-0.13

 Pros

-0.13

Chr

-0.13

POSITIVE LOGITS

-on

0.22

on

0.21

-in

0.17

ON

0.17

ischer

0.16

ìĺ¨

0.16

onation

0.16

ãĤªãĥ³

0.16

owanie

0.16

off

0.16

Activations Density 0.009%

phrases related to logging into accounts or systems

No Comments

No Known Activations